linux-stable.git/drivers/nvme, branch linux-6.11.y Linux kernel stable tree nvme-fabrics: fix kernel crash while shutting down controller 2024-12-05T12:54:30+00:00 Nilay Shroff nilay@linux.ibm.com 2024-11-05T06:12:09+00:00 30794f4952decb2ec8efa42f704cac5304499a41 [ Upstream commit e9869c85c81168a1275f909d5972a3fc435304be ] The nvme keep-alive operation, which executes at a periodic interval, could potentially sneak in while shutting down a fabric controller. This may lead to a race between the fabric controller admin queue destroy code path (invoked while shutting down controller) and hw/hctx queue dispatcher called from the nvme keep-alive async request queuing operation. This race could lead to the kernel crash shown below: Call Trace: autoremove_wake_function+0x0/0xbc (unreliable) __blk_mq_sched_dispatch_requests+0x114/0x24c blk_mq_sched_dispatch_requests+0x44/0x84 blk_mq_run_hw_queue+0x140/0x220 nvme_keep_alive_work+0xc8/0x19c [nvme_core] process_one_work+0x200/0x4e0 worker_thread+0x340/0x504 kthread+0x138/0x140 start_kernel_thread+0x14/0x18 While shutting down fabric controller, if nvme keep-alive request sneaks in then it would be flushed off. The nvme_keep_alive_end_io function is then invoked to handle the end of the keep-alive operation which decrements the admin->q_usage_counter and assuming this is the last/only request in the admin queue then the admin->q_usage_counter becomes zero. If that happens then blk-mq destroy queue operation (blk_mq_destroy_ queue()) which could be potentially running simultaneously on another cpu (as this is the controller shutdown code path) would forward progress and deletes the admin queue. So, now from this point onward we are not supposed to access the admin queue resources. However the issue here's that the nvme keep-alive thread running hw/hctx queue dispatch operation hasn't yet finished its work and so it could still potentially access the admin queue resource while the admin queue had been already deleted and that causes the above crash. The above kernel crash is regression caused due to changes implemented in commit a54a93d0e359 ("nvme: move stopping keep-alive into nvme_uninit_ctrl()"). Ideally we should stop keep-alive before destroyin g the admin queue and freeing the admin tagset so that it wouldn't sneak in during the shutdown operation. However we removed the keep alive stop operation from the beginning of the controller shutdown code path in commit a54a93d0e359 ("nvme: move stopping keep-alive into nvme_uninit_ctrl()") and added it under nvme_uninit_ctrl() which executes very late in the shutdown code path after the admin queue is destroyed and its tagset is removed. So this change created the possibility of keep-alive sneaking in and interfering with the shutdown operation and causing observed kernel crash. To fix the observed crash, we decided to move nvme_stop_keep_alive() from nvme_uninit_ctrl() to nvme_remove_admin_tag_set(). This change would ensure that we don't forward progress and delete the admin queue until the keep- alive operation is finished (if it's in-flight) or cancelled and that would help contain the race condition explained above and hence avoid the crash. Moving nvme_stop_keep_alive() to nvme_remove_admin_tag_set() instead of adding nvme_stop_keep_alive() to the beginning of the controller shutdown code path in nvme_stop_ctrl(), as was the case earlier before commit a54a93d0e359 ("nvme: move stopping keep-alive into nvme_uninit_ctrl()"), would help save one callsite of nvme_stop_keep_alive(). Fixes: a54a93d0e359 ("nvme: move stopping keep-alive into nvme_uninit_ctrl()") Link: https://lore.kernel.org/all/1a21f37b-0f2a-4745-8c56-4dc8628d3983@linux.ibm.com/ Reviewed-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit e9869c85c81168a1275f909d5972a3fc435304be ]

The nvme keep-alive operation, which executes at a periodic interval,
could potentially sneak in while shutting down a fabric controller.
This may lead to a race between the fabric controller admin queue
destroy code path (invoked while shutting down controller) and hw/hctx
queue dispatcher called from the nvme keep-alive async request queuing
operation. This race could lead to the kernel crash shown below:

Call Trace:
    autoremove_wake_function+0x0/0xbc (unreliable)
    __blk_mq_sched_dispatch_requests+0x114/0x24c
    blk_mq_sched_dispatch_requests+0x44/0x84
    blk_mq_run_hw_queue+0x140/0x220
    nvme_keep_alive_work+0xc8/0x19c [nvme_core]
    process_one_work+0x200/0x4e0
    worker_thread+0x340/0x504
    kthread+0x138/0x140
    start_kernel_thread+0x14/0x18

While shutting down fabric controller, if nvme keep-alive request sneaks
in then it would be flushed off. The nvme_keep_alive_end_io function is
then invoked to handle the end of the keep-alive operation which
decrements the admin->q_usage_counter and assuming this is the last/only
request in the admin queue then the admin->q_usage_counter becomes zero.
If that happens then blk-mq destroy queue operation (blk_mq_destroy_
queue()) which could be potentially running simultaneously on another
cpu (as this is the controller shutdown code path) would forward
progress and deletes the admin queue. So, now from this point onward
we are not supposed to access the admin queue resources. However the
issue here's that the nvme keep-alive thread running hw/hctx queue
dispatch operation hasn't yet finished its work and so it could still
potentially access the admin queue resource while the admin queue had
been already deleted and that causes the above crash.

