From: Alex Williamson <alex@shazbot.org>
To: John Levon <john.levon@nutanix.com>
Cc: "Cédric Le Goater" <clg@redhat.com>,
qemu-devel@nongnu.org,
"Alex Williamson" <alex.williamson@redhat.com>,
"John Johnson" <john.g.johnson@oracle.com>,
"Elena Ufimtseva" <elena.ufimtseva@oracle.com>,
"Jagannathan Raman" <jag.raman@oracle.com>
Subject: Re: [PULL 25/28] vfio: add region info cache
Date: Tue, 14 Oct 2025 08:29:49 -0600 [thread overview]
Message-ID: <20251014082937.7de6f960@shazbot.org> (raw)
In-Reply-To: <aO5XeknOrUtvLJO9@lent>
On Tue, 14 Oct 2025 16:00:26 +0200
John Levon <john.levon@nutanix.com> wrote:
> On Tue, Oct 14, 2025 at 07:49:24AM -0600, Alex Williamson wrote:
>
> > > My apologies - we hit the exact same issue internally, but with a much older
> > > codebase, so I did not realise this could be an upstream problem as well!
> > >
> > > We put this down to a bug in the nvidia driver - surely it shouldn't be
> > > reporting fewer regions than are actually in use. So we applied what we
> > > thought to be a gross hack of boundary checking, and not using the region
> > > cache in case it's beyond num_regions.
> > >
> > > To put it another way, the header file says:
> > >
> > > 217 __u32 num_regions; /* Max region index + 1 */
> > >
> > > If it's not actually the max region index + 1, what are the expected
> > > semantics of this field, or of region indices more generally? We could not
> > > find any clear documentation on the topic other than this comment.
> >
> > '9' only defines the end of the fixed, pre-defined region indexes for
> > vfio-pci, ie. VFIO_PCI_NUM_REGIONS. Beyond that, we support device specific
> > regions. The GFX region is one such device specific region.
>
> To be clear, nowhere in the code are we hard-coding 9, we ask the underlying
> vfio instance for the num_regions value.
>
> In this case, it appears that the underlying instance is returning 9, and then
> trying to use index 9.
That sounds like a bug, in vfio_pci_ioctl_get_info() we return
num_regions as (VFIO_PCI_NUM_REGIONS + vdev->num_regions) where
vdev->num_regions is incremented via
vfio_pci_core_register_dev_region(). Any read, write, or mmap access
done through vfio-pci-core would need to increment vdev->num_regions to
support that access. I think we're dealing with NVIDIA vGPU here,
which afaik is the only driver that uses a region for the display, so
I'd suspect they have a bug and don't use the vfio-pci-core callbacks
here.
Does QEMU's caching model need to fall through to GET_REGION_INFO and
expand the cache for an out-of-bounds index to handle such a bug?
> > There is no fixed limit to the number of regions a device may expose
>
> Sure.
>
> >, nor is vfio_device_info.num_regions necessarily a static value.
>
> Ah - is this documented somewhere? And do you have an example of where this
> varies over the lifetime of a device currently?
It's more the case that it's not specifically defined as static over
the lifetime of the device. I don't think we want device spontaneously
creating new regions, we have no device-check notification mechanism to
userspace to reevaluate regions, but a user initiated action defined to
create a new region seems like a fair vector where num_regions could be
updated. Also the current working idea would be that regions are only
added, not modified or deleted.
> IOW, are these devices actually changing the num_regions value they report
> sometime after initialization (namely, can it change after
> vfio_device_prepare()) ?
In theory, yes.
> Or would querying num_regions still return the original value, and in fact the
> semantics of the field are "it's not the max region index, but the max region
> index during initialization" or similar ?
