• Parav Pandit's avatar
    vfio/mdev: Improve the create/remove sequence · 522ecce0
    Parav Pandit authored
    This patch addresses below two issues and prepares the code to address
    3rd issue listed below.
    
    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, mdev driver's probe() gets
    triggered.  However there isn't a stable mdev available to work on.
    
       create_store()
         mdev_create_device()
           device_register()
              ...
             vfio_mdev_probe()
            [...]
            parent->ops->create()
              vfio_ap_mdev_create()
                mdev_set_drvdata(mdev, matrix_mdev);
                /* Valid pointer set above */
    
    Due to this way of initialization, mdev driver who wants to use the mdev,
    doesn't have a valid mdev to work on.
    
    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 fail because taking the
    device off the bus should terminate any usage.
    
    3. When remove operation fails, mdev sysfs removal attempts to add the
    file back on already removed device. Following call trace [1] is observed.
    
    [1] call trace:
    kernel: WARNING: CPU: 2 PID: 9348 at fs/sysfs/file.c:327 sysfs_create_file_ns+0x7f/0x90
    kernel: CPU: 2 PID: 9348 Comm: bash Kdump: loaded Not tainted 5.1.0-rc6-vdevbus+ #6
    kernel: Hardware name: Supermicro SYS-6028U-TR4+/X10DRU-i+, BIOS 2.0b 08/09/2016
    kernel: RIP: 0010:sysfs_create_file_ns+0x7f/0x90
    kernel: Call Trace:
    kernel: remove_store+0xdc/0x100 [mdev]
    kernel: kernfs_fop_write+0x113/0x1a0
    kernel: vfs_write+0xad/0x1b0
    kernel: ksys_write+0x5a/0xe0
    kernel: do_syscall_64+0x5a/0x210
    kernel: entry_SYSCALL_64_after_hwframe+0x49/0xbe
    
    Therefore, mdev core is improved in following ways.
    
    1. Split the device registration/deregistration sequence so that some
    things can be done between initialization of the device and hooking it
    up to the bus respectively after deregistering it from the bus but
    before giving up our final reference.
    In particular, this means invoking the ->create() and ->remove()
    callbacks in those new windows. This gives the vendor driver an
    initialized mdev device to work with during creation.
    At the same time, a bus driver who wish to bind to mdev driver also
    gets initialized mdev device.
    
    This follows standard Linux kernel bus and device model.
    
    2. During remove flow, first remove the device from the bus. This
    ensures that any bus specific devices are removed.
    Once device is taken off the mdev bus, invoke remove() of mdev
    from the vendor driver.
    
    3. The driver core device model provides way to register and auto
    unregister the device sysfs attribute groups at dev->groups.
    Make use of dev->groups to let core create the groups and eliminate
    code to avoid explicit groups creation and removal.
    
    To ensure, that new sequence is solid, a below stack dump of a
    process is taken who attempts to remove the device while device is in
    use by vfio driver and user application.
    This stack dump validates that vfio driver guards against such device
    removal when device is in use.
    
     cat /proc/21962/stack
    [<0>] vfio_del_group_dev+0x216/0x3c0 [vfio]
    [<0>] mdev_remove+0x21/0x40 [mdev]
    [<0>] device_release_driver_internal+0xe8/0x1b0
    [<0>] bus_remove_device+0xf9/0x170
    [<0>] device_del+0x168/0x350
    [<0>] mdev_device_remove_common+0x1d/0x50 [mdev]
    [<0>] mdev_device_remove+0x8c/0xd0 [mdev]
    [<0>] remove_store+0x71/0x90 [mdev]
    [<0>] kernfs_fop_write+0x113/0x1a0
    [<0>] vfs_write+0xad/0x1b0
    [<0>] ksys_write+0x5a/0xe0
    [<0>] do_syscall_64+0x5a/0x210
    [<0>] entry_SYSCALL_64_after_hwframe+0x49/0xbe
    [<0>] 0xffffffffffffffff
    
    This prepares the code to eliminate calling device_create_file() in
    subsequent patch.
    Reviewed-by: default avatarCornelia Huck <cohuck@redhat.com>
    Signed-off-by: default avatarParav Pandit <parav@mellanox.com>
    Signed-off-by: default avatarAlex Williamson <alex.williamson@redhat.com>
    522ecce0
mdev_sysfs.c 6.38 KB