On Fri, 22 Mar 2019 18:20:35 -0500
Parav Pandit <pa...@mellanox.com> wrote:

> There are five problems with current code structure.
> 1. mdev device is placed on the mdev bus before it is created in the
> vendor driver. Once a device is placed on the mdev bus without creating
> its supporting underlying vendor device, an open() can get triggered by
> userspace on partially initialized device.
> Below ladder diagram highlight it.
> 
>       cpu-0                                       cpu-1
>       -----                                       -----
>    create_store()
>      mdev_create_device()
>        device_register()
>           ...
>          vfio_mdev_probe()
>          ...creates char device
>                                         vfio_mdev_open()
>                                           parent->ops->open(mdev)
>                                             vfio_ap_mdev_open()
>                                               matrix_mdev = NULL
>         [...]
>         parent->ops->create()
>           vfio_ap_mdev_create()
>             mdev_set_drvdata(mdev, matrix_mdev);
>             /* Valid pointer set above */
> 
> 2. Current creation sequence is,
>    parent->ops_create()
>    groups_register()
> 
> Remove sequence is,
>    parent->ops->remove()
>    groups_unregister()
> However, remove sequence should be exact mirror of creation sequence.
> Once this is achieved, all users of the mdev will be terminated first
> before removing underlying vendor device.
> (Follow standard linux driver model).
> At that point vendor's remove() ops shouldn't failed because device is
> taken off the bus that should terminate the users.
> 
> 3. Additionally any new mdev driver that wants to work on mdev device
> during probe() routine registered using mdev_register_driver() needs to
> get stable mdev structure.
> 
> 4. In following sequence, child devices created while removing mdev parent
> device can be left out, or it may lead to race of removing half
> initialized child mdev devices.
> 
> issue-1:
> --------
>        cpu-0                         cpu-1
>        -----                         -----
>                                   mdev_unregister_device()
>                                      device_for_each_child()
>                                         mdev_device_remove_cb()
>                                             mdev_device_remove()
> create_store()
>   mdev_device_create()                   [...]
>        device_register()
>                                   parent_remove_sysfs_files()
>                                   /* BUG: device added by cpu-0
>                                    * whose parent is getting removed.
>                                    */
> 
> issue-2:
> --------
>        cpu-0                         cpu-1
>        -----                         -----
> create_store()
>   mdev_device_create()                   [...]
>        device_register()
> 
>        [...]                      mdev_unregister_device()
>                                      device_for_each_child()
>                                         mdev_device_remove_cb()
>                                             mdev_device_remove()
> 
>        mdev_create_sysfs_files()
>        /* BUG: create is adding
>         * sysfs files for a device
>         * which is undergoing removal.
>         */
>                                  parent_remove_sysfs_files()

In both cases above, it looks like the device will hold a reference to
the parent, so while there is a race, the parent object isn't released.

> 
> 5. Below crash is observed when user initiated remove is in progress
> and mdev_unregister_driver() completes parent unregistration.
> 
>        cpu-0                         cpu-1
>        -----                         -----
> remove_store()
>    mdev_device_remove()
>    active = false;
>                                   mdev_unregister_device()
>                                     remove type
>    [...]
>    mdev_remove_ops() crashes.
> 
> This is similar race like create() racing with mdev_unregister_device().

Not sure I catch this, the device should have a reference to the
parent, and we don't specifically clear parent->ops, so what's getting
removed that causes this oops?  Is .remove pointing at bad text
regardless?

> mtty mtty: MDEV: Registered
> iommu: Adding device 83b8f4f2-509f-382f-3c1e-e6bfe0fa1001 to group 57
> vfio_mdev 83b8f4f2-509f-382f-3c1e-e6bfe0fa1001: MDEV: group_id = 57
> mdev_device_remove sleep started
> mtty mtty: MDEV: Unregistering
> mtty_dev: Unloaded!
> BUG: unable to handle kernel paging request at ffffffffc027d668
> PGD af9818067 P4D af9818067 PUD af981a067 PMD 8583c3067 PTE 0
> Oops: 0000 [#1] SMP PTI
> CPU: 15 PID: 3517 Comm: bash Kdump: loaded Not tainted 5.0.0-rc7-vdevbus+ #2
> Hardware name: Supermicro SYS-6028U-TR4+/X10DRU-i+, BIOS 2.0b 08/09/2016
> RIP: 0010:mdev_device_remove_ops+0x1a/0x50 [mdev]
> Call Trace:
>  mdev_device_remove+0xef/0x130 [mdev]
>  remove_store+0x77/0xa0 [mdev]
>  kernfs_fop_write+0x113/0x1a0
>  __vfs_write+0x33/0x1b0
>  ? rcu_read_lock_sched_held+0x64/0x70
>  ? rcu_sync_lockdep_assert+0x2a/0x50
>  ? __sb_start_write+0x121/0x1b0
>  ? vfs_write+0x17c/0x1b0
>  vfs_write+0xad/0x1b0
>  ? trace_hardirqs_on_thunk+0x1a/0x1c
>  ksys_write+0x55/0xc0
>  do_syscall_64+0x5a/0x210
> 
> Therefore, mdev core is improved in following ways to overcome above
> issues.
> 
> 1. Before placing mdev devices on the bus, perform vendor drivers
> creation which supports the mdev creation.
> This ensures that mdev specific all necessary fields are initialized
> before a given mdev can be accessed by bus driver.
> 
> 2. During remove flow, first remove the device from the bus. This
> ensures that any bus specific devices and data is cleared.
> Once device is taken of the mdev bus, perform remove() of mdev from the
> vendor driver.
> 
> 3. Linux core device model provides way to register and auto unregister
> the device sysfs attribute groups at dev->groups.
> Make use of this groups to let core create the groups and simplify code
> to avoid explicit groups creation and removal.
> 
> 4. Wait for any ongoing mdev create() and remove() to finish before
> unregistering parent device using srcu. This continues to allow multiple
> create and remove to progress in parallel. At the same time guard parent
> removal while parent is being access by create() and remove callbacks.

