All of lore.kernel.org
 help / color / mirror / Atom feed
From: Pranjal Shrivastava <praan@google.com>
To: Alex Williamson <alex@shazbot.org>
Cc: "David Matlack" <dmatlack@google.com>,
	"Adithya Jayachandran" <ajayachandra@nvidia.com>,
	"Alexander Graf" <graf@amazon.com>,
	"Alex Mastro" <amastro@fb.com>,
	"Alistair Popple" <apopple@nvidia.com>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	"Ankit Agrawal" <ankita@nvidia.com>,
	"Bjorn Helgaas" <bhelgaas@google.com>,
	"Chris Li" <chrisl@kernel.org>,
	"David Rientjes" <rientjes@google.com>,
	"Jacob Pan" <jacob.pan@linux.microsoft.com>,
	"Jason Gunthorpe" <jgg@nvidia.com>,
	"Jason Gunthorpe" <jgg@ziepe.ca>,
	"Jonathan Corbet" <corbet@lwn.net>,
	"Josh Hilke" <jrhilke@google.com>,
	"Kevin Tian" <kevin.tian@intel.com>,
	kexec@lists.infradead.org, kvm@vger.kernel.org,
	"Leon Romanovsky" <leon@kernel.org>,
	"Leon Romanovsky" <leonro@nvidia.com>,
	linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-kselftest@vger.kernel.org, linux-mm@kvack.org,
	linux-pci@vger.kernel.org, "Lukas Wunner" <lukas@wunner.de>,
	"Michał Winiarski" <michal.winiarski@intel.com>,
	"Mike Rapoport" <rppt@kernel.org>,
	"Parav Pandit" <parav@nvidia.com>,
	"Pasha Tatashin" <pasha.tatashin@soleen.com>,
	"Pratyush Yadav" <pratyush@kernel.org>,
	"Raghavendra Rao Ananta" <rananta@google.com>,
	"Rodrigo Vivi" <rodrigo.vivi@intel.com>,
	"Saeed Mahameed" <saeedm@nvidia.com>,
	"Samiullah Khawaja" <skhawaja@google.com>,
	"Shuah Khan" <skhan@linuxfoundation.org>,
	"Thomas Hellström" <thomas.hellstrom@linux.intel.com>,
	"Tomita Moeko" <tomitamoeko@gmail.com>,
	"Vipin Sharma" <vipinsh@google.com>,
	"Vivek Kasireddy" <vivek.kasireddy@intel.com>,
	"William Tu" <witu@nvidia.com>, "Yi Liu" <yi.l.liu@intel.com>,
	"Zhu Yanjun" <yanjun.zhu@linux.dev>
Subject: Re: [PATCH v2 04/22] vfio/pci: Register a file handler with Live Update Orchestrator
Date: Wed, 25 Feb 2026 22:29:41 +0000	[thread overview]
Message-ID: <aZ931bYILhhkhW-Y@google.com> (raw)
In-Reply-To: <aZ9yWlcqs2b6FLxy@google.com>