The above kernel crash is regression caused due to changes implemented
in commit a54a93d0e359 ("nvme: move stopping keep-alive into
nvme_uninit_ctrl()"). Ideally we should stop keep-alive before destroyin
g the admin queue and freeing the admin tagset so that it wouldn't sneak
in during the shutdown operation. However we removed the keep alive stop
operation from the beginning of the controller shutdown code path in commit
a54a93d0e359 ("nvme: move stopping keep-alive into nvme_uninit_ctrl()")
and added it under nvme_uninit_ctrl() which executes very late in the
shutdown code path after the admin queue is destroyed and its tagset is
removed. So this change created the possibility of keep-alive sneaking in
and interfering with the shutdown operation and causing observed kernel
crash.

To fix the observed crash, we decided to move nvme_stop_keep_alive() from
nvme_uninit_ctrl() to nvme_remove_admin_tag_set(). This change would ensure
that we don't forward progress and delete the admin queue until the keep-
alive operation is finished (if it's in-flight) or cancelled and that would
help contain the race condition explained above and hence avoid the crash.

Moving nvme_stop_keep_alive() to nvme_remove_admin_tag_set() instead of
adding nvme_stop_keep_alive() to the beginning of the controller shutdown
code path in nvme_stop_ctrl(), as was the case earlier before commit
a54a93d0e359 ("nvme: move stopping keep-alive into nvme_uninit_ctrl()"),
would help save one callsite of nvme_stop_keep_alive().

Fixes: a54a93d0e359 ("nvme: move stopping keep-alive into nvme_uninit_ctrl()")
Link: https://lore.kernel.org/all/1a21f37b-0f2a-4745-8c56-4dc8628d3983@linux.ibm.com/
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Nilay Shroff <nilay@linux.ibm.com>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
nvme/multipath: Fix RCU list traversal to use SRCU primitive 2024-12-05T12:54:29+00:00 Breno Leitao leitao@debian.org 2024-11-05T14:42:46+00:00 c03cc062e3acccc6693a6f70d611784c39b9b2d4 [ Upstream commit 5dd18f09ce7399df6fffe80d1598add46c395ae9 ] The code currently uses list_for_each_entry_rcu() while holding an SRCU lock, triggering false positive warnings with CONFIG_PROVE_RCU=y enabled: drivers/nvme/host/multipath.c:168 RCU-list traversed in non-reader section!! drivers/nvme/host/multipath.c:227 RCU-list traversed in non-reader section!! drivers/nvme/host/multipath.c:260 RCU-list traversed in non-reader section!! While the list is properly protected by SRCU lock, the code uses the wrong list traversal primitive. Replace list_for_each_entry_rcu() with list_for_each_entry_srcu() to correctly indicate SRCU-based protection and eliminate the false warning. Signed-off-by: Breno Leitao <leitao@debian.org> Fixes: be647e2c76b2 ("nvme: use srcu for iterating namespace list") Signed-off-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 5dd18f09ce7399df6fffe80d1598add46c395ae9 ]

The code currently uses list_for_each_entry_rcu() while holding an SRCU
lock, triggering false positive warnings with CONFIG_PROVE_RCU=y
enabled:

	drivers/nvme/host/multipath.c:168 RCU-list traversed in non-reader section!!
	drivers/nvme/host/multipath.c:227 RCU-list traversed in non-reader section!!
	drivers/nvme/host/multipath.c:260 RCU-list traversed in non-reader section!!

While the list is properly protected by SRCU lock, the code uses the
wrong list traversal primitive. Replace list_for_each_entry_rcu() with
list_for_each_entry_srcu() to correctly indicate SRCU-based protection
and eliminate the false warning.

