This gives management tools like libvirt a chance to open the vfio
cdev with privilege and pass FD to qemu. This way qemu never needs
to have privilege to open a VFIO or iommu cdev node.
Add a helper function vfio_device_get_name() to check fd and get
device name, it will also be used by other vfio devices.
There is no easy way to check if a device is mdev with FD passing,
so fail the x-balloon-allowed check unconditionally in this case.
There is also no easy way to get BDF as name with FD passing, so
we fake a name by VFIO_FD[fd].
Signed-off-by: Zhenzhong Duan <zhenzhong.d...@intel.com>
---
hw/vfio/helpers.c | 28 ++++++++++++++++++++++++++++
hw/vfio/iommufd.c | 12 ++++++++----
hw/vfio/pci.c | 35 ++++++++++++++++++++++++++++-------
include/hw/vfio/vfio-common.h | 1 +
4 files changed, 65 insertions(+), 11 deletions(-)
diff --git a/hw/vfio/helpers.c b/hw/vfio/helpers.c
index 4338456b08..1a27efb075 100644
--- a/hw/vfio/helpers.c
+++ b/hw/vfio/helpers.c
@@ -596,3 +596,31 @@ bool vfio_has_region_cap(VFIODevice *vbasedev, int region,
uint16_t cap_type)
return ret;
}
+
+int vfio_device_get_name(VFIODevice *vbasedev, Error **errp)
+{
+ struct stat st;
+
+ if (vbasedev->fd < 0) {
+ if (stat(vbasedev->sysfsdev, &st) < 0) {
+ error_setg_errno(errp, errno, "no such host device");
+ error_prepend(errp, VFIO_MSG_PREFIX, vbasedev->sysfsdev);
+ return -errno;
+ }
+ vbasedev->name = g_path_get_basename(vbasedev->sysfsdev);
+ }
+#ifdef CONFIG_IOMMUFD
+ else {
+ if (!vbasedev->iommufd) {
+ error_setg(errp, "Use FD passing only with iommufd backend");
+ return -EINVAL;
+ }
+ /*
+ * Give a name with fd so any function printing out vbasedev->name
+ * will not break.
+ */
+ vbasedev->name = g_strdup_printf("VFIO_FD%d", vbasedev->fd);
+ }
+#endif
+ return 0;
+}
diff --git a/hw/vfio/iommufd.c b/hw/vfio/iommufd.c
index dd24e76e39..2cd2daebf4 100644
--- a/hw/vfio/iommufd.c
+++ b/hw/vfio/iommufd.c
@@ -376,11 +376,15 @@ static int iommufd_attach_device(char *name, VFIODevice
*vbasedev,
uint32_t ioas_id;
Error *err = NULL;
- devfd = vfio_get_devicefd(vbasedev->sysfsdev, errp);
- if (devfd < 0) {
- return devfd;
+ if (vbasedev->fd < 0) {
+ devfd = vfio_get_devicefd(vbasedev->sysfsdev, errp);
+ if (devfd < 0) {
+ return devfd;
+ }
+ vbasedev->fd = devfd;
+ } else {
+ devfd = vbasedev->fd;
}
- vbasedev->fd = devfd;
ret = iommufd_connect_and_bind(vbasedev, errp);
if (ret) {
diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c
index 99265253f8..eff52b5014 100644
--- a/hw/vfio/pci.c
+++ b/hw/vfio/pci.c
@@ -44,6 +44,7 @@
#include "migration/qemu-file.h"
#include "linux/iommufd.h"
#include "sysemu/iommufd.h"
+#include "monitor/monitor.h"
#define TYPE_VFIO_PCI_NOHOTPLUG "vfio-pci-nohotplug"
@@ -3171,18 +3172,23 @@ static void vfio_realize(PCIDevice *pdev, Error **errp)
VFIODevice *vbasedev = &vdev->vbasedev;