So there should be 4-5 separate patches here?  Wishful thinking?

> Fixes: 7b96953bc640 ("vfio: Mediated device Core driver")
> Signed-off-by: Parav Pandit <pa...@mellanox.com>
> ---
>  drivers/vfio/mdev/mdev_core.c    | 142 
> +++++++++++++++++++++------------------
>  drivers/vfio/mdev/mdev_private.h |   7 +-
>  drivers/vfio/mdev/mdev_sysfs.c   |   6 +-
>  3 files changed, 84 insertions(+), 71 deletions(-)
> 
> diff --git a/drivers/vfio/mdev/mdev_core.c b/drivers/vfio/mdev/mdev_core.c
> index 944a058..8fe0ed1 100644
> --- a/drivers/vfio/mdev/mdev_core.c
> +++ b/drivers/vfio/mdev/mdev_core.c
> @@ -84,6 +84,7 @@ static void mdev_release_parent(struct kref *kref)
>                                                 ref);
>       struct device *dev = parent->dev;
>  
> +     cleanup_srcu_struct(&parent->unreg_srcu);
>       kfree(parent);
>       put_device(dev);
>  }
> @@ -103,56 +104,30 @@ static inline void mdev_put_parent(struct mdev_parent 
> *parent)
>               kref_put(&parent->ref, mdev_release_parent);
>  }
>  
> -static int mdev_device_create_ops(struct kobject *kobj,
> -                               struct mdev_device *mdev)
> +static int mdev_device_must_remove(struct mdev_device *mdev)

Naming is off here, mdev_device_remove_common()?

>  {
> -     struct mdev_parent *parent = mdev->parent;
> +     struct mdev_parent *parent;
> +     struct mdev_type *type;
>       int ret;
>  
> -     ret = parent->ops->create(kobj, mdev);
> -     if (ret)
> -             return ret;
> +     type = to_mdev_type(mdev->type_kobj);
>  
> -     ret = sysfs_create_groups(&mdev->dev.kobj,
> -                               parent->ops->mdev_attr_groups);
> +     mdev_remove_sysfs_files(&mdev->dev, type);
> +     device_del(&mdev->dev);
> +     parent = mdev->parent;
> +     ret = parent->ops->remove(mdev);
>       if (ret)
> -             parent->ops->remove(mdev);
> +             dev_err(&mdev->dev, "Remove failed: err=%d\n", ret);

Let the caller decide whether to be verbose with the error, parent
removal might want to warn, sysfs remove might just return an error.

>  
> +     /* Balances with device_initialize() */
> +     put_device(&mdev->dev);
>       return ret;
>  }
>  
> -/*
> - * mdev_device_remove_ops gets called from sysfs's 'remove' and when parent
> - * device is being unregistered from mdev device framework.
> - * - 'force_remove' is set to 'false' when called from sysfs's 'remove' which
> - *   indicates that if the mdev device is active, used by VMM or userspace
> - *   application, vendor driver could return error then don't remove the 
> device.
> - * - 'force_remove' is set to 'true' when called from 
> mdev_unregister_device()
> - *   which indicate that parent device is being removed from mdev device
> - *   framework so remove mdev device forcefully.
> - */
> -static int mdev_device_remove_ops(struct mdev_device *mdev, bool 
> force_remove)
> -{
> -     struct mdev_parent *parent = mdev->parent;
> -     int ret;
> -
> -     /*
> -      * Vendor driver can return error if VMM or userspace application is
> -      * using this mdev device.
> -      */
> -     ret = parent->ops->remove(mdev);
> -     if (ret && !force_remove)
> -             return ret;
> -
> -     sysfs_remove_groups(&mdev->dev.kobj, parent->ops->mdev_attr_groups);
> -     return 0;
> -}

Seems like there's easily a separate patch in pushing the create/remove
ops into the calling function and separating for the iterator callback,
that would make this easier to review.