On Wed, Feb 25, 2026 at 10:06:18PM +0000, Pranjal Shrivastava wrote:
> On Wed, Feb 25, 2026 at 02:33:28PM -0700, Alex Williamson wrote:
> > On Thu, 29 Jan 2026 21:24:51 +0000
> > David Matlack <dmatlack@google.com> wrote:
> > > diff --git a/drivers/vfio/pci/vfio_pci.c b/drivers/vfio/pci/vfio_pci.c
> > > index 0c771064c0b8..19e88322af2c 100644
> > > --- a/drivers/vfio/pci/vfio_pci.c
> > > +++ b/drivers/vfio/pci/vfio_pci.c
> > > @@ -258,6 +258,10 @@ static int __init vfio_pci_init(void)
> > >  	int ret;
> > >  	bool is_disable_vga = true;
> > >  
> > > +	ret = vfio_pci_liveupdate_init();
> > > +	if (ret)
> > > +		return ret;
> > > +
> > >  #ifdef CONFIG_VFIO_PCI_VGA
> > >  	is_disable_vga = disable_vga;
> > >  #endif
> > > @@ -266,8 +270,10 @@ static int __init vfio_pci_init(void)
> > >  
> > >  	/* Register and scan for devices */
> > >  	ret = pci_register_driver(&vfio_pci_driver);
> > > -	if (ret)
> > > +	if (ret) {
> > > +		vfio_pci_liveupdate_cleanup();
> > >  		return ret;
> > > +	}
> > >  
> > >  	vfio_pci_fill_ids();
> > >  
> > > @@ -281,6 +287,7 @@ module_init(vfio_pci_init);
> > >  static void __exit vfio_pci_cleanup(void)
> > >  {
> > >  	pci_unregister_driver(&vfio_pci_driver);
> > > +	vfio_pci_liveupdate_cleanup();
> > >  }
> > >  module_exit(vfio_pci_cleanup);
> > >  
> > > diff --git a/drivers/vfio/pci/vfio_pci_liveupdate.c b/drivers/vfio/pci/vfio_pci_liveupdate.c
> > > new file mode 100644
> > > index 000000000000..b84e63c0357b
> > > --- /dev/null
> > > +++ b/drivers/vfio/pci/vfio_pci_liveupdate.c
> > > @@ -0,0 +1,69 @@
> > ...
> > > +static const struct liveupdate_file_ops vfio_pci_liveupdate_file_ops = {
> > > +	.can_preserve = vfio_pci_liveupdate_can_preserve,
> > > +	.preserve = vfio_pci_liveupdate_preserve,
> > > +	.unpreserve = vfio_pci_liveupdate_unpreserve,
> > > +	.retrieve = vfio_pci_liveupdate_retrieve,
> > > +	.finish = vfio_pci_liveupdate_finish,
> > > +	.owner = THIS_MODULE,
> > > +};
> > > +
> > > +static struct liveupdate_file_handler vfio_pci_liveupdate_fh = {
> > > +	.ops = &vfio_pci_liveupdate_file_ops,
> > > +	.compatible = VFIO_PCI_LUO_FH_COMPATIBLE,
> > > +};
> > > +
> > > +int __init vfio_pci_liveupdate_init(void)
> > > +{
> > > +	if (!liveupdate_enabled())
> > > +		return 0;
> > > +
> > > +	return liveupdate_register_file_handler(&vfio_pci_liveupdate_fh);
> > > +}
> > 
> > liveupdate_register_file_handler() "pins" vfio-pci with a
> > try_module_get().  Since this is done in our module_init function and
> > unregister occurs in our module_exit function, rather than relative
> > to any actual device binding or usage, this means vfio-pci CANNOT be
> > unloaded.  That seems bad.  Thanks,
> 
> Hmm... IIUC the concern here is about liveupdate policy if the user 
> wants to unload a module which was previously marked for preservation?
> 
> AFAICT, In such a case, the user is expected to close the LUO session FD,
> which "unpreserves" the FD. Finally, when rmmod is executed, the __exit 
> (vfio_pci_cleanup) calls vfio_pci_liveupdate_cleanup() which ends up 
> calling liveupdate_unregister_file_handler(), thereby dropping the ref
> held by the liveupdate orchestrator which allows the module to be
> unloaded.
> 

Ohh wait, You're right, Alex. I just realized the __exit won't even be 
reached because of the internal pin. The current implementation creates
a catch-22 where the module pins itself during init and can't reach the
unregister call in exit. I believe we should drop the ref when the user
closes the session FD? Additionally, should we move try_module_get out of
the global liveupdate_register_file_handler() and instead take the ref
only when a file is actually marked for preservation?

Thanks,
Praan

> I think we should document this policy somewhere or have a dev_warn to
> scream at the users when they try unloading the module without closing
> the session FD.
> 
> Thanks,
> Praan
> 
> > 
> > Alex
> > 
> > > +
> > > +void vfio_pci_liveupdate_cleanup(void)
> > > +{
> > > +	if (!liveupdate_enabled())
> > > +		return;
> > > +
> > > +	liveupdate_unregister_file_handler(&vfio_pci_liveupdate_fh);
> > > +}


  reply	other threads:[~2026-02-25 22:30 UTC|newest]

