| Message ID | 20230213151348.56451-12-yi.l.liu@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Add vfio_device cdev for iommufd support | expand |
> From: Liu, Yi L <yi.l.liu@intel.com> > Sent: Monday, February 13, 2023 11:14 PM > > for counting the devices that are opened via the cdev path. This count > is increased and decreased by the cdev path. The group path checks it > to achieve exclusion with the cdev path. With this, only one path (group > path or cdev path) will claim DMA ownership. This avoids scenarios in > which devices within the same group may be opened via different paths. please move vfio_device_claim/release_group() from patch 14 into this patch to make the exclusiveness part complete. > > Signed-off-by: Yi Liu <yi.l.liu@intel.com> > --- > drivers/vfio/group.c | 5 +++++ > drivers/vfio/vfio.h | 1 + > 2 files changed, 6 insertions(+) > > diff --git a/drivers/vfio/group.c b/drivers/vfio/group.c > index 9f3f6f0e4942..f3f5f4589cdd 100644 > --- a/drivers/vfio/group.c > +++ b/drivers/vfio/group.c > @@ -403,6 +403,11 @@ static int vfio_group_fops_open(struct inode *inode, > struct file *filep) > goto out_unlock; > } > > + if (group->cdev_device_open_cnt) { > + ret = -EBUSY; > + goto out_unlock; > + } > + > /* > * Do we need multiple instances of the group open? Seems not. > */ > diff --git a/drivers/vfio/vfio.h b/drivers/vfio/vfio.h > index 6f063e31d08a..7a77fb12bd2c 100644 > --- a/drivers/vfio/vfio.h > +++ b/drivers/vfio/vfio.h > @@ -84,6 +84,7 @@ struct vfio_group { > struct blocking_notifier_head notifier; > struct iommufd_ctx *iommufd; > spinlock_t kvm_ref_lock; > + unsigned int cdev_device_open_cnt; > }; > > int vfio_device_set_group(struct vfio_device *device, > -- > 2.34.1
diff --git a/drivers/vfio/group.c b/drivers/vfio/group.c index 9f3f6f0e4942..f3f5f4589cdd 100644 --- a/drivers/vfio/group.c +++ b/drivers/vfio/group.c @@ -403,6 +403,11 @@ static int vfio_group_fops_open(struct inode *inode, struct file *filep) goto out_unlock; } + if (group->cdev_device_open_cnt) { + ret = -EBUSY; + goto out_unlock; + } + /* * Do we need multiple instances of the group open? Seems not. */ diff --git a/drivers/vfio/vfio.h b/drivers/vfio/vfio.h index 6f063e31d08a..7a77fb12bd2c 100644 --- a/drivers/vfio/vfio.h +++ b/drivers/vfio/vfio.h @@ -84,6 +84,7 @@ struct vfio_group { struct blocking_notifier_head notifier; struct iommufd_ctx *iommufd; spinlock_t kvm_ref_lock; + unsigned int cdev_device_open_cnt; }; int vfio_device_set_group(struct vfio_device *device,
for counting the devices that are opened via the cdev path. This count is increased and decreased by the cdev path. The group path checks it to achieve exclusion with the cdev path. With this, only one path (group path or cdev path) will claim DMA ownership. This avoids scenarios in which devices within the same group may be opened via different paths. Signed-off-by: Yi Liu <yi.l.liu@intel.com> --- drivers/vfio/group.c | 5 +++++ drivers/vfio/vfio.h | 1 + 2 files changed, 6 insertions(+)