From: Jacob Pan <jacob.jun.pan@linux.intel.com> To: Baolu Lu <baolu.lu@linux.intel.com> Cc: "Tian, Kevin" <kevin.tian@intel.com>, Jason Gunthorpe <jgg@nvidia.com>, "iommu@lists.linux-foundation.org" <iommu@lists.linux-foundation.org>, LKML <linux-kernel@vger.kernel.org>, "dmaengine@vger.kernel.org" <dmaengine@vger.kernel.org>, Joerg Roedel <joro@8bytes.org>, David Woodhouse <dwmw2@infradead.org>, Jean-Philippe Brucker <jean-philippe@linaro.com>, Christoph Hellwig <hch@infradead.org>, "vkoul@kernel.org" <vkoul@kernel.org>, "robin.murphy@arm.com" <robin.murphy@arm.com>, "will@kernel.org" <will@kernel.org>, "Liu, Yi L" <yi.l.liu@intel.com>, "Jiang, Dave" <dave.jiang@intel.com>, "Raj, Ashok" <ashok.raj@intel.com>, Eric Auger <eric.auger@redhat.com>, jacob.jun.pan@linux.intel.com Subject: Re: [PATCH v4 1/6] iommu: Add a per domain PASID for DMA API Date: Tue, 31 May 2022 10:29:55 -0700 [thread overview] Message-ID: <20220531102955.6618b540@jacob-builder> (raw) In-Reply-To: <628aa885-dd12-8bcd-bfc6-446345bf69ed@linux.intel.com> Hi Baolu, On Tue, 31 May 2022 20:45:28 +0800, Baolu Lu <baolu.lu@linux.intel.com> wrote: > On 2022/5/31 18:12, Tian, Kevin wrote: > >>>> +++ b/include/linux/iommu.h > >>>> @@ -105,6 +105,8 @@ struct iommu_domain { > >>>> enum iommu_page_response_code (*iopf_handler)(struct > >> iommu_fault *fault, > >>>> void *data); > >>>> void *fault_data; > >>>> + ioasid_t pasid; /* Used for DMA requests > >>>> with PASID */ > >>>> + atomic_t pasid_users; > >>> These are poorly named, this is really the DMA API global PASID and > >>> shouldn't be used for other things. > >>> > >>> > >>> > >>> Perhaps I misunderstood, do you mind explaining more? > >> You still haven't really explained what this is for in this patch, > >> maybe it just needs a better commit message, or maybe something is > >> wrong. > >> > >> I keep saying the DMA API usage is not special, so why do we need to > >> create a new global pasid and refcount? Realistically this is only > >> going to be used by IDXD, why can't we just allocate a PASID and > >> return it to the driver every time a driver asks for DMA API on PASI > >> mode? Why does the core need to do anything special? > >> The reason why I store PASID at IOMMU domain is for IOTLB flush within the domain. Device driver is not aware of domain level IOTLB flush. We also have iova_cookie for each domain which essentially is for RIDPASID. > > Agree. I guess it was a mistake caused by treating ENQCMD as the > > only user although the actual semantics of the invented interfaces > > have already evolved to be quite general. > > > > This is very similar to what we have been discussing for iommufd. > > a PASID is just an additional routing info when attaching a device > > to an I/O address space (DMA API in this context) and by default > > it should be a per-device resource except when ENQCMD is > > explicitly opt in. > > > > Hence it's right time for us to develop common facility working > > for both this DMA API usage and iommufd, i.e.: > > > > for normal PASID attach to a domain, driver: > > > > allocates a local pasid from device local space; > > attaches the local pasid to a domain; > > > > for PASID attach in particular for ENQCMD, driver: > > > > allocates a global pasid in system-wide; > > attaches the global pasid to a domain; > > set the global pasid in PASID_MSR; > > > > In both cases the pasid is stored in the attach data instead of the > > domain. > > So during IOTLB flush for the domain, do we loop through the attach data? > > DMA API pasid is no special from above except it needs to allow > > one device attached to the same domain twice (one with RID > > and the other with RID+PASID). > > > > for iommufd those operations are initiated by userspace via > > iommufd uAPI. > > My understanding is that device driver owns its PASID policy. If ENQCMD > is supported on the device, the PASIDs should be allocated through > ioasid_alloc(). Otherwise, the whole PASID pool is managed by the device > driver. > It seems the changes we want for this patchset are: 1. move ioasid_alloc() from the core to device (allocation scope will be based on whether ENQCMD is intended or not) 2. store pasid in the attach data 3. use the same iommufd api to attach/set pasid on its default domain Am I summarizing correctly? > For kernel DMA w/ PASID, after the driver has a PASID for this purpose, > it can just set the default domain to the PASID on device. There's no > need for enable/disable() interfaces. > > Best regards, > baolu Thanks, Jacob
