From: Baolu Lu <baolu.lu@linux.intel.com>
To: Fenghua Yu <fenghua.yu@intel.com>, Vinod Koul <vkoul@kernel.org>,
Dave Jiang <dave.jiang@intel.com>
Cc: baolu.lu@linux.intel.com, dmaengine@vger.kernel.org,
linux-kernel <linux-kernel@vger.kernel.org>,
Alistair Popple <apopple@nvidia.com>,
Joerg Roedel <joro@8bytes.org>, Will Deacon <will@kernel.org>,
Robin Murphy <robin.murphy@arm.com>,
Lorenzo Stoakes <lstoakes@gmail.com>,
Christoph Hellwig <hch@infradead.org>,
iommu@lists.linux.dev
Subject: Re: [PATCH v2 08/16] iommu: define and export iommu_access_remote_vm()
Date: Wed, 8 Mar 2023 10:23:48 +0800 [thread overview]
Message-ID: <e37f4226-b4ee-9a80-187c-ac13733d353c@linux.intel.com> (raw)
In-Reply-To: <932aebd3-c655-3266-1acb-e41e8cbfb771@intel.com>
On 3/8/23 1:55 AM, Fenghua Yu wrote:
> Hi, Baolu,
>
> On 3/6/23 17:41, Baolu Lu wrote:
>> On 3/7/23 12:31 AM, Fenghua Yu wrote:
>>> Define and export iommu_access_remote_vm() to allow IOMMU related
>>> drivers to access user address space by PASID.
>>>
>>> The IDXD driver would like to use it to write the user's completion
>>> record that the hardware device is not able to write to due to user
>>> page fault.
>>
>> I don't quite follow here. Isn't I/O page fault already supported?
>
> The following patch 9 in this series explains in details why IDXD device
> cannot use page fault to write to user memory:
> https://lore.kernel.org/dmaengine/20230306163138.587484-10-fenghua.yu@intel.com/
>
> "DSA supports page fault handling through PRS. However, the DMA engine
> that's processing the descriptor is blocked until the PRS response is
> received. Other workqueues sharing the engine are also blocked.
> Page fault handing by the driver with PRS disabled can be used to
> mitigate the stalling.
Ah! Get your point now. Thanks for the explanation.
>
> With PRS disabled while ATS remain enabled, DSA handles page faults on
> a completion record by reporting an event in the event log. In this
> instance, the descriptor is completed and the event log contains the
> completion record address and the contents of the completion record."
>
> That's why IDXD driver needs this IOMMU's helper
> iommu_access_remote_vm() to copy the completion record from event log
> buffer to user space.
>
> Thanks.
>
> -Fenghua
Best regards,
baolu
next prev parent reply other threads:[~2023-03-08 2:24 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-06 16:31 [PATCH v2 00/16] Enable DSA 2.0 Event Log and completion record faulting features Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 01/16] dmaengine: idxd: make misc interrupt one shot Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 02/16] dmaengine: idxd: add event log size sysfs attribute Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 03/16] dmaengine: idxd: setup event log configuration Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 04/16] dmaengine: idxd: add interrupt handling for event log Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 05/16] dmanegine: idxd: add debugfs for event log dump Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 06/16] dmaengine: idxd: add per DSA wq workqueue for processing cr faults Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 07/16] dmaengine: idxd: create kmem cache for event log fault items Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 08/16] iommu: define and export iommu_access_remote_vm() Fenghua Yu
2023-03-07 1:41 ` Baolu Lu
2023-03-07 17:55 ` Fenghua Yu
2023-03-08 2:23 ` Baolu Lu [this message]
2023-03-20 13:35 ` Christoph Hellwig
2023-03-31 0:44 ` Fenghua Yu
2023-03-07 8:40 ` Jean-Philippe Brucker
2023-03-07 16:33 ` Fenghua Yu
2023-03-11 17:31 ` Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 09/16] dmaengine: idxd: process user page faults for completion record Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 10/16] dmaengine: idxd: add descs_completed field " Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 11/16] dmaengine: idxd: process batch descriptor completion record faults Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 12/16] dmaengine: idxd: add per file user counters for " Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 13/16] dmaengine: idxd: add a device to represent the file opened Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 14/16] dmaengine: idxd: expose fault counters to sysfs Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 15/16] dmaengine: idxd: add pid to exported sysfs attribute for opened file Fenghua Yu
2023-03-06 16:31 ` [PATCH v2 16/16] dmaengine: idxd: add per wq PRS disable Fenghua Yu
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=e37f4226-b4ee-9a80-187c-ac13733d353c@linux.intel.com \
--to=baolu.lu@linux.intel.com \
--cc=apopple@nvidia.com \
--cc=dave.jiang@intel.com \
--cc=dmaengine@vger.kernel.org \
--cc=fenghua.yu@intel.com \
--cc=hch@infradead.org \
--cc=iommu@lists.linux.dev \
--cc=joro@8bytes.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lstoakes@gmail.com \
--cc=robin.murphy@arm.com \
--cc=vkoul@kernel.org \
--cc=will@kernel.org \
/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).