From: Matt Evans <matt@ozlabs.org>
To: Jason Gunthorpe <jgg@nvidia.com>
Cc: "Leon Romanovsky" <leon@kernel.org>,
"Alex Williamson" <alex@shazbot.org>,
"Alex Mastro" <amastro@fb.com>,
"Christian König" <christian.koenig@amd.com>,
"Bjorn Helgaas" <bhelgaas@google.com>,
"Logan Gunthorpe" <logang@deltatee.com>,
"Kevin Tian" <kevin.tian@intel.com>,
"Pranjal Shrivastava" <praan@google.com>,
"Longfang Liu" <liulongfang@huawei.com>,
"Mahmoud Adam" <mngyadam@amazon.de>,
"David Matlack" <dmatlack@google.com>,
"Björn Töpel" <bjorn@kernel.org>,
"Sumit Semwal" <sumit.semwal@linaro.org>,
"Ankit Agrawal" <ankita@nvidia.com>,
"Alistair Popple" <apopple@nvidia.com>,
"Vivek Kasireddy" <vivek.kasireddy@intel.com>,
linux-kernel@vger.kernel.org, linux-media@vger.kernel.org,
dri-devel@lists.freedesktop.org, linaro-mm-sig@lists.linaro.org,
kvm@vger.kernel.org, linux-pci@vger.kernel.org
Subject: Re: [PATCH v6 9/9] vfio/pci: Permanently revoke a DMABUF on request
Date: Tue, 15 Sep 2026 15:30:32 +0100 [thread overview]
Message-ID: <9ebcebb9-94a5-40b4-8764-5d70ea465f9c@ozlabs.org> (raw)
In-Reply-To: <20260915123517.GS3968357@nvidia.com>
Hi Jason,
On 15/09/2026 13:35, Jason Gunthorpe wrote:
> On Mon, Sep 14, 2026 at 01:13:36PM +0100, Matt Evans wrote:
>> "This is useful for lifecycle management, to reclaim VFIO PCI BAR
>> ranges previously delegated to a subordinate client process: by
>> revoking, the driver process can ensure that the loaned resources are
>> made inaccessible when the client is deemed "done". The original
>> DMABUF is defunct, and BAR resources can then be safely re-exported
>> for use by new clients."
>
> This makes the point clear, you want a way to permanently revoke a
> specific dmabuf fd and render it forever unable to access any
> underlying memory.
>
> You want to do this so the vendor can fence the vendee without
> requiring its co-operation.
Exactly! (I'll reword this commit message still, re other requests in
this thread, but you got the concept.)
> "temporary revoke" ie what VFIO already does around reset is a
> different thing.
Yes, right. From the DMABUF importer (and now mmap) perspective the
mechanism is unchanged. priv->revoked was always "temporary"/transient
(OK, splitting hairs, but at VFIO shutdown an exported DMABUF becomes
permanently inaccessible).
Here the only change is adding another cause of becoming inaccessible
(direct request from userspace) and the length of time (it can never be
re-attached, for the reasons you mentioned).
Cheers,
Matt
prev parent reply other threads:[~2026-09-15 14:30 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-11 21:41 [PATCH v6 0/9] vfio/pci: Add mmap() for DMABUFs Matt Evans
2026-09-11 21:41 ` [PATCH v6 1/9] vfio/pci: Remove DMABUF export dependency on vdev->memory_lock Matt Evans
2026-09-11 21:41 ` [PATCH v6 2/9] vfio/pci: Un-revoke DMABUFs in LOW_POWER_ENTRY_WITH_WAKEUP resume Matt Evans
2026-09-11 21:41 ` [PATCH v6 3/9] dma-buf: Export dma_buf_set_name() Matt Evans
2026-09-14 11:06 ` Christian König
2026-09-15 13:35 ` Matt Evans
2026-09-11 21:41 ` [PATCH v6 4/9] vfio/pci: Add a helper to look up PFNs for DMABUFs Matt Evans
2026-09-11 21:41 ` [PATCH v6 5/9] vfio/pci: Add a helper to create a DMABUF for a BAR-map VMA Matt Evans
2026-09-15 12:16 ` liulongfang
2026-09-21 13:24 ` Matt Evans
2026-09-22 9:16 ` liulongfang
2026-09-24 12:43 ` Matt Evans
2026-09-11 21:41 ` [PATCH v6 6/9] vfio/pci: Convert BAR mmap() to use a DMABUF Matt Evans
2026-09-11 21:41 ` [PATCH v6 7/9] vfio/pci: Clean up BAR zap and revocation Matt Evans
2026-09-11 21:41 ` [PATCH v6 8/9] vfio/pci: Support mmap() of a VFIO DMABUF Matt Evans
2026-09-11 21:41 ` [PATCH v6 9/9] vfio/pci: Permanently revoke a DMABUF on request Matt Evans
2026-09-13 16:52 ` Leon Romanovsky
2026-09-14 11:36 ` Jason Gunthorpe
2026-09-14 11:54 ` Leon Romanovsky
2026-09-14 11:58 ` Jason Gunthorpe
2026-09-14 12:06 ` Leon Romanovsky
2026-09-14 12:08 ` Jason Gunthorpe
2026-09-14 12:13 ` Matt Evans
2026-09-15 7:20 ` Leon Romanovsky
2026-09-15 11:13 ` Christian König
2026-09-15 14:22 ` Matt Evans
2026-09-16 14:19 ` Christian König
2026-09-21 13:08 ` Matt Evans
2026-09-21 13:22 ` Jason Gunthorpe
2026-09-21 13:45 ` Christian König
2026-09-21 13:49 ` Jason Gunthorpe
2026-09-21 14:09 ` Christian König
2026-09-22 12:41 ` Jason Gunthorpe
2026-09-22 12:46 ` Leon Romanovsky
2026-09-22 12:54 ` Jason Gunthorpe
2026-09-22 22:31 ` Alex Mastro
2026-09-22 22:57 ` Jason Gunthorpe
2026-09-23 15:40 ` Matt Evans
2026-09-23 16:13 ` Jason Gunthorpe
2026-09-23 16:52 ` Leon Romanovsky
2026-09-23 17:06 ` Matt Evans
2026-09-24 17:36 ` Alex Mastro
2026-09-22 11:40 ` Leon Romanovsky
2026-09-15 12:35 ` Jason Gunthorpe
2026-09-15 14:30 ` Matt Evans [this message]
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=9ebcebb9-94a5-40b4-8764-5d70ea465f9c@ozlabs.org \
--to=matt@ozlabs.org \
--cc=alex@shazbot.org \
--cc=amastro@fb.com \
--cc=ankita@nvidia.com \
--cc=apopple@nvidia.com \
--cc=bhelgaas@google.com \
--cc=bjorn@kernel.org \
--cc=christian.koenig@amd.com \
--cc=dmatlack@google.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=jgg@nvidia.com \
--cc=kevin.tian@intel.com \
--cc=kvm@vger.kernel.org \
--cc=leon@kernel.org \
--cc=linaro-mm-sig@lists.linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=liulongfang@huawei.com \
--cc=logang@deltatee.com \
--cc=mngyadam@amazon.de \
--cc=praan@google.com \
--cc=sumit.semwal@linaro.org \
--cc=vivek.kasireddy@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
all inboxes | Powered by JetHome®