[char-misc 4.2] mei: nfc: fix deadlock on shutdown/suspend path

[char-misc 4.2] mei: nfc: fix deadlock on shutdown/suspend path

[Tomas Winkler]

In function mei_nfc_host_exit mei_cl_remove_device cannot be called
under the device mutex as device removing flow invokes the device driver
remove handler that calls in turn to mei_cl_disable_device which
naturally acquires the device mutex.

Also remove mei_cl_bus_remove_devices which has the same issue, but is
never executed as currently the only device on the mei client bus is
NFC and a new device cannot be easily added till
the bus revamp is completed.

This fixes regression caused by commit
be9b720a0ccba096d669bc86634f900b82b9bf71
Author: Tomas Winkler <tomas.winkler@intel.com>
Date:   Thu May 7 15:54:04 2015 +0300
NFC: mei_phy: move all nfc logic from mei driver to nfc

Prior to this change the nfc driver remove handler
called to no-op disable function while actual nfc
device was disabled directly from the mei driver.

Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
---
 drivers/misc/mei/bus.c  | 16 ----------------
 drivers/misc/mei/init.c |  2 --
 drivers/misc/mei/nfc.c  |  3 ++-
 3 files changed, 2 insertions(+), 19 deletions(-)

diff --git a/drivers/misc/mei/bus.c b/drivers/misc/mei/bus.c
index 357b6ae4d207..458aa5a09c52 100644
--- a/drivers/misc/mei/bus.c
+++ b/drivers/misc/mei/bus.c
@@ -552,22 +552,6 @@ void mei_cl_bus_rx_event(struct mei_cl *cl)
 	schedule_work(&device->event_work);
 }
 
-void mei_cl_bus_remove_devices(struct mei_device *dev)
-{
-	struct mei_cl *cl, *next;
-
-	mutex_lock(&dev->device_lock);
-	list_for_each_entry_safe(cl, next, &dev->device_list, device_link) {
-		if (cl->device)
-			mei_cl_remove_device(cl->device);
-
-		list_del(&cl->device_link);
-		mei_cl_unlink(cl);
-		kfree(cl);
-	}
-	mutex_unlock(&dev->device_lock);
-}
-
 int __init mei_cl_bus_init(void)
 {
 	return bus_register(&mei_cl_bus_type);
diff --git a/drivers/misc/mei/init.c b/drivers/misc/mei/init.c
index 94514b2c7a50..00c3865ca3b1 100644
--- a/drivers/misc/mei/init.c
+++ b/drivers/misc/mei/init.c
@@ -333,8 +333,6 @@ void mei_stop(struct mei_device *dev)
 
 	mei_nfc_host_exit(dev);
 
-	mei_cl_bus_remove_devices(dev);
-
 	mutex_lock(&dev->device_lock);
 
 	mei_wd_stop(dev);
diff --git a/drivers/misc/mei/nfc.c b/drivers/misc/mei/nfc.c
index b983c4ecad38..290ef3037437 100644
--- a/drivers/misc/mei/nfc.c
+++ b/drivers/misc/mei/nfc.c
@@ -402,11 +402,12 @@ void mei_nfc_host_exit(struct mei_device *dev)
 
 	cldev->priv_data = NULL;
 
-	mutex_lock(&dev->device_lock);
 	/* Need to remove the device here
 	 * since mei_nfc_free will unlink the clients
 	 */
 	mei_cl_remove_device(cldev);
+
+	mutex_lock(&dev->device_lock);
 	mei_nfc_free(ndev);
 	mutex_unlock(&dev->device_lock);
 }
-- 
2.4.3

Re: [char-misc 4.2] mei: nfc: fix deadlock on shutdown/suspend path

[Linus Torvalds]

On Tue, Jul 7, 2015 at 2:22 PM, Tomas Winkler <tomas.winkler@intel.com> wrote:
> In function mei_nfc_host_exit mei_cl_remove_device cannot be called
> under the device mutex as device removing flow invokes the device driver
> remove handler that calls in turn to mei_cl_disable_device which
> naturally acquires the device mutex.

Yup, tested and confirmed. This patch fixes things for me without the
revert. Thanks.

Greg, should I just apply it directly, or do we go through your tree?

                   Linus

Re: [char-misc 4.2] mei: nfc: fix deadlock on shutdown/suspend path

[Greg Kroah-Hartman]

On Tue, Jul 07, 2015 at 02:40:13PM -0700, Linus Torvalds wrote:
> On Tue, Jul 7, 2015 at 2:22 PM, Tomas Winkler <tomas.winkler@intel.com> wrote:
> > In function mei_nfc_host_exit mei_cl_remove_device cannot be called
> > under the device mutex as device removing flow invokes the device driver
> > remove handler that calls in turn to mei_cl_disable_device which
> > naturally acquires the device mutex.
> 
> Yup, tested and confirmed. This patch fixes things for me without the
> revert. Thanks.
> 
> Greg, should I just apply it directly, or do we go through your tree?

Feel free to take it directly, that way it fixes your laptop now, I
don't have any other pending char-misc patches at the moment.

Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

thanks,

greg k-h