Thread overview: 120+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-29 21:24 [PATCH v2 00/22] vfio/pci: Base Live Update support for VFIO device files David Matlack
2026-01-29 21:24 ` [PATCH v2 01/22] liveupdate: Export symbols needed by modules David Matlack
2026-02-24  8:26   ` Pranjal Shrivastava
2026-02-24 17:08   ` Samiullah Khawaja
2026-01-29 21:24 ` [PATCH v2 02/22] PCI: Add API to track PCI devices preserved across Live Update David Matlack
2026-02-01  6:38   ` Zhu Yanjun
2026-02-02 18:14     ` David Matlack
2026-02-04  0:10       ` Yanjun.Zhu
2026-02-20 19:03         ` David Matlack
2026-02-23 22:04   ` Samiullah Khawaja
2026-02-23 23:08     ` David Matlack
2026-02-23 23:43       ` Samiullah Khawaja
2026-02-24  0:00         ` David Matlack
2026-02-24  9:17   ` Pranjal Shrivastava
2026-02-24 17:33     ` David Matlack
2026-02-24 19:02       ` Pranjal Shrivastava
2026-02-24 19:05         ` Pranjal Shrivastava
2026-02-25 22:46   ` Bjorn Helgaas
2026-02-26  0:28     ` David Matlack
2026-02-27 16:32       ` Alex Williamson
2026-02-27 17:19         ` David Matlack
2026-02-27 18:25           ` Alex Williamson
2026-02-27 22:19             ` David Matlack
2026-02-27 22:23               ` Alex Williamson
2026-02-27 22:35                 ` David Matlack
2026-03-01 19:20                   ` Jason Gunthorpe
2026-03-01 19:29             ` Jason Gunthorpe
2026-03-05 19:42               ` David Matlack
2026-03-01 19:14         ` Jason Gunthorpe
2026-01-29 21:24 ` [PATCH v2 03/22] PCI: Inherit bus numbers from previous kernel during " David Matlack
2026-02-24  9:36   ` Pranjal Shrivastava
2026-02-24 17:36     ` David Matlack
2026-02-25 22:47   ` Bjorn Helgaas
2026-02-25 23:20     ` David Matlack
2026-02-26 14:40       ` Jason Gunthorpe
2026-02-27 16:04         ` Alex Williamson
2026-03-01 19:22           ` Jason Gunthorpe
2026-03-02 18:14             ` David Matlack
2026-03-03 23:54               ` Jason Gunthorpe
2026-01-29 21:24 ` [PATCH v2 04/22] vfio/pci: Register a file handler with Live Update Orchestrator David Matlack
2026-02-06 22:37   ` Yanjun.Zhu
2026-02-06 23:14     ` David Matlack
2026-02-24  9:58       ` Pranjal Shrivastava
2026-02-25 21:33   ` Alex Williamson
2026-02-25 22:06     ` Pranjal Shrivastava
2026-02-25 22:29       ` Pranjal Shrivastava [this message]
2026-02-25 22:50         ` Samiullah Khawaja
2026-02-25 23:15     ` David Matlack
2026-02-25 23:54       ` Samiullah Khawaja
2026-03-19 22:09         ` David Matlack
2026-03-03  6:21   ` Zhu Yanjun
2026-01-29 21:24 ` [PATCH v2 05/22] vfio/pci: Preserve vfio-pci device files across Live Update David Matlack
2026-02-23 22:29   ` Samiullah Khawaja
2026-02-24 18:37   ` Pranjal Shrivastava
2026-02-24 19:16     ` David Matlack
2026-02-25 22:41   ` Alex Williamson
2026-02-25 23:41     ` David Matlack
2026-01-29 21:24 ` [PATCH v2 06/22] vfio/pci: Retrieve preserved device files after " David Matlack
2026-02-23 23:27   ` Samiullah Khawaja
2026-02-24 19:19   ` Pranjal Shrivastava
2026-02-26 22:52   ` Alex Williamson
2026-02-26 23:40     ` David Matlack
2026-04-01 14:36       ` Pratyush Yadav
2026-01-29 21:24 ` [PATCH v2 07/22] vfio/pci: Notify PCI subsystem about devices preserved across " David Matlack
2026-02-25  7:55   ` Pranjal Shrivastava
2026-02-26  0:45     ` David Matlack
2026-02-26 23:03   ` Alex Williamson
2026-02-26 23:31     ` David Matlack
2026-03-03 21:07     ` Jason Gunthorpe
2026-03-13  0:33       ` Pasha Tatashin
2026-03-13 12:25         ` Jason Gunthorpe
2026-03-13 14:27           ` Pasha Tatashin
2026-03-16 16:03   ` Vipin Sharma
2026-03-16 17:09     ` David Matlack
2026-01-29 21:24 ` [PATCH v2 08/22] vfio: Enforce preserved devices are retrieved via LIVEUPDATE_SESSION_RETRIEVE_FD David Matlack
2026-02-25  8:03   ` Pranjal Shrivastava
2026-02-26 23:15   ` Alex Williamson
2026-02-26 23:27     ` David Matlack
2026-02-26 23:42       ` Alex Williamson
2026-01-29 21:24 ` [PATCH v2 09/22] vfio/pci: Store incoming Live Update state in struct vfio_pci_core_device David Matlack
2026-02-25  8:38   ` Pranjal Shrivastava
2026-02-26  0:51     ` David Matlack
2026-01-29 21:24 ` [PATCH v2 10/22] vfio/pci: Skip reset of preserved device after Live Update David Matlack
2026-01-29 22:21   ` Jacob Pan
2026-01-29 22:33     ` David Matlack
2026-01-30  0:31       ` Jacob Pan
2026-02-27  0:00   ` Alex Williamson
2026-02-27  0:51     ` David Matlack
2026-02-27 15:46       ` Alex Williamson
2026-02-27 17:07         ` David Matlack
2026-02-27 17:57           ` Alex Williamson
2026-03-09 17:32             ` David Matlack
2026-03-12 23:39               ` David Matlack
2026-03-13 15:39                 ` Jacob Pan
2026-03-16 16:28                   ` Vipin Sharma
2026-03-19 16:28                     ` Jacob Pan
2026-03-16 18:11                   ` David Matlack
2026-03-19 16:39                     ` Jacob Pan
2026-03-19 17:23                       ` David Matlack
2026-03-16 16:22                 ` Vipin Sharma
2026-03-16 17:18                   ` David Matlack
2026-03-16 21:49                     ` Vipin Sharma
2026-03-16 22:14                       ` David Matlack
2026-03-17 23:36                         ` Vipin Sharma
2026-01-29 21:24 ` [PATCH v2 11/22] docs: liveupdate: Document VFIO device file preservation David Matlack
2026-03-16 21:55   ` Vipin Sharma
2026-03-16 22:30     ` David Matlack
2026-01-29 21:24 ` [PATCH v2 12/22] selftests/liveupdate: Move luo_test_utils.* into a reusable library David Matlack
2026-01-29 21:25 ` [PATCH v2 13/22] selftests/liveupdate: Add helpers to preserve/retrieve FDs David Matlack
2026-01-29 21:25 ` [PATCH v2 14/22] vfio: selftests: Build liveupdate library in VFIO selftests David Matlack
2026-01-29 21:25 ` [PATCH v2 15/22] vfio: selftests: Add Makefile support for TEST_GEN_PROGS_EXTENDED David Matlack
2026-01-29 21:25 ` [PATCH v2 16/22] vfio: selftests: Add vfio_pci_liveupdate_uapi_test David Matlack
2026-03-16 16:04   ` Vipin Sharma
2026-03-16 17:09     ` David Matlack
2026-01-29 21:25 ` [PATCH v2 17/22] vfio: selftests: Initialize vfio_pci_device using a VFIO cdev FD David Matlack
2026-01-29 21:25 ` [PATCH v2 18/22] vfio: selftests: Add vfio_pci_liveupdate_kexec_test David Matlack
2026-01-29 21:25 ` [PATCH v2 19/22] vfio: selftests: Expose iommu_modes to tests David Matlack
2026-01-29 21:25 ` [PATCH v2 20/22] vfio: selftests: Expose low-level helper routines for setting up struct vfio_pci_device David Matlack
2026-01-29 21:25 ` [PATCH v2 21/22] vfio: selftests: Verify that opening VFIO device fails during Live Update David Matlack
2026-01-29 21:25 ` [PATCH v2 22/22] vfio: selftests: Add continuous DMA to vfio_pci_liveupdate_kexec_test David Matlack

