scsi: Fix a bdi reregistration race
PATCH 1/1 scsi: Fix a bdi reregistration race
Unregister and reregister BDI devices in the proper order. This patch avoids that the following kernel warning can get triggered:
WARNING: CPU: 7 PID: 203 at fs/sysfs/dir.c:31 sysfs_warn_dup+0x68/0x80() sysfs: cannot create duplicate filename '/devices/virtual/bdi/8:32' Workqueue: events_unbound async_run_entry_fn Call Trace: [<ffffffff814ff5a4>] dump_stack+0x4c/0x65 [<ffffffff810746ba>] warn_slowpath_common+0x8a/0xc0 [<ffffffff81074736>] warn_slowpath_fmt+0x46/0x50 [<ffffffff81237ca8>] sysfs_warn_dup+0x68/0x80 [<ffffffff81237d8e>] sysfs_create_dir_ns+0x7e/0x90 [<ffffffff81291f58>] kobject_add_internal+0xa8/0x320 [<ffffffff812923a0>] kobject_add+0x60/0xb0 [<ffffffff8138c937>] device_add+0x107/0x5e0 [<ffffffff8138d018>] device_create_groups_vargs+0xd8/0x100 [<ffffffff8138d05c>] device_create_vargs+0x1c/0x20 [<ffffffff8117f233>] bdi_register+0x63/0x2a0 [<ffffffff8117f497>] bdi_register_dev+0x27/0x30 [<ffffffff81281549>] add_disk+0x1a9/0x4e0 [<ffffffffa00c5739>] sd_probe_async+0x119/0x1d0 [sd_mod] [<ffffffff8109a81a>] async_run_entry_fn+0x4a/0x140 [<ffffffff81091078>] process_one_work+0x1d8/0x7c0 [<ffffffff81091774>] worker_thread+0x114/0x460 [<ffffffff81097878>] kthread+0xf8/0x110 [<ffffffff8150801f>] ret_from_fork+0x3f/0x70
See also patch "block: destroy bdi before blockdev is unregistered" (commit ID 6cd18e711dd8).
--- drivers/scsi/scsi_sysfs.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-)
drivers/scsi/scsi_sysfs.c +11 -3
@@ -1084,9 +1084,7 @@ void __scsi_remove_device(struct scsi_device *sdev)device_unregister (&sdev->sdev_dev);transport_remove_device (dev);scsi_dh_remove_device (sdev); -device_del (dev); - }else -put_device (&sdev->sdev_dev); + }/* * Stop accepting new requests and wait until all queuecommand() and
@@ -1097,6 +1095,16 @@ void __scsi_remove_device(struct scsi_device *sdev)blk_cleanup_queue (sdev->request_queue);cancel_work_sync (&sdev->requeue_work); +/* +* Remove the device after blk_cleanup_queue() has been called such +* a possible bdi_register() call with the same name occurs after +* blk_cleanup_queue() has called bdi_destroy(). +*/ +if (sdev->is_visible) +device_del (dev); +else +put_device (&sdev->sdev_dev); +if (sdev->host->hostt->slave_destroy) sdev->host->hostt->slave_destroy (sdev);transport_destroy_device (dev);
Signature · 1 line
2.1.4
Quoted 62 lines · Bart Van Assche
Unregister and reregister BDI devices in the proper order. This patch avoids that the following kernel warning can get triggered:
WARNING: CPU: 7 PID: 203 at fs/sysfs/dir.c:31 sysfs_warn_dup+0x68/0x80() sysfs: cannot create duplicate filename '/devices/virtual/bdi/8:32' Workqueue: events_unbound async_run_entry_fn Call Trace: [<ffffffff814ff5a4>] dump_stack+0x4c/0x65 [<ffffffff810746ba>] warn_slowpath_common+0x8a/0xc0 [<ffffffff81074736>] warn_slowpath_fmt+0x46/0x50 [<ffffffff81237ca8>] sysfs_warn_dup+0x68/0x80 [<ffffffff81237d8e>] sysfs_create_dir_ns+0x7e/0x90 [<ffffffff81291f58>] kobject_add_internal+0xa8/0x320 [<ffffffff812923a0>] kobject_add+0x60/0xb0 [<ffffffff8138c937>] device_add+0x107/0x5e0 [<ffffffff8138d018>] device_create_groups_vargs+0xd8/0x100 [<ffffffff8138d05c>] device_create_vargs+0x1c/0x20 [<ffffffff8117f233>] bdi_register+0x63/0x2a0 [<ffffffff8117f497>] bdi_register_dev+0x27/0x30 [<ffffffff81281549>] add_disk+0x1a9/0x4e0 [<ffffffffa00c5739>] sd_probe_async+0x119/0x1d0 [sd_mod] [<ffffffff8109a81a>] async_run_entry_fn+0x4a/0x140 [<ffffffff81091078>] process_one_work+0x1d8/0x7c0 [<ffffffff81091774>] worker_thread+0x114/0x460 [<ffffffff81097878>] kthread+0xf8/0x110 [<ffffffff8150801f>] ret_from_fork+0x3f/0x70See also patch "block: destroy bdi before blockdev is unregistered" (commit ID