> -
>  static int mdev_device_remove_cb(struct device *dev, void *data)
>  {
>       if (dev_is_mdev(dev))
> -             mdev_device_remove(dev, true);
> -
> +             mdev_device_must_remove(to_mdev_device(dev));
>       return 0;
>  }
>  
> @@ -194,6 +169,7 @@ int mdev_register_device(struct device *dev, const struct 
> mdev_parent_ops *ops)
>       }
>  
>       kref_init(&parent->ref);
> +     init_srcu_struct(&parent->unreg_srcu);
>  
>       parent->dev = dev;
>       parent->ops = ops;
> @@ -214,6 +190,7 @@ int mdev_register_device(struct device *dev, const struct 
> mdev_parent_ops *ops)
>       if (ret)
>               dev_warn(dev, "Failed to create compatibility class link\n");
>  
> +     rcu_assign_pointer(parent->self, parent);
>       list_add(&parent->next, &parent_list);
>       mutex_unlock(&parent_list_lock);
>  
> @@ -244,21 +221,36 @@ void mdev_unregister_device(struct device *dev)
>  
>       mutex_lock(&parent_list_lock);
>       parent = __find_parent_device(dev);
> -
>       if (!parent) {
>               mutex_unlock(&parent_list_lock);
>               return;
>       }
> +     list_del(&parent->next);
> +     mutex_unlock(&parent_list_lock);
> +
>       dev_info(dev, "MDEV: Unregistering\n");
>  
> -     list_del(&parent->next);
> +     /* Publish that this mdev parent is unregistering. So any new
> +      * create/remove cannot start on this parent anymore by user.
> +      */

Comment style, we're not in netdev.

> +     rcu_assign_pointer(parent->self, NULL);
> +
> +     /*
> +      * Wait for any active create() or remove() mdev ops on the parent
> +      * to complete.
> +      */
> +     synchronize_srcu(&parent->unreg_srcu);
> +
> +     /* At this point it is confirmed that any pending user initiated
> +      * create or remove callbacks accessing the parent are completed.
> +      * It is safe to remove the parent now.
> +      */
>       class_compat_remove_link(mdev_bus_compat_class, dev, NULL);
>  
>       device_for_each_child(dev, NULL, mdev_device_remove_cb);
>  
>       parent_remove_sysfs_files(parent);
>  
> -     mutex_unlock(&parent_list_lock);
>       mdev_put_parent(parent);
>  }
>  EXPORT_SYMBOL(mdev_unregister_device);
> @@ -278,14 +270,24 @@ static void mdev_device_release(struct device *dev)
>  int mdev_device_create(struct kobject *kobj, struct device *dev, uuid_le 
> uuid)
>  {
>       int ret;
> +     struct mdev_parent *valid_parent;
>       struct mdev_device *mdev, *tmp;
>       struct mdev_parent *parent;
>       struct mdev_type *type = to_mdev_type(kobj);
> +     int srcu_idx;
>  
>       parent = mdev_get_parent(type->parent);
>       if (!parent)
>               return -EINVAL;
>  
> +     srcu_idx = srcu_read_lock(&parent->unreg_srcu);
> +     valid_parent = srcu_dereference(parent->self, &parent->unreg_srcu);
> +     if (!valid_parent) {
> +             /* parent is undergoing unregistration */
> +             ret = -ENODEV;
> +             goto mdev_fail;
> +     }
> +
>       mutex_lock(&mdev_list_lock);
>  
>       /* Check for duplicate */
> @@ -310,68 +312,76 @@ int mdev_device_create(struct kobject *kobj, struct 
> device *dev, uuid_le uuid)
>  
>       mdev->parent = parent;
>  
> +     device_initialize(&mdev->dev);
>       mdev->dev.parent  = dev;
>       mdev->dev.bus     = &mdev_bus_type;
>       mdev->dev.release = mdev_device_release;
> +     mdev->dev.groups = type->parent->ops->mdev_attr_groups;
>       dev_set_name(&mdev->dev, "%pUl", uuid.b);
>  
> -     ret = device_register(&mdev->dev);
> +     ret = type->parent->ops->create(kobj, mdev);
>       if (ret)
> -             goto mdev_fail;
> +             goto create_fail;
>  
> -     ret = mdev_device_create_ops(kobj, mdev);
> +     ret = device_add(&mdev->dev);

Separating device_initialize() and device_add() also looks like a
separate patch, then the srcu could be added at the end.  Thanks,

