Skip to content

Commit 488c10b

Browse files
Farhan Aligregkh
authored andcommitted
vfio: Fix WARNING "do not call blocking ops when !TASK_RUNNING"
[ Upstream commit 41be3e2 ] vfio_dev_present() which is the condition to wait_event_interruptible_timeout(), will call vfio_group_get_device and try to acquire the mutex group->device_lock. wait_event_interruptible_timeout() will set the state of the current task to TASK_INTERRUPTIBLE, before doing the condition check. This means that we will try to acquire the mutex while already in a sleeping state. The scheduler warns us by giving the following warning: [ 4050.264464] ------------[ cut here ]------------ [ 4050.264508] do not call blocking ops when !TASK_RUNNING; state=1 set at [<00000000b33c00e2>] prepare_to_wait_event+0x14a/0x188 [ 4050.264529] WARNING: CPU: 12 PID: 35924 at kernel/sched/core.c:6112 __might_sleep+0x76/0x90 .... 4050.264756] Call Trace: [ 4050.264765] ([<000000000017bbaa>] __might_sleep+0x72/0x90) [ 4050.264774] [<0000000000b97edc>] __mutex_lock+0x44/0x8c0 [ 4050.264782] [<0000000000b9878a>] mutex_lock_nested+0x32/0x40 [ 4050.264793] [<000003ff800d7abe>] vfio_group_get_device+0x36/0xa8 [vfio] [ 4050.264803] [<000003ff800d87c0>] vfio_del_group_dev+0x238/0x378 [vfio] [ 4050.264813] [<000003ff8015f67c>] mdev_remove+0x3c/0x68 [mdev] [ 4050.264825] [<00000000008e01b0>] device_release_driver_internal+0x168/0x268 [ 4050.264834] [<00000000008de692>] bus_remove_device+0x162/0x190 [ 4050.264843] [<00000000008daf42>] device_del+0x1e2/0x368 [ 4050.264851] [<00000000008db12c>] device_unregister+0x64/0x88 [ 4050.264862] [<000003ff8015ed84>] mdev_device_remove+0xec/0x130 [mdev] [ 4050.264872] [<000003ff8015f074>] remove_store+0x6c/0xa8 [mdev] [ 4050.264881] [<000000000046f494>] kernfs_fop_write+0x14c/0x1f8 [ 4050.264890] [<00000000003c1530>] __vfs_write+0x38/0x1a8 [ 4050.264899] [<00000000003c187c>] vfs_write+0xb4/0x198 [ 4050.264908] [<00000000003c1af2>] ksys_write+0x5a/0xb0 [ 4050.264916] [<0000000000b9e270>] system_call+0xdc/0x2d8 [ 4050.264925] 4 locks held by sh/35924: [ 4050.264933] #0: 000000001ef90325 (sb_writers#4){.+.+}, at: vfs_write+0x9e/0x198 [ 4050.264948] #1: 000000005c1ab0b3 (&of->mutex){+.+.}, at: kernfs_fop_write+0x1cc/0x1f8 [ 4050.264963] #2: 0000000034831ab8 (kn->count#297){++++}, at: kernfs_remove_self+0x12e/0x150 [ 4050.264979] #3: 00000000e152484f (&dev->mutex){....}, at: device_release_driver_internal+0x5c/0x268 [ 4050.264993] Last Breaking-Event-Address: [ 4050.265002] [<000000000017bbaa>] __might_sleep+0x72/0x90 [ 4050.265010] irq event stamp: 7039 [ 4050.265020] hardirqs last enabled at (7047): [<00000000001cee7a>] console_unlock+0x6d2/0x740 [ 4050.265029] hardirqs last disabled at (7054): [<00000000001ce87e>] console_unlock+0xd6/0x740 [ 4050.265040] softirqs last enabled at (6416): [<0000000000b8fe26>] __udelay+0xb6/0x100 [ 4050.265049] softirqs last disabled at (6415): [<0000000000b8fe06>] __udelay+0x96/0x100 [ 4050.265057] ---[ end trace d04a07d39d99a9f9 ]--- Let's fix this as described in the article https://lwn.net/Articles/628628/. Signed-off-by: Farhan Ali <alifm@linux.ibm.com> [remove now redundant vfio_dev_present()] Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
1 parent 76f53b8 commit 488c10b

1 file changed

Lines changed: 10 additions & 20 deletions

File tree

‎drivers/vfio/vfio.c‎

Lines changed: 10 additions & 20 deletions
Original file line numberDiff line numberDiff line change
@@ -34,6 +34,7 @@
3434
#include <linux/uaccess.h>
3535
#include <linux/vfio.h>
3636
#include <linux/wait.h>
37+
#include <linux/sched/signal.h>
3738

3839
#define DRIVER_VERSION "0.3"
3940
#define DRIVER_AUTHOR "Alex Williamson <alex.williamson@redhat.com>"
@@ -909,30 +910,17 @@ void *vfio_device_data(struct vfio_device *device)
909910
}
910911
EXPORT_SYMBOL_GPL(vfio_device_data);
911912

912-
/* Given a referenced group, check if it contains the device */
913-
static bool vfio_dev_present(struct vfio_group *group, struct device *dev)
914-
{
915-
struct vfio_device *device;
916-
917-
device = vfio_group_get_device(group, dev);
918-
if (!device)
919-
return false;
920-
921-
vfio_device_put(device);
922-
return true;
923-
}
924-
925913
/*
926914
* Decrement the device reference count and wait for the device to be
927915
* removed. Open file descriptors for the device... */
928916
void *vfio_del_group_dev(struct device *dev)
929917
{
918+
DEFINE_WAIT_FUNC(wait, woken_wake_function);
930919
struct vfio_device *device = dev_get_drvdata(dev);
931920
struct vfio_group *group = device->group;
932921
void *device_data = device->device_data;
933922
struct vfio_unbound_dev *unbound;
934923
unsigned int i = 0;
935-
long ret;
936924
bool interrupted = false;
937925

938926
/*
@@ -969,6 +957,8 @@ void *vfio_del_group_dev(struct device *dev)
969957
* interval with counter to allow the driver to take escalating
970958
* measures to release the device if it has the ability to do so.
971959
*/
960+
add_wait_queue(&vfio.release_q, &wait);
961+
972962
do {
973963
device = vfio_group_get_device(group, dev);
974964
if (!device)
@@ -980,12 +970,10 @@ void *vfio_del_group_dev(struct device *dev)
980970
vfio_device_put(device);
981971

982972
if (interrupted) {
983-
ret = wait_event_timeout(vfio.release_q,
984-
!vfio_dev_present(group, dev), HZ * 10);
973+
wait_woken(&wait, TASK_UNINTERRUPTIBLE, HZ * 10);
985974
} else {
986-
ret = wait_event_interruptible_timeout(vfio.release_q,
987-
!vfio_dev_present(group, dev), HZ * 10);
988-
if (ret == -ERESTARTSYS) {
975+
wait_woken(&wait, TASK_INTERRUPTIBLE, HZ * 10);
976+
if (signal_pending(current)) {
989977
interrupted = true;
990978
dev_warn(dev,
991979
"Device is currently in use, task"
@@ -994,8 +982,10 @@ void *vfio_del_group_dev(struct device *dev)
994982
current->comm, task_pid_nr(current));
995983
}
996984
}
997-
} while (ret <= 0);
998985

986+
} while (1);
987+
988+
remove_wait_queue(&vfio.release_q, &wait);
999989
/*
1000990
* In order to support multiple devices per group, devices can be
1001991
* plucked from the group while other devices in the group are still

0 commit comments

Comments
 (0)