char *tmp, *subsys;
Error *err = NULL;
- struct stat st;
int i, ret;
bool is_mdev;
char uuid[UUID_FMT_LEN];
char *name;
- if (!vbasedev->sysfsdev) {
+ if (vbasedev->fd < 0 && !vbasedev->sysfsdev) {
if (!(~vdev->host.domain || ~vdev->host.bus ||
~vdev->host.slot || ~vdev->host.function)) {
error_setg(errp, "No provided host device");
+#ifdef CONFIG_IOMMUFD
+ error_append_hint(errp, "Use -device vfio-pci,host=DDDD:BB:DD.F, "
+ "-device vfio-pci,sysfsdev=PATH_TO_DEVICE "
+ "or -device vfio-pci,fd=DEVICE_FD\n");
+#else
error_append_hint(errp, "Use -device vfio-pci,host=DDDD:BB:DD.F "
"or -device vfio-pci,sysfsdev=PATH_TO_DEVICE\n");
+#endif
return;
}
vbasedev->sysfsdev =
@@ -3191,13 +3197,9 @@ static void vfio_realize(PCIDevice *pdev, Error **errp)
vdev->host.slot, vdev->host.function);
}
- if (stat(vbasedev->sysfsdev, &st) < 0) {
- error_setg_errno(errp, errno, "no such host device");
- error_prepend(errp, VFIO_MSG_PREFIX, vbasedev->sysfsdev);
+ if (vfio_device_get_name(vbasedev, errp)) {
return;
}
-
- vbasedev->name = g_path_get_basename(vbasedev->sysfsdev);
vbasedev->ops = &vfio_pci_ops;
vbasedev->type = VFIO_DEVICE_TYPE_PCI;
vbasedev->dev = DEVICE(vdev);
@@ -3559,6 +3561,7 @@ static void vfio_instance_init(Object *obj)
vdev->host.bus = ~0U;
vdev->host.slot = ~0U;
vdev->host.function = ~0U;
+ vdev->vbasedev.fd = -1;
vdev->nv_gpudirect_clique = 0xFF;
@@ -3619,6 +3622,21 @@ static Property vfio_pci_dev_properties[] = {
DEFINE_PROP_END_OF_LIST(),
};
+#ifdef CONFIG_IOMMUFD
+static void vfio_pci_set_fd(Object *obj, const char *str, Error **errp)
+{
+ VFIOPCIDevice *vdev = VFIO_PCI(obj);
+ int fd = -1;
+
+ fd = monitor_fd_param(monitor_cur(), str, errp);
+ if (fd == -1) {
+ error_prepend(errp, "Could not parse remote object fd %s:", str);
+ return;
+ }
+ vdev->vbasedev.fd = fd;
+}
+#endif
+
static void vfio_pci_dev_class_init(ObjectClass *klass, void *data)
{
DeviceClass *dc = DEVICE_CLASS(klass);
@@ -3626,6 +3644,9 @@ static void vfio_pci_dev_class_init(ObjectClass *klass,
void *data)
dc->reset = vfio_pci_reset;
device_class_set_props(dc, vfio_pci_dev_properties);
+#ifdef CONFIG_IOMMUFD
+ object_class_property_add_str(klass, "fd", NULL, vfio_pci_set_fd);
+#endif
dc->desc = "VFIO-based PCI device assignment";
set_bit(DEVICE_CATEGORY_MISC, dc->categories);
pdc->realize = vfio_realize;
diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h
index 027a59a13a..41c8eeaa54 100644
--- a/include/hw/vfio/vfio-common.h
+++ b/include/hw/vfio/vfio-common.h
@@ -242,6 +242,7 @@ struct vfio_info_cap_header *
vfio_get_device_info_cap(struct vfio_device_info *info, uint16_t id);
struct vfio_info_cap_header *
vfio_get_cap(void *ptr, uint32_t cap_offset, uint16_t id);
+int vfio_device_get_name(VFIODevice *vbasedev, Error **errp);
#endif
extern const MemoryListener vfio_prereg_listener;
--
2.34.1