Alex

>       if (ret)
> -             goto create_fail;
> +             goto dev_fail;
>  
>       ret = mdev_create_sysfs_files(&mdev->dev, type);
> -     if (ret) {
> -             mdev_device_remove_ops(mdev, true);
> -             goto create_fail;
> -     }
> +     if (ret)
> +             goto sysfs_fail;
>  
>       mdev->type_kobj = kobj;
>       mdev->active = true;
>       dev_dbg(&mdev->dev, "MDEV: created\n");
> +     srcu_read_unlock(&parent->unreg_srcu, srcu_idx);
>  
>       return 0;
>  
> +sysfs_fail:
> +     device_del(&mdev->dev);
> +dev_fail:
> +     type->parent->ops->remove(mdev);
>  create_fail:
> -     device_unregister(&mdev->dev);
> +     put_device(&mdev->dev);
>  mdev_fail:
> +     srcu_read_unlock(&parent->unreg_srcu, srcu_idx);
>       mdev_put_parent(parent);
>       return ret;
>  }
>  
> -int mdev_device_remove(struct device *dev, bool force_remove)
> +int mdev_device_remove(struct device *dev)
>  {
> +     struct mdev_parent *valid_parent;
>       struct mdev_device *mdev;
>       struct mdev_parent *parent;
> -     struct mdev_type *type;
> +     int srcu_idx;
>       int ret;
>  
>       mdev = to_mdev_device(dev);
> +     parent = mdev->parent;
> +     srcu_idx = srcu_read_lock(&parent->unreg_srcu);
> +     valid_parent = srcu_dereference(parent->self, &parent->unreg_srcu);
> +     if (!valid_parent) {
> +             srcu_read_unlock(&parent->unreg_srcu, srcu_idx);
> +             /* parent is undergoing unregistration */
> +             return -ENODEV;
> +     }
> +
> +     mutex_lock(&mdev_list_lock);
>       if (!mdev->active) {
>               mutex_unlock(&mdev_list_lock);
> -             return -EAGAIN;
> +             srcu_read_unlock(&parent->unreg_srcu, srcu_idx);
> +             return -ENODEV;
>       }
> -
>       mdev->active = false;
>       mutex_unlock(&mdev_list_lock);
>  
> -     type = to_mdev_type(mdev->type_kobj);
> -     parent = mdev->parent;
> -
> -     ret = mdev_device_remove_ops(mdev, force_remove);
> -     if (ret) {
> -             mdev->active = true;
> -             return ret;
> -     }
> +     ret = mdev_device_must_remove(mdev);
> +     srcu_read_unlock(&parent->unreg_srcu, srcu_idx);
>  
> -     mdev_remove_sysfs_files(dev, type);
> -     device_unregister(dev);
>       mdev_put_parent(parent);
> -
> -     return 0;
> +     return ret;
>  }
>  
>  static int __init mdev_init(void)
> diff --git a/drivers/vfio/mdev/mdev_private.h 
> b/drivers/vfio/mdev/mdev_private.h
> index 84b2b6c..3d17db9 100644
> --- a/drivers/vfio/mdev/mdev_private.h
> +++ b/drivers/vfio/mdev/mdev_private.h
> @@ -23,6 +23,11 @@ struct mdev_parent {
>       struct list_head next;
>       struct kset *mdev_types_kset;
>       struct list_head type_list;
> +     /* Protects unregistration to wait until create/remove
> +      * are completed.
> +      */
> +     struct srcu_struct unreg_srcu;
> +     struct mdev_parent __rcu *self;
>  };
>  
>  struct mdev_device {
> @@ -58,6 +63,6 @@ struct mdev_type {
>  void mdev_remove_sysfs_files(struct device *dev, struct mdev_type *type);
>  
>  int  mdev_device_create(struct kobject *kobj, struct device *dev, uuid_le 
> uuid);
> -int  mdev_device_remove(struct device *dev, bool force_remove);
> +int  mdev_device_remove(struct device *dev);
>  
>  #endif /* MDEV_PRIVATE_H */
> diff --git a/drivers/vfio/mdev/mdev_sysfs.c b/drivers/vfio/mdev/mdev_sysfs.c
> index c782fa9..68a8191 100644
> --- a/drivers/vfio/mdev/mdev_sysfs.c
> +++ b/drivers/vfio/mdev/mdev_sysfs.c
> @@ -236,11 +236,9 @@ static ssize_t remove_store(struct device *dev, struct 
> device_attribute *attr,
>       if (val && device_remove_file_self(dev, attr)) {
>               int ret;
>  
> -             ret = mdev_device_remove(dev, false);
> -             if (ret) {
> -                     device_create_file(dev, attr);
> +             ret = mdev_device_remove(dev);
> +             if (ret)
>                       return ret;
> -             }
>       }
>  
>       return count;

Reply via email to