From: Sinan Kaya <okaya@codeaurora.org>
To: Jean-Philippe Brucker <jean-philippe.brucker@arm.com>,
linux-arm-kernel@lists.infradead.org, linux-pci@vger.kernel.org,
linux-acpi@vger.kernel.org, devicetree@vger.kernel.org,
iommu@lists.linux-foundation.org, kvm@vger.kernel.org,
linux-mm@kvack.org
Cc: joro@8bytes.org, will.deacon@arm.com, robin.murphy@arm.com,
alex.williamson@redhat.com, tn@semihalf.com, liubo95@huawei.com,
thunder.leizhen@huawei.com, xieyisheng1@huawei.com,
xuzaibo@huawei.com, ilias.apalodimas@linaro.org,
jonathan.cameron@huawei.com, liudongdong3@huawei.com,
shunyong.yang@hxt-semitech.com, nwatters@codeaurora.org,
jcrouse@codeaurora.org, rfranz@cavium.com, dwmw2@infradead.org,
jacob.jun.pan@linux.intel.com, yi.l.liu@intel.com,
ashok.raj@intel.com, kevin.tian@intel.com,
baolu.lu@linux.intel.com, robdclark@gmail.com,
christian.koenig@amd.com, bharatku@xilinx.com,
rgummal@xilinx.com
Subject: Re: [PATCH v2 35/40] iommu/arm-smmu-v3: Add support for PCI ATS
Date: Sat, 19 May 2018 13:25:06 -0400 [thread overview]
Message-ID: <922474e8-0aa5-e022-0502-f1e51b0d4859@codeaurora.org> (raw)
In-Reply-To: <20180511190641.23008-36-jean-philippe.brucker@arm.com>
On 5/11/2018 3:06 PM, Jean-Philippe Brucker wrote:
> PCIe devices can implement their own TLB, named Address Translation Cache
> (ATC). Enable Address Translation Service (ATS) for devices that support
> it and send them invalidation requests whenever we invalidate the IOTLBs.
>
> Range calculation
> -----------------
>
> The invalidation packet itself is a bit awkward: range must be naturally
> aligned, which means that the start address is a multiple of the range
> size. In addition, the size must be a power of two number of 4k pages. We
> have a few options to enforce this constraint:
>
> (1) Find the smallest naturally aligned region that covers the requested
> range. This is simple to compute and only takes one ATC_INV, but it
> will spill on lots of neighbouring ATC entries.
>
> (2) Align the start address to the region size (rounded up to a power of
> two), and send a second invalidation for the next range of the same
> size. Still not great, but reduces spilling.
>
> (3) Cover the range exactly with the smallest number of naturally aligned
> regions. This would be interesting to implement but as for (2),
> requires multiple ATC_INV.
>
> As I suspect ATC invalidation packets will be a very scarce resource, I'll
> go with option (1) for now, and only send one big invalidation. We can
> move to (2), which is both easier to read and more gentle with the ATC,
> once we've observed on real systems that we can send multiple smaller
> Invalidation Requests for roughly the same price as a single big one.
>
> Note that with io-pgtable, the unmap function is called for each page, so
> this doesn't matter. The problem shows up when sharing page tables with
> the MMU.
>
> Timeout
> -------
>
> ATC invalidation is allowed to take up to 90 seconds, according to the
> PCIe spec, so it is possible to hit the SMMU command queue timeout during
> normal operations.
>
> Some SMMU implementations will raise a CERROR_ATC_INV_SYNC when a CMD_SYNC
> fails because of an ATC invalidation. Some will just abort the CMD_SYNC.
> Others might let CMD_SYNC complete and have an asynchronous IMPDEF
> mechanism to record the error. When we receive a CERROR_ATC_INV_SYNC, we
> could retry sending all ATC_INV since last successful CMD_SYNC. When a
> CMD_SYNC fails without CERROR_ATC_INV_SYNC, we could retry sending *all*
> commands since last successful CMD_SYNC.