6cd18e711dd8).Signed-off-by Bart Van Assche <[email protected]>Cc Hannes Reinecke <[email protected]>Cc Christoph Hellwig <[email protected]>Cc Martin K. Petersen <[email protected]>Cc <[email protected]>--- drivers/scsi/scsi_sysfs.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-)drivers/scsi/scsi_sysfs.c +4
-3@@ -1084,9 +1084,7 @@ void __scsi_remove_device(struct scsi_device *sdev)device_unregister (&sdev->sdev_dev);transport_remove_device (dev);scsi_dh_remove_device (sdev); -device_del (dev); - }else -put_device (&sdev->sdev_dev); + }/* * Stop accepting new requests and wait until all queuecommand() and @@ -1097,6 +1095,16 @@ void __scsi_remove_device(struct scsi_device *sdev)blk_cleanup_queue (sdev->request_queue);cancel_work_sync (&sdev->requeue_work); +/* +* Remove the device after blk_cleanup_queue() has been called such +* a possible bdi_register() call with the same name occurs after
drivers/scsi/scsi_sysfs.c void __scsi_remove_device(struct scsi_device *sdev) · line 1110
Quoted 10 lines
drivers/scsi/scsi_sysfs.c +7
-0+* blk_cleanup_queue() has called bdi_destroy(). +*/ +if (sdev->is_visible) +device_del (dev); +else +put_device (&sdev->sdev_dev); +if (sdev->host->hostt->slave_destroy) sdev->host->hostt->slave_destroy (sdev);transport_destroy_device (dev);
Cheers,
Hannes
Signature · 8 lines
Dr. Hannes Reinecke zSeries & Storage [email protected] +49 911 74053 688 SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: J. Hawn, J. Guild, F. Imendörffer, HRB 16746 (AG Nürnberg) -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 2 lines · Bart Van Assche
Unregister and reregister BDI devices in the proper order. This patch avoids that the following kernel warning can get triggered:
Hi Bart,
Can you share the scenario that reproduced this? I think I might have seen this before.
Thanks,
Sagi.
Quoted 6 lines · Sagi Grimberg
Quoted 2 lines · Bart Van Assche
Unregister and reregister BDI devices in the proper order. This patch avoids that the following kernel warning can get triggered:
Can you share the scenario that reproduced this? I think I might have seen this before.
Hello Sagi,
The details of the setup on which I can reproduce the reported behavior easily are as follows:
* Several kernel debugging options were enabled on the initiator system (PROVE_LOCKING, SLUB_DEBUG, KMEMLEAK, ...).
* srp_daemon and multipathd were running on the initiator system.
* Four IB ports were present the initiator system.
* Eight IB ports were present the target system.
* 100 LUNs were defined on the target system.
* As a result, 3200 /dev/sd* device nodes were created on the
initiator system by the SRP initiator driver.
* The following command was run on the initiator system:
for p in /sys/class/srp_remote_ports/*; do echo 1 >$p/delete & done;
wait; dmsetup remove_all
Bart.