Signed-off-by: Breno Leitao <leitao@debian.org>
Fixes: be647e2c76b2 ("nvme: use srcu for iterating namespace list")
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
nvme-multipath: avoid hang on inaccessible namespaces 2024-12-05T12:54:29+00:00 Hannes Reinecke hare@kernel.org 2024-09-14T12:01:23+00:00 64184804a74e5cab8bb71bc856b4693d5d06cdb0 [ Upstream commit 3b97f5a05cfc55e7729ff3769f63eef64e2178bb ] During repetitive namespace remapping operations on the target the namespace might have changed between the time the initial scan was performed, and partition scan was invoked by device_add_disk() in nvme_mpath_set_live(). We then end up with a stuck scanning process: [<0>] folio_wait_bit_common+0x12a/0x310 [<0>] filemap_read_folio+0x97/0xd0 [<0>] do_read_cache_folio+0x108/0x390 [<0>] read_part_sector+0x31/0xa0 [<0>] read_lba+0xc5/0x160 [<0>] efi_partition+0xd9/0x8f0 [<0>] bdev_disk_changed+0x23d/0x6d0 [<0>] blkdev_get_whole+0x78/0xc0 [<0>] bdev_open+0x2c6/0x3b0 [<0>] bdev_file_open_by_dev+0xcb/0x120 [<0>] disk_scan_partitions+0x5d/0x100 [<0>] device_add_disk+0x402/0x420 [<0>] nvme_mpath_set_live+0x4f/0x1f0 [nvme_core] [<0>] nvme_mpath_add_disk+0x107/0x120 [nvme_core] [<0>] nvme_alloc_ns+0xac6/0xe60 [nvme_core] [<0>] nvme_scan_ns+0x2dd/0x3e0 [nvme_core] [<0>] nvme_scan_work+0x1a3/0x490 [nvme_core] This happens when we have several paths, some of which are inaccessible, and the active paths are removed first. Then nvme_find_path() will requeue I/O in the ns_head (as paths are present), but the requeue list is never triggered as all remaining paths are inactive. This patch checks for NVME_NSHEAD_DISK_LIVE in nvme_available_path(), and requeue I/O after NVME_NSHEAD_DISK_LIVE has been cleared once the last path has been removed to properly terminate pending I/O. Signed-off-by: Hannes Reinecke <hare@kernel.org> Reviewed-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Keith Busch <kbusch@kernel.org> Stable-dep-of: 5dd18f09ce73 ("nvme/multipath: Fix RCU list traversal to use SRCU primitive") Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 3b97f5a05cfc55e7729ff3769f63eef64e2178bb ]

During repetitive namespace remapping operations on the target the
namespace might have changed between the time the initial scan
was performed, and partition scan was invoked by device_add_disk()
in nvme_mpath_set_live(). We then end up with a stuck scanning process:

[<0>] folio_wait_bit_common+0x12a/0x310
[<0>] filemap_read_folio+0x97/0xd0
[<0>] do_read_cache_folio+0x108/0x390
[<0>] read_part_sector+0x31/0xa0
[<0>] read_lba+0xc5/0x160
[<0>] efi_partition+0xd9/0x8f0
[<0>] bdev_disk_changed+0x23d/0x6d0
[<0>] blkdev_get_whole+0x78/0xc0
[<0>] bdev_open+0x2c6/0x3b0
[<0>] bdev_file_open_by_dev+0xcb/0x120
[<0>] disk_scan_partitions+0x5d/0x100
[<0>] device_add_disk+0x402/0x420
[<0>] nvme_mpath_set_live+0x4f/0x1f0 [nvme_core]
[<0>] nvme_mpath_add_disk+0x107/0x120 [nvme_core]
[<0>] nvme_alloc_ns+0xac6/0xe60 [nvme_core]
[<0>] nvme_scan_ns+0x2dd/0x3e0 [nvme_core]
[<0>] nvme_scan_work+0x1a3/0x490 [nvme_core]

This happens when we have several paths, some of which are inaccessible,
and the active paths are removed first. Then nvme_find_path() will requeue
I/O in the ns_head (as paths are present), but the requeue list is never
triggered as all remaining paths are inactive.

This patch checks for NVME_NSHEAD_DISK_LIVE in nvme_available_path(),
and requeue I/O after NVME_NSHEAD_DISK_LIVE has been cleared once
the last path has been removed to properly terminate pending I/O.

Signed-off-by: Hannes Reinecke <hare@kernel.org>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Stable-dep-of: 5dd18f09ce73 ("nvme/multipath: Fix RCU list traversal to use SRCU primitive")
Signed-off-by: Sasha Levin <sashal@kernel.org>
nvme-pci: reverse request order in nvme_queue_rqs 2024-12-05T12:52:48+00:00 Christoph Hellwig hch@lst.de 2024-11-13T15:20:41+00:00 0595c647d45f6599e7e9cc63df8361e7f6bf59a0 [ Upstream commit beadf0088501d9dcf2454b05d90d5d31ea3ba55f ] blk_mq_flush_plug_list submits requests in the reverse order that they were submitted, which leads to a rather suboptimal I/O pattern especially in rotational devices. Fix this by rewriting nvme_queue_rqs so that it always pops the requests from the passed in request list, and then adds them to the head of a local submit list. This actually simplifies the code a bit as it removes the complicated list splicing, at the cost of extra updates of the rq_next pointer. As that should be cache hot anyway it should be an easy price to pay. Fixes: d62cbcf62f2f ("nvme: add support for mq_ops->queue_rqs()") Signed-off-by: Christoph Hellwig <hch@lst.de> Link: https://lore.kernel.org/r/20241113152050.157179-2-hch@lst.de Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit beadf0088501d9dcf2454b05d90d5d31ea3ba55f ]