>
> We cannot afford to wait 90 seconds in iommu_unmap, let alone MMU
> notifiers. So we'd have to introduce a more clever system if this timeout
> becomes a problem, like keeping hold of mappings and invalidating in the
> background. Implementing safe delayed invalidations is a very complex
> problem and deserves a series of its own. We'll assess whether more work
> is needed to properly handle ATC invalidation timeouts once this code runs
> on real hardware.
>
> Misc
> ----
>
> I didn't put ATC and TLB invalidations in the same functions for three
> reasons:
>
> * TLB invalidation by range is batched and committed with a single sync.
> Batching ATC invalidation is inconvenient, endpoints limit the number of
> inflight invalidations. We'd have to count the number of invalidations
> queued and send a sync periodically. In addition, I suspect we always
> need a sync between TLB and ATC invalidation for the same page.
>
> * Doing ATC invalidation outside tlb_inv_range also allows to send less
> requests, since TLB invalidations are done per page or block, while ATC
> invalidations target IOVA ranges.
>
> * TLB invalidation by context is performed when freeing the domain, at
> which point there isn't any device attached anymore.
>
> Signed-off-by: Jean-Philippe Brucker <jean-philippe.brucker@arm.com>
Nothing specific about this patch but just a general observation. Last time I
looked at the code, it seemed to require both ATS and PRI support from a given
hardware.
I think you can assume that for ATS 1.1 specification but ATS 1.0 specification
allows a system to have ATS+PASID without PRI.
QDF2400 is ATS 1.0 compatible as an example.
Is this an assumption / my misinterpretation?
--
Sinan Kaya
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.
next prev parent reply other threads:[~2018-05-19 17:25 UTC|newest]
Thread overview: 125+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-05-11 19:06 [PATCH v2 00/40] Shared Virtual Addressing for the IOMMU Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 01/40] iommu: Introduce Shared Virtual Addressing API Jean-Philippe Brucker
2018-05-16 20:41 ` Jacob Pan
2018-05-17 10:02 ` Jean-Philippe Brucker
2018-05-17 17:00 ` Jacob Pan
2018-09-05 11:29 ` Auger Eric
2018-09-06 11:09 ` Jean-Philippe Brucker
2018-09-06 11:12 ` Christian König
2018-09-06 12:45 ` Jean-Philippe Brucker
2018-09-07 8:55 ` Christian König
2018-09-07 15:45 ` Jean-Philippe Brucker
2018-09-07 18:02 ` Christian König
2018-09-07 21:25 ` Jacob Pan
2018-09-08 7:29 ` Christian König
2018-09-12 12:40 ` Jean-Philippe Brucker
2018-09-12 12:56 ` Christian König
2018-09-13 7:15 ` Tian, Kevin
2018-09-13 7:26 ` Tian, Kevin
2018-05-11 19:06 ` [PATCH v2 02/40] iommu/sva: Bind process address spaces to devices Jean-Philippe Brucker
2018-05-17 13:10 ` Jonathan Cameron
2018-05-21 14:43 ` Jean-Philippe Brucker
2018-09-05 11:29 ` Auger Eric
2018-09-06 11:09 ` Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 03/40] iommu/sva: Manage process address spaces Jean-Philippe Brucker
2018-05-16 23:31 ` Jacob Pan
2018-05-17 10:02 ` Jean-Philippe Brucker
2018-05-22 16:43 ` Jacob Pan
2018-05-24 11:44 ` Jean-Philippe Brucker
2018-05-24 11:50 ` Ilias Apalodimas
2018-05-24 15:04 ` Jean-Philippe Brucker
2018-05-25 6:33 ` Ilias Apalodimas
2018-05-25 8:39 ` Jonathan Cameron
2018-05-26 2:24 ` Kenneth Lee
[not found] ` <20180525093959.000040a7-hv44wF8Li93QT0dZR+AlfA@public.gmane.org>
2018-05-26 2:24 ` Kenneth Lee
2018-05-26 2:24 ` Kenneth Lee
[not found] ` <20180526022445.GA6069@kllp05>