Ultimately it's just a marker that *should* indicate there are no
regions to query above (num_regions - 1). It should be valid at the
time it's queried. I don't think it would be valid that it could be
dynamic in any other way than a monotonically increasing value. Thanks,
Alex
next prev parent reply other threads:[~2025-10-14 14:31 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-09 13:12 [PULL 00/28] vfio queue Cédric Le Goater
2025-05-09 13:12 ` [PULL 01/28] vfio/container: ram discard disable helper Cédric Le Goater
2025-05-09 13:12 ` [PULL 02/28] vfio/container: reform vfio_container_connect cleanup Cédric Le Goater
2025-05-09 13:12 ` [PULL 03/28] vfio/container: vfio_container_group_add Cédric Le Goater
2025-05-09 13:12 ` [PULL 04/28] vfio/igd: Restrict legacy mode to Gen6-9 devices Cédric Le Goater
2025-05-09 13:12 ` [PULL 05/28] vfio/igd: Always emulate ASLS (OpRegion) register Cédric Le Goater
2025-05-09 13:12 ` [PULL 06/28] vfio/igd: Detect IGD device by OpRegion Cédric Le Goater
2025-05-09 13:12 ` [PULL 07/28] vfio/igd: Check vendor and device ID on GVT-g mdev Cédric Le Goater
2025-05-09 13:12 ` [PULL 08/28] vfio/igd: Check OpRegion support " Cédric Le Goater
2025-05-09 13:12 ` [PULL 09/28] vfio/igd: Enable OpRegion by default Cédric Le Goater
2025-05-09 13:12 ` [PULL 10/28] vfio/igd: Allow overriding GMS with 0xf0 to 0xfe on Gen9+ Cédric Le Goater
2025-05-09 13:13 ` [PULL 11/28] vfio/igd: Only emulate GGC register when x-igd-gms is set Cédric Le Goater
2025-05-09 13:13 ` [PULL 12/28] vfio/igd: Remove generation limitation for IGD passthrough Cédric Le Goater
2025-05-09 13:13 ` [PULL 13/28] linux-header: update-linux-header script changes Cédric Le Goater
2025-05-09 13:13 ` [PULL 14/28] linux-headers: Update to Linux v6.15-rc3 Cédric Le Goater
2025-05-09 13:13 ` [PULL 15/28] vfio: add vfio_device_prepare() Cédric Le Goater
2025-05-09 13:13 ` [PULL 16/28] vfio: add vfio_device_unprepare() Cédric Le Goater
2025-05-09 13:13 ` [PULL 17/28] vfio: add vfio_attach_device_by_iommu_type() Cédric Le Goater
2025-05-09 13:13 ` [PULL 18/28] vfio: add vfio_device_get_irq_info() helper Cédric Le Goater
2025-05-09 13:13 ` [PULL 19/28] vfio: consistently handle return value for helpers Cédric Le Goater
2025-05-09 13:13 ` [PULL 20/28] vfio: add strread/writeerror() Cédric Le Goater
2025-05-09 13:13 ` [PULL 21/28] vfio: add vfio_pci_config_space_read/write() Cédric Le Goater
2025-05-09 13:13 ` [PULL 22/28] vfio: add unmap_all flag to DMA unmap callback Cédric Le Goater
2025-05-09 13:13 ` [PULL 23/28] vfio: implement unmap all for DMA unmap callbacks Cédric Le Goater
2025-05-09 13:13 ` [PULL 24/28] vfio: add device IO ops vector Cédric Le Goater
2025-05-09 13:13 ` [PULL 25/28] vfio: add region info cache Cédric Le Goater
2025-10-14 13:16 ` Cédric Le Goater
2025-10-14 13:32 ` John Levon
2025-10-14 13:49 ` Alex Williamson
2025-10-14 13:58 ` Cédric Le Goater
2025-10-14 14:01 ` John Levon
2025-10-14 14:00 ` John Levon
2025-10-14 14:29 ` Alex Williamson [this message]
2025-10-14 14:36 ` John Levon
2025-05-09 13:13 ` [PULL 26/28] vfio: add read/write to device IO ops vector Cédric Le Goater
2025-05-09 13:13 ` [PULL 27/28] vfio: add vfio-pci-base class Cédric Le Goater
2025-05-09 13:13 ` [PULL 28/28] vfio/container: pass listener_begin/commit callbacks Cédric Le Goater
2025-05-10 18:36 ` [PULL 00/28] vfio queue Stefan Hajnoczi
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=20251014082937.7de6f960@shazbot.org \
--to=alex@shazbot.org \
--cc=alex.williamson@redhat.com \
--cc=clg@redhat.com \
--cc=elena.ufimtseva@oracle.com \
--cc=jag.raman@oracle.com \
--cc=john.g.johnson@oracle.com \
--cc=john.levon@nutanix.com \
--cc=qemu-devel@nongnu.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).