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=aZ931bYILhhkhW-Y@google.com \
    --to=praan@google.com \
    --cc=ajayachandra@nvidia.com \
    --cc=akpm@linux-foundation.org \
    --cc=alex@shazbot.org \
    --cc=amastro@fb.com \
    --cc=ankita@nvidia.com \
    --cc=apopple@nvidia.com \
    --cc=bhelgaas@google.com \
    --cc=chrisl@kernel.org \
    --cc=corbet@lwn.net \
    --cc=dmatlack@google.com \
    --cc=graf@amazon.com \
    --cc=jacob.pan@linux.microsoft.com \
    --cc=jgg@nvidia.com \
    --cc=jgg@ziepe.ca \
    --cc=jrhilke@google.com \
    --cc=kevin.tian@intel.com \
    --cc=kexec@lists.infradead.org \
    --cc=kvm@vger.kernel.org \
    --cc=leon@kernel.org \
    --cc=leonro@nvidia.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=lukas@wunner.de \
    --cc=michal.winiarski@intel.com \
    --cc=parav@nvidia.com \
    --cc=pasha.tatashin@soleen.com \
    --cc=pratyush@kernel.org \
    --cc=rananta@google.com \
    --cc=rientjes@google.com \
    --cc=rodrigo.vivi@intel.com \
    --cc=rppt@kernel.org \
    --cc=saeedm@nvidia.com \
    --cc=skhan@linuxfoundation.org \
    --cc=skhawaja@google.com \
    --cc=thomas.hellstrom@linux.intel.com \
    --cc=tomitamoeko@gmail.com \
    --cc=vipinsh@google.com \
    --cc=vivek.kasireddy@intel.com \
    --cc=witu@nvidia.com \
    --cc=yanjun.zhu@linux.dev \
    --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 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.