2018-06-11 16:10 ` Kenneth Lee
2018-06-11 16:10 ` Kenneth Lee
2018-06-11 16:10 ` Kenneth Lee
2018-06-11 16:32 ` Kenneth Lee
2018-05-17 14:25 ` Jonathan Cameron
2018-05-21 14:44 ` Jean-Philippe Brucker
2018-09-05 12:14 ` Auger Eric
2018-09-05 18:18 ` Jacob Pan
2018-09-06 17:40 ` Jean-Philippe Brucker
2018-09-06 11:10 ` Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 04/40] iommu/sva: Add a mm_exit callback for device drivers Jean-Philippe Brucker
2018-09-05 13:23 ` Auger Eric
2018-09-06 11:10 ` Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 05/40] iommu/sva: Track mm changes with an MMU notifier Jean-Philippe Brucker
2018-05-17 14:25 ` Jonathan Cameron
2018-05-21 14:44 ` Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 06/40] iommu/sva: Search mm by PASID Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 07/40] iommu: Add a page fault handler Jean-Philippe Brucker
2018-05-17 15:25 ` Jonathan Cameron
2018-05-21 14:48 ` Jean-Philippe Brucker
2018-05-18 18:04 ` Jacob Pan
2018-05-21 14:49 ` Jean-Philippe Brucker
2018-05-22 23:35 ` Jacob Pan
2018-05-24 11:44 ` Jean-Philippe Brucker
2018-05-26 0:35 ` Jacob Pan
2018-05-29 10:00 ` Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 08/40] iommu/iopf: Handle mm faults Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 09/40] iommu/sva: Register page fault handler Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 10/40] mm: export symbol mm_access Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 11/40] mm: export symbol find_get_task_by_vpid Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 12/40] mm: export symbol mmput_async Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 13/40] vfio: Add support for Shared Virtual Addressing Jean-Philippe Brucker
2018-05-17 15:58 ` Jonathan Cameron
2018-05-21 14:51 ` Jean-Philippe Brucker
2018-05-23 9:38 ` Xu Zaibo
2018-05-24 11:44 ` Jean-Philippe Brucker
2018-05-24 12:35 ` Xu Zaibo
2018-05-24 15:04 ` Jean-Philippe Brucker
2018-05-25 2:39 ` Xu Zaibo
2018-05-25 9:47 ` Jean-Philippe Brucker
2018-05-26 3:53 ` Xu Zaibo
2018-05-29 11:55 ` Jean-Philippe Brucker
2018-05-29 12:24 ` Xu Zaibo
2018-08-27 8:06 ` Xu Zaibo
2018-08-31 13:34 ` Jean-Philippe Brucker
2018-09-01 2:23 ` Xu Zaibo
2018-09-03 10:34 ` Jean-Philippe Brucker
2018-09-04 2:12 ` Xu Zaibo
2018-09-04 10:57 ` Jean-Philippe Brucker
2018-09-05 3:15 ` Xu Zaibo
2018-09-05 11:02 ` Jean-Philippe Brucker
2018-09-06 7:26 ` Xu Zaibo
2018-05-11 19:06 ` [PATCH v2 14/40] dt-bindings: document stall and PASID properties for IOMMU masters Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 15/40] iommu/of: Add stall and pasid properties to iommu_fwspec Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 16/40] arm64: mm: Pin down ASIDs for sharing mm with devices Jean-Philippe Brucker
2018-05-15 14:16 ` Catalin Marinas
2018-05-17 10:01 ` Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 17/40] iommu/arm-smmu-v3: Link domains and devices Jean-Philippe Brucker
2018-05-17 16:07 ` Jonathan Cameron
2018-05-21 14:49 ` Jean-Philippe Brucker
2018-09-10 15:16 ` Auger Eric
2018-05-11 19:06 ` [PATCH v2 18/40] iommu/io-pgtable-arm: Factor out ARM LPAE register defines Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 19/40] iommu: Add generic PASID table library Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 20/40] iommu/arm-smmu-v3: Move context descriptor code Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 21/40] iommu/arm-smmu-v3: Add support for Substream IDs Jean-Philippe Brucker
2018-05-31 11:01 ` Bharat Kumar Gogada