WARNING: multiple messages have this Message-ID (diff)
From: Jacob Pan <jacob.jun.pan@linux.intel.com> To: Baolu Lu <baolu.lu@linux.intel.com> Cc: "vkoul@kernel.org" <vkoul@kernel.org>, "Tian, Kevin" <kevin.tian@intel.com>, "Jiang, Dave" <dave.jiang@intel.com>, "Raj, Ashok" <ashok.raj@intel.com>, "will@kernel.org" <will@kernel.org>, Jean-Philippe Brucker <jean-philippe@linaro.com>, LKML <linux-kernel@vger.kernel.org>, Christoph Hellwig <hch@infradead.org>, "iommu@lists.linux-foundation.org" <iommu@lists.linux-foundation.org>, Jason Gunthorpe <jgg@nvidia.com>, "dmaengine@vger.kernel.org" <dmaengine@vger.kernel.org>, David Woodhouse <dwmw2@infradead.org>, "robin.murphy@arm.com" <robin.murphy@arm.com> Subject: Re: [PATCH v4 1/6] iommu: Add a per domain PASID for DMA API Date: Tue, 31 May 2022 10:29:55 -0700 [thread overview] Message-ID: <20220531102955.6618b540@jacob-builder> (raw) In-Reply-To: <628aa885-dd12-8bcd-bfc6-446345bf69ed@linux.intel.com> Hi Baolu, On Tue, 31 May 2022 20:45:28 +0800, Baolu Lu <baolu.lu@linux.intel.com> wrote: > On 2022/5/31 18:12, Tian, Kevin wrote: > >>>> +++ b/include/linux/iommu.h > >>>> @@ -105,6 +105,8 @@ struct iommu_domain { > >>>> enum iommu_page_response_code (*iopf_handler)(struct > >> iommu_fault *fault, > >>>> void *data); > >>>> void *fault_data; > >>>> + ioasid_t pasid; /* Used for DMA requests > >>>> with PASID */ > >>>> + atomic_t pasid_users; > >>> These are poorly named, this is really the DMA API global PASID and > >>> shouldn't be used for other things. > >>> > >>> > >>> > >>> Perhaps I misunderstood, do you mind explaining more? > >> You still haven't really explained what this is for in this patch, > >> maybe it just needs a better commit message, or maybe something is > >> wrong. > >> > >> I keep saying the DMA API usage is not special, so why do we need to > >> create a new global pasid and refcount? Realistically this is only > >> going to be used by IDXD, why can't we just allocate a PASID and > >> return it to the driver every time a driver asks for DMA API on PASI > >> mode? Why does the core need to do anything special? > >> The reason why I store PASID at IOMMU domain is for IOTLB flush within the domain. Device driver is not aware of domain level IOTLB flush. We also have iova_cookie for each domain which essentially is for RIDPASID. > > Agree. I guess it was a mistake caused by treating ENQCMD as the > > only user although the actual semantics of the invented interfaces > > have already evolved to be quite general. > > > > This is very similar to what we have been discussing for iommufd. > > a PASID is just an additional routing info when attaching a device > > to an I/O address space (DMA API in this context) and by default > > it should be a per-device resource except when ENQCMD is > > explicitly opt in. > > > > Hence it's right time for us to develop common facility working > > for both this DMA API usage and iommufd, i.e.: > > > > for normal PASID attach to a domain, driver: > > > > allocates a local pasid from device local space; > > attaches the local pasid to a domain; > > > > for PASID attach in particular for ENQCMD, driver: > > > > allocates a global pasid in system-wide; > > attaches the global pasid to a domain; > > set the global pasid in PASID_MSR; > > > > In both cases the pasid is stored in the attach data instead of the > > domain. > > So during IOTLB flush for the domain, do we loop through the attach data? > > DMA API pasid is no special from above except it needs to allow > > one device attached to the same domain twice (one with RID > > and the other with RID+PASID). > > > > for iommufd those operations are initiated by userspace via > > iommufd uAPI. > > My understanding is that device driver owns its PASID policy. If ENQCMD > is