Commit d2b59bd4 authored by Alexandra Winter's avatar Alexandra Winter Committed by Jakub Kicinski
Browse files

s390/qeth: fix deadlock during failing recovery



Commit 0b9902c1 ("s390/qeth: fix deadlock during recovery") removed
taking discipline_mutex inside qeth_do_reset(), fixing potential
deadlocks. An error path was missed though, that still takes
discipline_mutex and thus has the original deadlock potential.

Intermittent deadlocks were seen when a qeth channel path is configured
offline, causing a race between qeth_do_reset and ccwgroup_remove.
Call qeth_set_offline() directly in the qeth_do_reset() error case and
then a new variant of ccwgroup_set_offline(), without taking
discipline_mutex.

Fixes: b41b554c ("s390/qeth: fix locking for discipline setup / removal")
Signed-off-by: default avatarAlexandra Winter <wintera@linux.ibm.com>
Reviewed-by: default avatarJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: default avatarJulian Wiedmann <jwi@linux.ibm.com>
Signed-off-by: default avatarJakub Kicinski <kuba@kernel.org>
parent ee909d0b
Loading
Loading
Loading
Loading
+1 −1
Original line number Original line Diff line number Diff line
@@ -55,7 +55,7 @@ int ccwgroup_create_dev(struct device *root, struct ccwgroup_driver *gdrv,
			int num_devices, const char *buf);
			int num_devices, const char *buf);


extern int ccwgroup_set_online(struct ccwgroup_device *gdev);
extern int ccwgroup_set_online(struct ccwgroup_device *gdev);
extern int ccwgroup_set_offline(struct ccwgroup_device *gdev);
int ccwgroup_set_offline(struct ccwgroup_device *gdev, bool call_gdrv);


extern int ccwgroup_probe_ccwdev(struct ccw_device *cdev);
extern int ccwgroup_probe_ccwdev(struct ccw_device *cdev);
extern void ccwgroup_remove_ccwdev(struct ccw_device *cdev);
extern void ccwgroup_remove_ccwdev(struct ccw_device *cdev);
+8 −2
Original line number Original line Diff line number Diff line
@@ -77,12 +77,13 @@ EXPORT_SYMBOL(ccwgroup_set_online);
/**
/**
 * ccwgroup_set_offline() - disable a ccwgroup device
 * ccwgroup_set_offline() - disable a ccwgroup device
 * @gdev: target ccwgroup device
 * @gdev: target ccwgroup device
 * @call_gdrv: Call the registered gdrv set_offline function
 *
 *
 * This function attempts to put the ccwgroup device into the offline state.
 * This function attempts to put the ccwgroup device into the offline state.
 * Returns:
 * Returns:
 *  %0 on success and a negative error value on failure.
 *  %0 on success and a negative error value on failure.
 */
 */
int ccwgroup_set_offline(struct ccwgroup_device *gdev)
int ccwgroup_set_offline(struct ccwgroup_device *gdev, bool call_gdrv)
{
{
	struct ccwgroup_driver *gdrv = to_ccwgroupdrv(gdev->dev.driver);
	struct ccwgroup_driver *gdrv = to_ccwgroupdrv(gdev->dev.driver);
	int ret = -EINVAL;
	int ret = -EINVAL;
@@ -91,11 +92,16 @@ int ccwgroup_set_offline(struct ccwgroup_device *gdev)
		return -EAGAIN;
		return -EAGAIN;
	if (gdev->state == CCWGROUP_OFFLINE)
	if (gdev->state == CCWGROUP_OFFLINE)
		goto out;
		goto out;
	if (!call_gdrv) {
		ret = 0;
		goto offline;
	}
	if (gdrv->set_offline)
	if (gdrv->set_offline)
		ret = gdrv->set_offline(gdev);
		ret = gdrv->set_offline(gdev);
	if (ret)
	if (ret)
		goto out;
		goto out;


offline:
	gdev->state = CCWGROUP_OFFLINE;
	gdev->state = CCWGROUP_OFFLINE;
out:
out:
	atomic_set(&gdev->onoff, 0);
	atomic_set(&gdev->onoff, 0);
@@ -124,7 +130,7 @@ static ssize_t ccwgroup_online_store(struct device *dev,
	if (value == 1)
	if (value == 1)
		ret = ccwgroup_set_online(gdev);
		ret = ccwgroup_set_online(gdev);
	else if (value == 0)
	else if (value == 0)
		ret = ccwgroup_set_offline(gdev);
		ret = ccwgroup_set_offline(gdev, true);
	else
	else
		ret = -EINVAL;
		ret = -EINVAL;
out:
out:
+2 −1
Original line number Original line Diff line number Diff line
@@ -5514,7 +5514,8 @@ static int qeth_do_reset(void *data)
		dev_info(&card->gdev->dev,
		dev_info(&card->gdev->dev,
			 "Device successfully recovered!\n");
			 "Device successfully recovered!\n");
	} else {
	} else {
		ccwgroup_set_offline(card->gdev);
		qeth_set_offline(card, disc, true);
		ccwgroup_set_offline(card->gdev, false);
		dev_warn(&card->gdev->dev,
		dev_warn(&card->gdev->dev,
			 "The qeth device driver failed to recover an error on the device\n");
			 "The qeth device driver failed to recover an error on the device\n");
	}
	}