2018-06-01 10:46 ` Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 22/40] iommu/arm-smmu-v3: Add second level of context descriptor table Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 23/40] iommu/arm-smmu-v3: Share process page tables Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 24/40] iommu/arm-smmu-v3: Seize private ASID Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 25/40] iommu/arm-smmu-v3: Add support for VHE Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 26/40] iommu/arm-smmu-v3: Enable broadcast TLB maintenance Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 27/40] iommu/arm-smmu-v3: Add SVA feature checking Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 28/40] iommu/arm-smmu-v3: Implement mm operations Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 29/40] iommu/arm-smmu-v3: Add support for Hardware Translation Table Update Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 30/40] iommu/arm-smmu-v3: Register I/O Page Fault queue Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 31/40] iommu/arm-smmu-v3: Improve add_device error handling Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 32/40] iommu/arm-smmu-v3: Maintain a SID->device structure Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 33/40] iommu/arm-smmu-v3: Add stall support for platform devices Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 34/40] ACPI/IORT: Check ATS capability in root complex nodes Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 35/40] iommu/arm-smmu-v3: Add support for PCI ATS Jean-Philippe Brucker
2018-05-19 17:25 ` Sinan Kaya [this message]
2018-05-21 14:52 ` Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 36/40] iommu/arm-smmu-v3: Hook up ATC invalidation to mm ops Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 37/40] iommu/arm-smmu-v3: Disable tagged pointers Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 38/40] PCI: Make "PRG Response PASID Required" handling common Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 39/40] iommu/arm-smmu-v3: Add support for PRI Jean-Philippe Brucker
2018-05-25 14:08 ` Bharat Kumar Gogada
2018-05-29 10:27 ` Jean-Philippe Brucker
2018-05-11 19:06 ` [PATCH v2 40/40] iommu/arm-smmu-v3: Add support for PCI PASID Jean-Philippe Brucker
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=922474e8-0aa5-e022-0502-f1e51b0d4859@codeaurora.org \
--to=okaya@codeaurora.org \
--cc=alex.williamson@redhat.com \
--cc=ashok.raj@intel.com \
--cc=baolu.lu@linux.intel.com \
--cc=bharatku@xilinx.com \
--cc=christian.koenig@amd.com \
--cc=devicetree@vger.kernel.org \
--cc=dwmw2@infradead.org \
--cc=ilias.apalodimas@linaro.org \
--cc=iommu@lists.linux-foundation.org \
--cc=jacob.jun.pan@linux.intel.com \
--cc=jcrouse@codeaurora.org \
--cc=jean-philippe.brucker@arm.com \
--cc=jonathan.cameron@huawei.com \
--cc=joro@8bytes.org \
--cc=kevin.tian@intel.com \
--cc=kvm@vger.kernel.org \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-mm@kvack.org \
--cc=linux-pci@vger.kernel.org \
--cc=liubo95@huawei.com \
--cc=liudongdong3@huawei.com \
--cc=nwatters@codeaurora.org \
--cc=rfranz@cavium.com \
--cc=rgummal@xilinx.com \
--cc=robdclark@gmail.com \
--cc=robin.murphy@arm.com \
--cc=shunyong.yang@hxt-semitech.com \
--cc=thunder.leizhen@huawei.com \
--cc=tn@semihalf.com \
--cc=will.deacon@arm.com \
--cc=xieyisheng1@huawei.com \
--cc=xuzaibo@huawei.com \
--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: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).