supported on the device, the PASIDs should be allocated through > ioasid_alloc(). Otherwise, the whole PASID pool is managed by the device > driver. > It seems the changes we want for this patchset are: 1. move ioasid_alloc() from the core to device (allocation scope will be based on whether ENQCMD is intended or not) 2. store pasid in the attach data 3. use the same iommufd api to attach/set pasid on its default domain Am I summarizing correctly? > For kernel DMA w/ PASID, after the driver has a PASID for this purpose, > it can just set the default domain to the PASID on device. There's no > need for enable/disable() interfaces. > > Best regards, > baolu Thanks, Jacob _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
next prev parent reply other threads:[~2022-05-31 17:26 UTC|newest] Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-05-18 18:21 [PATCH v4 0/6] Enable PASID for DMA API users Jacob Pan 2022-05-18 18:21 ` Jacob Pan 2022-05-18 18:21 ` [PATCH v4 1/6] iommu: Add a per domain PASID for DMA API Jacob Pan 2022-05-18 18:21 ` Jacob Pan 2022-05-19 6:50 ` Baolu Lu 2022-05-19 6:50 ` Baolu Lu 2022-05-24 13:50 ` Jason Gunthorpe 2022-05-24 13:50 ` Jason Gunthorpe via iommu 2022-05-24 15:17 ` Jacob Pan 2022-05-24 15:17 ` Jacob Pan 2022-05-30 12:22 ` Jason Gunthorpe 2022-05-30 12:22 ` Jason Gunthorpe via iommu 2022-05-31 10:12 ` Tian, Kevin 2022-05-31 10:12 ` Tian, Kevin 2022-05-31 12:45 ` Baolu Lu 2022-05-31 12:45 ` Baolu Lu 2022-05-31 16:03 ` Jason Gunthorpe 2022-05-31 16:03 ` Jason Gunthorpe via iommu 2022-05-31 17:29 ` Jacob Pan [this message] 2022-05-31 17:29 ` Jacob Pan 2022-05-31 19:05 ` Jason Gunthorpe 2022-05-31 19:05 ` Jason Gunthorpe via iommu 2022-05-31 20:44 ` Jacob Pan 2022-05-31 20:44 ` Jacob Pan 2022-06-01 1:50 ` Tian, Kevin 2022-06-01 1:50 ` Tian, Kevin 2022-06-01 1:43 ` Tian, Kevin 2022-06-01 1:43 ` Tian, Kevin 2022-06-01 9:37 ` Baolu Lu 2022-06-01 9:37 ` Baolu Lu 2022-06-01 10:05 ` Tian, Kevin 2022-06-01 10:05 ` Tian, Kevin 2022-05-18 18:21 ` [PATCH v4 2/6] iommu: Add a helper to do PASID lookup from domain Jacob Pan 2022-05-18 18:21 ` Jacob Pan 2022-05-19 6:41 ` Baolu Lu 2022-05-19 6:41 ` Baolu Lu 2022-05-19 20:10 ` Jacob Pan 2022-05-19 20:10 ` Jacob Pan 2022-05-19 6:48 ` Christoph Hellwig 2022-05-19 6:48 ` Christoph Hellwig 2022-05-20 15:18 ` Jacob Pan 2022-05-20 15:18 ` Jacob Pan 2022-05-23 7:55 ` Tian, Kevin 2022-05-23 7:55 ` Tian, Kevin 2022-05-23 9:14 ` Tian, Kevin 2022-05-23 9:14 ` Tian, Kevin 2022-05-23 18:01 ` Jacob Pan 2022-05-23 18:01 ` Jacob Pan 2022-05-18 18:21 ` [PATCH v4 3/6] iommu/vt-d: Implement domain ops for attach_dev_pasid Jacob Pan 2022-05-18 18:21 ` Jacob Pan 2022-05-24 13:51 ` Jason Gunthorpe 2022-05-24 13:51 ` Jason Gunthorpe via iommu 2022-05-24 16:12 ` Jacob Pan 2022-05-24 16:12 ` Jacob Pan 2022-05-24 18:02 ` Jason Gunthorpe 2022-05-24 18:02 ` Jason Gunthorpe via iommu 2022-05-24 20:45 ` Jacob Pan 2022-05-24 20:45 ` Jacob Pan 2022-05-24 21:10 ` Jason Gunthorpe 2022-05-24 21:10 ` Jason Gunthorpe via iommu 2022-05-18 18:21 ` [PATCH v4 4/6] iommu: Add PASID support for DMA mapping API users Jacob Pan 2022-05-18 18:21 ` Jacob Pan 2022-05-23 8:25 ` Tian, Kevin 2022-05-23 8:25 ` Tian, Kevin 2022-05-23 15:23 ` Jacob Pan 2022-05-23 15:23 ` Jacob Pan 2022-05-18 18:21 ` [PATCH v4 5/6] dmaengine: idxd: Use DMA API for in-kernel DMA with PASID Jacob Pan 2022-05-18 18:21 ` Jacob Pan 2022-05-18 18:21 ` [PATCH v4 6/6] iommu/vt-d: Delete unused SVM flag Jacob Pan 2022-05-18 18:21 ` Jacob Pan
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20220531102955.6618b540@jacob-builder \ --to=jacob.jun.pan@linux.intel.com \ --cc=ashok.raj@intel.com \ --cc=baolu.lu@linux.intel.com \ --cc=dave.jiang@intel.com \ --cc=dmaengine@vger.kernel.org \ --cc=dwmw2@infradead.org \ --cc=eric.auger@redhat.com \ --cc=hch@infradead.org \ --cc=iommu@lists.linux-foundation.org \ --cc=jean-philippe@linaro.com \ --cc=jgg@nvidia.com \ --cc=joro@8bytes.org \ --cc=kevin.tian@intel.com \ --cc=linux-kernel@vger.kernel.org \ --cc=robin.murphy@arm.com \ --cc=vkoul@kernel.org \ --cc=will@kernel.org \ --cc=yi.l.liu@intel.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.