From: Vipin Sharma <vipinsh@google.com>
To: kexec@lists.infradead.org, linux-kernel@vger.kernel.org,
linux-doc@vger.kernel.org, kvm@vger.kernel.org,
linux-mm@kvack.org, linux-kselftest@vger.kernel.org
Cc: ajayachandra@nvidia.com, alex@shazbot.org, amastro@fb.com,
ankita@nvidia.com, apopple@nvidia.com, bhelgaas@google.com,
chrisl@kernel.org, christian.koenig@amd.com, corbet@lwn.net,
dmatlack@google.com, graf@amazon.com,
jacob.pan@linux.microsoft.com, jgg@nvidia.com, jgg@ziepe.ca,
jrhilke@google.com, julianr@linux.ibm.com, kees@kernel.org,
kevin.tian@intel.com, leon@kernel.org, leonro@nvidia.com,
lukas@wunner.de, mattev@meta.com, michal.winiarski@intel.com,
parav@nvidia.com, pasha.tatashin@soleen.com, praan@google.com,
pratyush@kernel.org, rananta@google.com, rientjes@google.com,
rodrigo.vivi@intel.com, rppt@kernel.org, saeedm@nvidia.com,
schnelle@linux.ibm.com, skhan@linuxfoundation.org,
skhawaja@google.com, vipinsh@google.com,
vivek.kasireddy@intel.com, witu@nvidia.com,
yanjun.zhu@linux.dev, yi.l.liu@intel.com
Subject: [PATCH v5 07/20] vfio: Introduce vfio_find_device() helper
Date: Tue, 14 Jul 2026 08:14:52 -0700 [thread overview]
Message-ID: <20260714151505.3466855-8-vipinsh@google.com> (raw)
In-Reply-To: <20260714151505.3466855-1-vipinsh@google.com>
Introduce vfio_find_device() to discover a registered vfio_device from
vfio_device_class matching user-provided criteria.
To ensure safe lifecycle handling, vfio_find_device() acquires a
registration reference (device->refcount) via
vfio_device_try_get_registration() rather than relying solely on
device.kref taken in class_find_device(). Holding only kref allows
vendor modules to unload while external callers hold device pointers,
leading to a kernel panic on release. Callers must release the device
via vfio_device_put_registration().
Additionally, move refcount_set() prior to vfio_device_add() in
__vfio_register_dev() so device->refcount is initialized before the
device is published to class lookups, avoiding registration data races.
On vfio_device_add() failure, the error path drops the initial
registration reference and waits for completion to drain any reference
taken by a concurrent lookup before returning the error to the driver.
Signed-off-by: Vipin Sharma <vipinsh@google.com>
---
drivers/vfio/vfio_main.c | 64 +++++++++++++++++++++++++++++++++++++---
include/linux/vfio.h | 2 ++
2 files changed, 62 insertions(+), 4 deletions(-)
diff --git a/drivers/vfio/vfio_main.c b/drivers/vfio/vfio_main.c
index 2437af031169..e91fd83dc58d 100644
--- a/drivers/vfio/vfio_main.c
+++ b/drivers/vfio/vfio_main.c
@@ -13,6 +13,7 @@
#include <linux/cdev.h>
#include <linux/compat.h>
#include <linux/device.h>
+#include <linux/device/class.h>
#include <linux/fs.h>
#include <linux/idr.h>
#include <linux/iommu.h>
@@ -359,17 +360,20 @@ static int __vfio_register_dev(struct vfio_device *device,
goto err_out;
}
- ret = vfio_device_add(device);
- if (ret)
- goto err_out;
-
/* Refcounting can't start until the driver calls register */
refcount_set(&device->refcount, 1);
+ ret = vfio_device_add(device);
+ if (ret)
+ goto err_add;
+
vfio_device_group_register(device);
vfio_device_debugfs_init(device);
return 0;
+err_add:
+ vfio_device_put_registration(device);
+ wait_for_completion(&device->comp);
err_out:
vfio_device_remove_group(device);
return ret;
@@ -1797,6 +1801,58 @@ int vfio_dma_rw(struct vfio_device *device, dma_addr_t iova, void *data,
}
EXPORT_SYMBOL(vfio_dma_rw);
+struct vfio_device_match_data {
+ const void *data;
+ device_match_t match;
+};
+
+static int vfio_device_match_fn(struct device *dev, const void *data)
+{
+ struct vfio_device *vdev = container_of(dev, struct vfio_device, device);
+ const struct vfio_device_match_data *mdata = data;
+
+ if (mdata->match(dev, mdata->data)) {
+ if (vfio_device_try_get_registration(vdev))
+ return 1;
+ }
+ return 0;
+}
+
+/**
+ * vfio_find_device - Find a registered VFIO device matching search criteria
+ * @data: Context data passed to the @match function
+ * @match: Device match callback function
+ *
+ * Finds a registered VFIO device in the vfio_device_class matching the search
+ * criteria specified by @match.
+ *
+ * On success, this function acquires a registration reference on the matched
+ * device via vfio_device_try_get_registration(). This pins the device and
+ * prevents its vendor driver module from unloading or unbinding while the
+ * reference is held.
+ *
+ * Return: Pointer to the matching vfio_device on success, or NULL if no
+ * matching device is found. The caller MUST release the acquired registration
+ * reference using vfio_device_put_registration() when finished with the device.