blk_mq_flush_plug_list submits requests in the reverse order that they
were submitted, which leads to a rather suboptimal I/O pattern especially
in rotational devices.  Fix this by rewriting nvme_queue_rqs so that it
always pops the requests from the passed in request list, and then adds
them to the head of a local submit list.  This actually simplifies the
code a bit as it removes the complicated list splicing, at the cost of
extra updates of the rq_next pointer.  As that should be cache hot
anyway it should be an easy price to pay.

Fixes: d62cbcf62f2f ("nvme: add support for mq_ops->queue_rqs()")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20241113152050.157179-2-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
nvme-pci: fix freeing of the HMB descriptor table 2024-12-05T12:52:45+00:00 Christoph Hellwig hch@lst.de 2024-11-01T04:40:04+00:00 6d0f599db73b099aa724a12736369c4d4d92849d [ Upstream commit 3c2fb1ca8086eb139b2a551358137525ae8e0d7a ] The HMB descriptor table is sized to the maximum number of descriptors that could be used for a given device, but __nvme_alloc_host_mem could break out of the loop earlier on memory allocation failure and end up using less descriptors than planned for, which leads to an incorrect size passed to dma_free_coherent. In practice this was not showing up because the number of descriptors tends to be low and the dma coherent allocator always allocates and frees at least a page. Fixes: 87ad72a59a38 ("nvme-pci: implement host memory buffer support") Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 3c2fb1ca8086eb139b2a551358137525ae8e0d7a ]

The HMB descriptor table is sized to the maximum number of descriptors
that could be used for a given device, but __nvme_alloc_host_mem could
break out of the loop earlier on memory allocation failure and end up
using less descriptors than planned for, which leads to an incorrect
size passed to dma_free_coherent.

In practice this was not showing up because the number of descriptors
tends to be low and the dma coherent allocator always allocates and
frees at least a page.

Fixes: 87ad72a59a38 ("nvme-pci: implement host memory buffer support")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
nvme: make keep-alive synchronous operation 2024-11-17T14:09:52+00:00 Nilay Shroff nilay@linux.ibm.com 2024-10-16T03:03:15+00:00 ccc1d82dfaad0ad27d21139da22e57add73d2a5e [ Upstream commit d06923670b5a5f609603d4a9fee4dec02d38de9c ] The nvme keep-alive operation, which executes at a periodic interval, could potentially sneak in while shutting down a fabric controller. This may lead to a race between the fabric controller admin queue destroy code path (invoked while shutting down controller) and hw/hctx queue dispatcher called from the nvme keep-alive async request queuing operation. This race could lead to the kernel crash shown below: Call Trace: autoremove_wake_function+0x0/0xbc (unreliable) __blk_mq_sched_dispatch_requests+0x114/0x24c blk_mq_sched_dispatch_requests+0x44/0x84 blk_mq_run_hw_queue+0x140/0x220 nvme_keep_alive_work+0xc8/0x19c [nvme_core] process_one_work+0x200/0x4e0 worker_thread+0x340/0x504 kthread+0x138/0x140 start_kernel_thread+0x14/0x18 While shutting down fabric controller, if nvme keep-alive request sneaks in then it would be flushed off. The nvme_keep_alive_end_io function is then invoked to handle the end of the keep-alive operation which decrements the admin->q_usage_counter and assuming this is the last/only request in the admin queue then the admin->q_usage_counter becomes zero. If that happens then blk-mq destroy queue operation (blk_mq_destroy_ queue()) which could be potentially running simultaneously on another cpu (as this is the controller shutdown code path) would forward progress and deletes the admin queue. So, now from this point onward we are not supposed to access the admin queue resources. However the issue here's that the nvme keep-alive thread running hw/hctx queue dispatch operation hasn't yet finished its work and so it could still potentially access the admin queue resource while the admin queue had been already deleted and that causes the above crash. This fix helps avoid the observed crash by implementing keep-alive as a synchronous operation so that we decrement admin->q_usage_counter only after keep-alive command finished its execution and returns the command status back up to its caller (blk_execute_rq()). This would ensure that fabric shutdown code path doesn't destroy the fabric admin queue until keep-alive request finished execution and also keep-alive thread is not running hw/hctx queue dispatch operation. Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit d06923670b5a5f609603d4a9fee4dec02d38de9c ]

The nvme keep-alive operation, which executes at a periodic interval,
could potentially sneak in while shutting down a fabric controller.
This may lead to a race between the fabric controller admin queue
destroy code path (invoked while shutting down controller) and hw/hctx
queue dispatcher called from the nvme keep-alive async request queuing
operation. This race could lead to the kernel crash shown below:

Call Trace:
    autoremove_wake_function+0x0/0xbc (unreliable)
    __blk_mq_sched_dispatch_requests+0x114/0x24c
    blk_mq_sched_dispatch_requests+0x44/0x84
    blk_mq_run_hw_queue+0x140/0x220
    nvme_keep_alive_work+0xc8/0x19c [nvme_core]
    process_one_work+0x200/0x4e0
    worker_thread+0x340/0x504
    kthread+0x138/0x140
    start_kernel_thread+0x14/0x18

While shutting down fabric controller, if nvme keep-alive request sneaks
in then it would be flushed off. The nvme_keep_alive_end_io function is
then invoked to handle the end of the keep-alive operation which
decrements the admin->q_usage_counter and assuming this is the last/only
request in the admin queue then the admin->q_usage_counter becomes zero.
If that happens then blk-mq destroy queue operation (blk_mq_destroy_
queue()) which could be potentially running simultaneously on another
cpu (as this is the controller shutdown code path) would forward
progress and deletes the admin queue. So, now from this point onward
we are not supposed to access the admin queue resources. However the
issue here's that the nvme keep-alive thread running hw/hctx queue
dispatch operation hasn't yet finished its work and so it could still
potentially access the admin queue resource while the admin queue had
been already deleted and that causes the above crash.

This fix helps avoid the observed crash by implementing keep-alive as a
synchronous operation so that we decrement admin->q_usage_counter only
after keep-alive command finished its execution and returns the command
status back up to its caller (blk_execute_rq()). This would ensure that
fabric shutdown code path doesn't destroy the fabric admin queue until
keep-alive request finished execution and also keep-alive thread is not
running hw/hctx queue dispatch operation.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nilay Shroff <nilay@linux.ibm.com>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
nvme-loop: flush off pending I/O while shutting down loop controller 2024-11-17T14:09:52+00:00 Nilay Shroff nilay@linux.ibm.com 2024-10-16T03:03:14+00:00 2934b11b504c11e2ee88084b4ae4d8c8edd552e1 [ Upstream commit c199fac88fe7c749f88a0653e9f621b9f5a71cf1 ] While shutting down loop controller, we first quiesce the admin/IO queue, delete the admin/IO tag-set and then at last destroy the admin/IO queue. However it's quite possible that during the window between quiescing and destroying of the admin/IO queue, some admin/IO request might sneak in and if that happens then we could potentially encounter a hung task because shutdown operation can't forward progress until any pending I/O is flushed off. This commit helps ensure that before destroying the admin/IO queue, we unquiesce the admin/IO queue so that any outstanding requests, which are added after the admin/IO queue is quiesced, are now flushed to its completion. Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit c199fac88fe7c749f88a0653e9f621b9f5a71cf1 ]

While shutting down loop controller, we first quiesce the admin/IO queue,
delete the admin/IO tag-set and then at last destroy the admin/IO queue.
However it's quite possible that during the window between quiescing and
destroying of the admin/IO queue, some admin/IO request might sneak in
and if that happens then we could potentially encounter a hung task
because shutdown operation can't forward progress until any pending I/O
is flushed off.

This commit helps ensure that before destroying the admin/IO queue, we
unquiesce the admin/IO queue so that any outstanding requests, which are
added after the admin/IO queue is quiesced, are now flushed to its
completion.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nilay Shroff <nilay@linux.ibm.com>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
nvme-multipath: defer partition scanning 2024-11-17T14:09:52+00:00 Keith Busch kbusch@kernel.org 2024-10-15T14:30:17+00:00 a91b7eddf45afeeb9c5ece11dddff5de0921b00f [ Upstream commit 1f021341eef41e77a633186e9be5223de2ce5d48 ] We need to suppress the partition scan from occuring within the controller's scan_work context. If a path error occurs here, the IO will wait until a path becomes available or all paths are torn down, but that action also occurs within scan_work, so it would deadlock. Defer the partion scan to a different context that does not block scan_work. Reported-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 1f021341eef41e77a633186e9be5223de2ce5d48 ]

We need to suppress the partition scan from occuring within the
controller's scan_work context. If a path error occurs here, the IO will
wait until a path becomes available or all paths are torn down, but that
action also occurs within scan_work, so it would deadlock. Defer the
partion scan to a different context that does not block scan_work.

Reported-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
nvme: disable CC.CRIME (NVME_CC_CRIME) 2024-11-17T14:09:51+00:00 Greg Joyce gjoyce@linux.ibm.com 2024-10-07T19:33:24+00:00 f3604ead6c846ac4ac0bddb26b5a6d6257c42968 [ Upstream commit 0ce96a6708f34280a536263ee5c67e20c433dcce ] Disable NVME_CC_CRIME so that CSTS.RDY indicates that the media is ready and able to handle commands without returning NVME_SC_ADMIN_COMMAND_MEDIA_NOT_READY. Signed-off-by: Greg Joyce <gjoyce@linux.ibm.com> Reviewed-by: Nilay Shroff <nilay@linux.ibm.com> Tested-by: Nilay Shroff <nilay@linux.ibm.com> Signed-off-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 0ce96a6708f34280a536263ee5c67e20c433dcce ]

Disable NVME_CC_CRIME so that CSTS.RDY indicates that the media
is ready and able to handle commands without returning
NVME_SC_ADMIN_COMMAND_MEDIA_NOT_READY.

Signed-off-by: Greg Joyce <gjoyce@linux.ibm.com>
Reviewed-by: Nilay Shroff <nilay@linux.ibm.com>
Tested-by: Nilay Shroff <nilay@linux.ibm.com>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
nvme: tcp: avoid race between queue_lock lock and destroy 2024-11-17T14:09:51+00:00 Hannes Reinecke hare@suse.de 2024-10-02T04:51:41+00:00 e15cebc1b21856944b387f4abd03b66bd3d4f027 [ Upstream commit 782373ba27660ba7d330208cf5509ece6feb4545 ] Commit 76d54bf20cdc ("nvme-tcp: don't access released socket during error recovery") added a mutex_lock() call for the queue->queue_lock in nvme_tcp_get_address(). However, the mutex_lock() races with mutex_destroy() in nvme_tcp_free_queue(), and causes the WARN below. DEBUG_LOCKS_WARN_ON(lock->magic != lock) WARNING: CPU: 3 PID: 34077 at kernel/locking/mutex.c:587 __mutex_lock+0xcf0/0x1220 Modules linked in: nvmet_tcp nvmet nvme_tcp nvme_fabrics iw_cm ib_cm ib_core pktcdvd nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ip_set nf_tables qrtr sunrpc ppdev 9pnet_virtio 9pnet pcspkr netfs parport_pc parport e1000 i2c_piix4 i2c_smbus loop fuse nfnetlink zram bochs drm_vram_helper drm_ttm_helper ttm drm_kms_helper xfs drm sym53c8xx floppy nvme scsi_transport_spi nvme_core nvme_auth serio_raw ata_generic pata_acpi dm_multipath qemu_fw_cfg [last unloaded: ib_uverbs] CPU: 3 UID: 0 PID: 34077 Comm: udisksd Not tainted 6.11.0-rc7 #319 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-2.fc40 04/01/2014 RIP: 0010:__mutex_lock+0xcf0/0x1220 Code: 08 84 d2 0f 85 c8 04 00 00 8b 15 ef b6 c8 01 85 d2 0f 85 78 f4 ff ff 48 c7 c6 20 93 ee af 48 c7 c7 60 91 ee af e8 f0 a7 6d fd <0f> 0b e9 5e f4 ff ff 48 b8 00 00 00 00 00 fc ff df 4c 89 f2 48 c1 RSP: 0018:ffff88811305f760 EFLAGS: 00010286 RAX: 0000000000000000 RBX: ffff88812c652058 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000004 RDI: 0000000000000001 RBP: ffff88811305f8b0 R08: 0000000000000001 R09: ffffed1075c36341 R10: ffff8883ae1b1a0b R11: 0000000000010498 R12: 0000000000000000 R13: 0000000000000000 R14: dffffc0000000000 R15: ffff88812c652058 FS: 00007f9713ae4980(0000) GS:ffff8883ae180000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fcd78483c7c CR3: 0000000122c38000 CR4: 00000000000006f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> ? __warn.cold+0x5b/0x1af ? __mutex_lock+0xcf0/0x1220 ? report_bug+0x1ec/0x390 ? handle_bug+0x3c/0x80 ? exc_invalid_op+0x13/0x40 ? asm_exc_invalid_op+0x16/0x20 ? __mutex_lock+0xcf0/0x1220 ? nvme_tcp_get_address+0xc2/0x1e0 [nvme_tcp] ? __pfx___mutex_lock+0x10/0x10 ? __lock_acquire+0xd6a/0x59e0 ? nvme_tcp_get_address+0xc2/0x1e0 [nvme_tcp] nvme_tcp_get_address+0xc2/0x1e0 [nvme_tcp] ? __pfx_nvme_tcp_get_address+0x10/0x10 [nvme_tcp] nvme_sysfs_show_address+0x81/0xc0 [nvme_core] dev_attr_show+0x42/0x80 ? __asan_memset+0x1f/0x40 sysfs_kf_seq_show+0x1f0/0x370 seq_read_iter+0x2cb/0x1130 ? rw_verify_area+0x3b1/0x590 ? __mutex_lock+0x433/0x1220 vfs_read+0x6a6/0xa20 ? lockdep_hardirqs_on+0x78/0x100 ? __pfx_vfs_read+0x10/0x10 ksys_read+0xf7/0x1d0 ? __pfx_ksys_read+0x10/0x10 ? __x64_sys_openat+0x105/0x1d0 do_syscall_64+0x93/0x180 ? lockdep_hardirqs_on_prepare+0x16d/0x400 ? do_syscall_64+0x9f/0x180 ? lockdep_hardirqs_on+0x78/0x100 ? do_syscall_64+0x9f/0x180 ? __pfx_ksys_read+0x10/0x10 ? lockdep_hardirqs_on_prepare+0x16d/0x400 ? do_syscall_64+0x9f/0x180 ? lockdep_hardirqs_on+0x78/0x100 ? do_syscall_64+0x9f/0x180 ? lockdep_hardirqs_on_prepare+0x16d/0x400 ? do_syscall_64+0x9f/0x180 ? lockdep_hardirqs_on+0x78/0x100 ? do_syscall_64+0x9f/0x180 ? lockdep_hardirqs_on_prepare+0x16d/0x400 ? do_syscall_64+0x9f/0x180 ? lockdep_hardirqs_on+0x78/0x100 ? do_syscall_64+0x9f/0x180 ? lockdep_hardirqs_on_prepare+0x16d/0x400 ? do_syscall_64+0x9f/0x180 ? lockdep_hardirqs_on+0x78/0x100 ? do_syscall_64+0x9f/0x180 ? do_syscall_64+0x9f/0x180 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f9713f55cfa Code: 55 48 89 e5 48 83 ec 20 48 89 55 e8 48 89 75 f0 89 7d f8 e8 e8 74 f8 ff 48 8b 55 e8 48 8b 75 f0 41 89 c0 8b 7d f8 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 2e 44 89 c7 48 89 45 f8 e8 42 75 f8 ff 48 8b RSP: 002b:00007ffd7f512e70 EFLAGS: 00000246 ORIG_RAX: 0000000000000000 RAX: ffffffffffffffda RBX: 000055c38f316859 RCX: 00007f9713f55cfa RDX: 0000000000000fff RSI: 00007ffd7f512eb0 RDI: 0000000000000011 RBP: 00007ffd7f512e90 R08: 0000000000000000 R09: 00000000ffffffff R10: 0000000000000000 R11: 0000000000000246 R12: 000055c38f317148 R13: 0000000000000000 R14: 00007f96f4004f30 R15: 000055c3b6b623c0 </TASK> The WARN is observed when the blktests test case nvme/014 is repeated with tcp transport. It is rare, and 200 times repeat is required to recreate in some test environments. To avoid the WARN, check the NVME_TCP_Q_LIVE flag before locking queue->queue_lock. The flag is cleared long time before the lock gets destroyed. Signed-off-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com> Signed-off-by: Keith Busch <kbusch@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 782373ba27660ba7d330208cf5509ece6feb4545 ]

Commit 76d54bf20cdc ("nvme-tcp: don't access released socket during
error recovery") added a mutex_lock() call for the queue->queue_lock
in nvme_tcp_get_address(). However, the mutex_lock() races with
mutex_destroy() in nvme_tcp_free_queue(), and causes the WARN below.

DEBUG_LOCKS_WARN_ON(lock->magic != lock)
WARNING: CPU: 3 PID: 34077 at kernel/locking/mutex.c:587 __mutex_lock+0xcf0/0x1220
Modules linked in: nvmet_tcp nvmet nvme_tcp nvme_fabrics iw_cm ib_cm ib_core pktcdvd nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ip_set nf_tables qrtr sunrpc ppdev 9pnet_virtio 9pnet pcspkr netfs parport_pc parport e1000 i2c_piix4 i2c_smbus loop fuse nfnetlink zram bochs drm_vram_helper drm_ttm_helper ttm drm_kms_helper xfs drm sym53c8xx floppy nvme scsi_transport_spi nvme_core nvme_auth serio_raw ata_generic pata_acpi dm_multipath qemu_fw_cfg [last unloaded: ib_uverbs]
CPU: 3 UID: 0 PID: 34077 Comm: udisksd Not tainted 6.11.0-rc7 #319
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-2.fc40 04/01/2014
RIP: 0010:__mutex_lock+0xcf0/0x1220
Code: 08 84 d2 0f 85 c8 04 00 00 8b 15 ef b6 c8 01 85 d2 0f 85 78 f4 ff ff 48 c7 c6 20 93 ee af 48 c7 c7 60 91 ee af e8 f0 a7 6d fd <0f> 0b e9 5e f4 ff ff 48 b8 00 00 00 00 00 fc ff df 4c 89 f2 48 c1
RSP: 0018:ffff88811305f760 EFLAGS: 00010286
RAX: 0000000000000000 RBX: ffff88812c652058 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000004 RDI: 0000000000000001
RBP: ffff88811305f8b0 R08: 0000000000000001 R09: ffffed1075c36341
R10: ffff8883ae1b1a0b R11: 0000000000010498 R12: 0000000000000000
R13: 0000000000000000 R14: dffffc0000000000 R15: ffff88812c652058
FS:  00007f9713ae4980(0000) GS:ffff8883ae180000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fcd78483c7c CR3: 0000000122c38000 CR4: 00000000000006f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 ? __warn.cold+0x5b/0x1af
 ? __mutex_lock+0xcf0/0x1220
 ? report_bug+0x1ec/0x390
 ? handle_bug+0x3c/0x80
 ? exc_invalid_op+0x13/0x40
 ? asm_exc_invalid_op+0x16/0x20
 ? __mutex_lock+0xcf0/0x1220
 ? nvme_tcp_get_address+0xc2/0x1e0 [nvme_tcp]
 ? __pfx___mutex_lock+0x10/0x10
 ? __lock_acquire+0xd6a/0x59e0
 ? nvme_tcp_get_address+0xc2/0x1e0 [nvme_tcp]
 nvme_tcp_get_address+0xc2/0x1e0 [nvme_tcp]
 ? __pfx_nvme_tcp_get_address+0x10/0x10 [nvme_tcp]
 nvme_sysfs_show_address+0x81/0xc0 [nvme_core]
 dev_attr_show+0x42/0x80
 ? __asan_memset+0x1f/0x40
 sysfs_kf_seq_show+0x1f0/0x370
 seq_read_iter+0x2cb/0x1130
 ? rw_verify_area+0x3b1/0x590
 ? __mutex_lock+0x433/0x1220
 vfs_read+0x6a6/0xa20
 ? lockdep_hardirqs_on+0x78/0x100
 ? __pfx_vfs_read+0x10/0x10
 ksys_read+0xf7/0x1d0
 ? __pfx_ksys_read+0x10/0x10
 ? __x64_sys_openat+0x105/0x1d0
 do_syscall_64+0x93/0x180
 ? lockdep_hardirqs_on_prepare+0x16d/0x400
 ? do_syscall_64+0x9f/0x180
 ? lockdep_hardirqs_on+0x78/0x100
 ? do_syscall_64+0x9f/0x180
 ? __pfx_ksys_read+0x10/0x10
 ? lockdep_hardirqs_on_prepare+0x16d/0x400
 ? do_syscall_64+0x9f/0x180
 ? lockdep_hardirqs_on+0x78/0x100
 ? do_syscall_64+0x9f/0x180
 ? lockdep_hardirqs_on_prepare+0x16d/0x400
 ? do_syscall_64+0x9f/0x180
 ? lockdep_hardirqs_on+0x78/0x100
 ? do_syscall_64+0x9f/0x180
 ? lockdep_hardirqs_on_prepare+0x16d/0x400
 ? do_syscall_64+0x9f/0x180
 ? lockdep_hardirqs_on+0x78/0x100
 ? do_syscall_64+0x9f/0x180
 ? lockdep_hardirqs_on_prepare+0x16d/0x400
 ? do_syscall_64+0x9f/0x180
 ? lockdep_hardirqs_on+0x78/0x100
 ? do_syscall_64+0x9f/0x180
 ? do_syscall_64+0x9f/0x180
 entry_SYSCALL_64_after_hwframe+0x76/0x7e
RIP: 0033:0x7f9713f55cfa
Code: 55 48 89 e5 48 83 ec 20 48 89 55 e8 48 89 75 f0 89 7d f8 e8 e8 74 f8 ff 48 8b 55 e8 48 8b 75 f0 41 89 c0 8b 7d f8 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 2e 44 89 c7 48 89 45 f8 e8 42 75 f8 ff 48 8b
RSP: 002b:00007ffd7f512e70 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
RAX: ffffffffffffffda RBX: 000055c38f316859 RCX: 00007f9713f55cfa
RDX: 0000000000000fff RSI: 00007ffd7f512eb0 RDI: 0000000000000011
RBP: 00007ffd7f512e90 R08: 0000000000000000 R09: 00000000ffffffff
R10: 0000000000000000 R11: 0000000000000246 R12: 000055c38f317148
R13: 0000000000000000 R14: 00007f96f4004f30 R15: 000055c3b6b623c0
 </TASK>

The WARN is observed when the blktests test case nvme/014 is repeated
with tcp transport. It is rare, and 200 times repeat is required to
recreate in some test environments.

To avoid the WARN, check the NVME_TCP_Q_LIVE flag before locking
queue->queue_lock. The flag is cleared long time before the lock gets
destroyed.

Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>