[PATCH 4.9 32/49] scsi: core: try to get module before removing device

From: Greg Kroah-Hartman
Date: Sun Oct 27 2019 - 17:36:30 EST


From: Yufen Yu <yuyufen@xxxxxxxxxx>

commit 77c301287ebae86cc71d03eb3806f271cb14da79 upstream.

We have a test case like block/001 in blktests, which will create a scsi
device by loading scsi_debug module and then try to delete the device by
sysfs interface. At the same time, it may remove the scsi_debug module.

And getting a invalid paging request BUG_ON as following:

[ 34.625854] BUG: unable to handle page fault for address: ffffffffa0016bb8
[ 34.629189] Oops: 0000 [#1] SMP PTI
[ 34.629618] CPU: 1 PID: 450 Comm: bash Tainted: G W 5.4.0-rc3+ #473
[ 34.632524] RIP: 0010:scsi_proc_hostdir_rm+0x5/0xa0
[ 34.643555] CR2: ffffffffa0016bb8 CR3: 000000012cd88000 CR4: 00000000000006e0
[ 34.644545] Call Trace:
[ 34.644907] scsi_host_dev_release+0x6b/0x1f0
[ 34.645511] device_release+0x74/0x110
[ 34.646046] kobject_put+0x116/0x390
[ 34.646559] put_device+0x17/0x30
[ 34.647041] scsi_target_dev_release+0x2b/0x40
[ 34.647652] device_release+0x74/0x110
[ 34.648186] kobject_put+0x116/0x390
[ 34.648691] put_device+0x17/0x30
[ 34.649157] scsi_device_dev_release_usercontext+0x2e8/0x360
[ 34.649953] execute_in_process_context+0x29/0x80
[ 34.650603] scsi_device_dev_release+0x20/0x30
[ 34.651221] device_release+0x74/0x110
[ 34.651732] kobject_put+0x116/0x390
[ 34.652230] sysfs_unbreak_active_protection+0x3f/0x50
[ 34.652935] sdev_store_delete.cold.4+0x71/0x8f
[ 34.653579] dev_attr_store+0x1b/0x40
[ 34.654103] sysfs_kf_write+0x3d/0x60
[ 34.654603] kernfs_fop_write+0x174/0x250
[ 34.655165] __vfs_write+0x1f/0x60
[ 34.655639] vfs_write+0xc7/0x280
[ 34.656117] ksys_write+0x6d/0x140
[ 34.656591] __x64_sys_write+0x1e/0x30
[ 34.657114] do_syscall_64+0xb1/0x400
[ 34.657627] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 34.658335] RIP: 0033:0x7f156f337130

During deleting scsi target, the scsi_debug module have been removed. Then,
sdebug_driver_template belonged to the module cannot be accessd, resulting
in scsi_proc_hostdir_rm() BUG_ON.

To fix the bug, we add scsi_device_get() in sdev_store_delete() to try to
increase refcount of module, avoiding the module been removed.

Cc: stable@xxxxxxxxxxxxxxx
Link: https://lore.kernel.org/r/20191015130556.18061-1-yuyufen@xxxxxxxxxx
Signed-off-by: Yufen Yu <yuyufen@xxxxxxxxxx>
Reviewed-by: Bart Van Assche <bvanassche@xxxxxxx>
Signed-off-by: Martin K. Petersen <martin.petersen@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
drivers/scsi/scsi_sysfs.c | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)

--- a/drivers/scsi/scsi_sysfs.c
+++ b/drivers/scsi/scsi_sysfs.c
@@ -710,6 +710,14 @@ sdev_store_delete(struct device *dev, st
const char *buf, size_t count)
{
struct kernfs_node *kn;
+ struct scsi_device *sdev = to_scsi_device(dev);
+
+ /*
+ * We need to try to get module, avoiding the module been removed
+ * during delete.
+ */
+ if (scsi_device_get(sdev))
+ return -ENODEV;

kn = sysfs_break_active_protection(&dev->kobj, &attr->attr);
WARN_ON_ONCE(!kn);
@@ -724,9 +732,10 @@ sdev_store_delete(struct device *dev, st
* state into SDEV_DEL.
*/
device_remove_file(dev, attr);
- scsi_remove_device(to_scsi_device(dev));
+ scsi_remove_device(sdev);
if (kn)
sysfs_unbreak_active_protection(kn);
+ scsi_device_put(sdev);
return count;
};
static DEVICE_ATTR(delete, S_IWUSR, NULL, sdev_store_delete);