+ */
+struct vfio_device *vfio_find_device(const void *data, device_match_t match)
+{
+ struct vfio_device_match_data mdata = { .data = data, .match = match };
+ struct device *device;
+
+ device = class_find_device(&vfio_device_class, NULL, &mdata, vfio_device_match_fn);
+ if (!device)
+ return NULL;
+
+ /*
+ * Drop standard kref reference. Caller holds reference from
+ * vfio_device_try_get_registration().
+ */
+ put_device(device);
+ return container_of(device, struct vfio_device, device);
+}
+EXPORT_SYMBOL_GPL(vfio_find_device);
+
/*
* Module/class support
*/
diff --git a/include/linux/vfio.h b/include/linux/vfio.h
index cf0d941d0410..42c3a9ae3a85 100644
--- a/include/linux/vfio.h
+++ b/include/linux/vfio.h
@@ -430,4 +430,6 @@ int vfio_virqfd_enable(void *opaque, int (*handler)(void *, void *),
void vfio_virqfd_disable(struct virqfd **pvirqfd);
void vfio_virqfd_flush_thread(struct virqfd **pvirqfd);
+struct vfio_device *vfio_find_device(const void *data, device_match_t match);
+
#endif /* VFIO_H */
--
2.55.0.795.g602f6c329a-goog
next prev parent reply other threads:[~2026-07-14 15:15 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-14 15:14 [PATCH v5 00/20] vfio/pci: Base Live Update support for VFIO Vipin Sharma
2026-07-14 15:14 ` [PATCH v5 01/20] vfio/pci: Factor out the reset logic in VFIO PCI device close path Vipin Sharma
2026-07-14 15:14 ` [PATCH v5 02/20] vfio: Export various helpers from VFIO Vipin Sharma
2026-07-14 15:14 ` [PATCH v5 03/20] vfio/pci: Export vfio_pci_dma_buf_move for vfio-pci module Vipin Sharma
2026-07-14 15:14 ` [PATCH v5 04/20] liveupdate: Export symbols needed by modules Vipin Sharma
2026-07-14 15:14 ` [PATCH v5 05/20] vfio/pci: Register a file handler with Live Update Orchestrator Vipin Sharma
2026-07-14 15:14 ` [PATCH v5 06/20] vfio/pci: Preserve vfio-pci device files across Live Update Vipin Sharma
2026-07-21 23:16 ` Josh Hilke
2026-07-14 15:14 ` Vipin Sharma [this message]
2026-07-14 15:14 ` [PATCH v5 08/20] vfio: Refactor vfio_device_fops_cdev_open() Vipin Sharma
2026-07-14 15:14 ` [PATCH v5 09/20] vfio: Add API to open cdev device for Live Update restore Vipin Sharma
2026-07-14 15:14 ` [PATCH v5 10/20] vfio/pci: Retrieve preserved device files after Live Update Vipin Sharma
2026-07-21 16:23 ` Josh Hilke
2026-07-14 15:14 ` [PATCH v5 11/20] vfio: Enforce preserved devices are retrieved via LIVEUPDATE_SESSION_RETRIEVE_FD Vipin Sharma
2026-07-14 15:14 ` [PATCH v5 12/20] docs: liveupdate: Add documentation for VFIO PCI Vipin Sharma
2026-07-14 15:14 ` [PATCH v5 13/20] vfio: selftests: Build liveupdate library in VFIO selftests Vipin Sharma
2026-07-14 15:14 ` [PATCH v5 14/20] vfio: selftests: Add vfio_pci_liveupdate_uapi_test Vipin Sharma
2026-07-14 15:15 ` [PATCH v5 15/20] vfio: selftests: Initialize vfio_pci_device using a VFIO cdev FD Vipin Sharma
2026-07-14 15:15 ` [PATCH v5 16/20] vfio: selftests: Add Makefile support for TEST_GEN_PROGS_EXTENDED Vipin Sharma
2026-07-14 15:15 ` [PATCH v5 17/20] vfio: selftests: Add vfio_pci_liveupdate_kexec_test Vipin Sharma
2026-07-14 15:15 ` [PATCH v5 18/20] vfio: selftests: Expose iommu_modes to tests Vipin Sharma
2026-07-14 15:15 ` [PATCH v5 19/20] vfio: selftests: Verify that opening VFIO device fails during Live Update Vipin Sharma
2026-07-14 15:15 ` [PATCH v5 20/20] vfio: selftests: Add continuous DMA to vfio_pci_liveupdate_kexec_test Vipin Sharma
[not found] ` <2c03ac97-60a6-4352-a942-27dcebae8d22@linux.dev>
[not found] ` <8fd94fad-e457-4839-9e41-ae257a138bee@linux.dev>
2026-07-20 17:08 ` [PATCH v5 00/20] vfio/pci: Base Live Update support for VFIO Vipin Sharma
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=20260714151505.3466855-8-vipinsh@google.com \
--to=vipinsh@google.com \
--cc=ajayachandra@nvidia.com \
--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=christian.koenig@amd.com \
--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=julianr@linux.ibm.com \
--cc=kees@kernel.org \
--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=lukas@wunner.de \
--cc=mattev@meta.com \
--cc=michal.winiarski@intel.com \
--cc=parav@nvidia.com \
--cc=pasha.tatashin@soleen.com \
--cc=praan@google.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=schnelle@linux.ibm.com \
--cc=skhan@linuxfoundation.org \
--cc=skhawaja@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox