* blktests/nvme/008
@ 2018-07-25 15:56 Bart Van Assche
2018-07-25 16:05 ` blktests/nvme/008 Chaitanya Kulkarni
0 siblings, 1 reply; 4+ messages in thread
From: Bart Van Assche @ 2018-07-25 15:56 UTC (permalink / raw)
Hello,
If I run blktests test nvme/008 against kernel v4.18-rc6 then the following
appears:
run blktests nvme/008 at 2018-07-25 08:38:04
nvmet: adding nsid 1 to subsystem blktests-subsystem-1
nvmet: creating controller 1 for subsystem blktests-subsystem-1 for NQN nqn.2014-08.org.nvmexpress:uuid:5a9f0ba8-37f1-48e9-b3f0-7715fc55c789.
nvme nvme0: creating 32 I/O queues.
nvme nvme0: new ctrl: "blktests-subsystem-1"
nvme nvme0: Removing ctrl: NQN "blktests-subsystem-1"
nvmet: ctrl 1 keep-alive timer (15 seconds) expired!
nvmet: ctrl 1 fatal error occurred!
INFO: task kworker/u64:4:375 blocked for more than 120 seconds.
Not tainted 4.18.0-rc6-dbg+ #15
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
kworker/u64:4 D24720 375 2 0x80000000
Workqueue: nvme-delete-wq nvme_delete_ctrl_work [nvme_core]
Call Trace:
__schedule+0x444/0xff0
schedule+0x4e/0xe0
blk_mq_freeze_queue_wait+0xca/0x180
blk_freeze_queue+0x30/0x50
blk_cleanup_queue+0x153/0x340
nvme_ns_remove+0x20c/0x260 [nvme_core]
nvme_remove_namespaces+0x18b/0x1e0 [nvme_core]
nvme_delete_ctrl_work+0x7d/0xe0 [nvme_core]
process_one_work+0x4fe/0xad0
worker_thread+0x3a9/0x5a0
kthread+0x1c1/0x1e0
ret_from_fork+0x24/0x30
Is this a bug in the test script or in the kernel driver?
Thanks,
Bart.
^ permalink raw reply [flat|nested] 4+ messages in thread
* blktests/nvme/008
2018-07-25 15:56 blktests/nvme/008 Bart Van Assche
@ 2018-07-25 16:05 ` Chaitanya Kulkarni
2018-07-25 16:56 ` blktests/nvme/008 Chaitanya Kulkarni
0 siblings, 1 reply; 4+ messages in thread
From: Chaitanya Kulkarni @ 2018-07-25 16:05 UTC (permalink / raw)
I just ran the same test to check if the test is broken on the
different kernel nvme-4.19, and it did pass without any problem
multiple times.
[root at qemu blktests]# for i in `seq 10` ; do ./check tests/nvme/008; done
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
runtime 0.541s ... 0.523s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
runtime 0.523s ... 0.520s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
runtime 0.520s ... 0.513s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
runtime 0.513s ... 0.504s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
runtime 0.504s ... 0.502s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
runtime 0.502s ... 0.517s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
runtime 0.517s ... 0.505s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
runtime 0.505s ... 0.521s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
runtime 0.521s ... 0.516s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
runtime 0.516s ... 0.510s
[root at qemu blktests]#
I'll build the v4.18-rc6 and run it on that tree.
From: Bart Van Assche
Sent: Wednesday, July 25, 2018 8:56 AM
To: linux-nvme at lists.infradead.org
Cc: Chaitanya Kulkarni
Subject: blktests/nvme/008
?
Hello,
If I run blktests test nvme/008 against kernel v4.18-rc6 then the following
appears:
run blktests nvme/008 at 2018-07-25 08:38:04
nvmet: adding nsid 1 to subsystem blktests-subsystem-1
nvmet: creating controller 1 for subsystem blktests-subsystem-1 for NQN nqn.2014-08.org.nvmexpress:uuid:5a9f0ba8-37f1-48e9-b3f0-7715fc55c789.
nvme nvme0: creating 32 I/O queues.
nvme nvme0: new ctrl: "blktests-subsystem-1"
nvme nvme0: Removing ctrl: NQN "blktests-subsystem-1"
nvmet: ctrl 1 keep-alive timer (15 seconds) expired!
nvmet: ctrl 1 fatal error occurred!
INFO: task kworker/u64:4:375 blocked for more than 120 seconds.
????? Not tainted 4.18.0-rc6-dbg+ #15
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
kworker/u64:4?? D24720?? 375????? 2 0x80000000
Workqueue: nvme-delete-wq nvme_delete_ctrl_work [nvme_core]
Call Trace:
?__schedule+0x444/0xff0
?schedule+0x4e/0xe0
?blk_mq_freeze_queue_wait+0xca/0x180
?blk_freeze_queue+0x30/0x50
?blk_cleanup_queue+0x153/0x340
?nvme_ns_remove+0x20c/0x260 [nvme_core]
?nvme_remove_namespaces+0x18b/0x1e0 [nvme_core]
?nvme_delete_ctrl_work+0x7d/0xe0 [nvme_core]
?process_one_work+0x4fe/0xad0
?worker_thread+0x3a9/0x5a0
?kthread+0x1c1/0x1e0
?ret_from_fork+0x24/0x30
Is this a bug in the test script or in the kernel driver?
Thanks,
Bart.
^ permalink raw reply [flat|nested] 4+ messages in thread
* blktests/nvme/008
2018-07-25 16:05 ` blktests/nvme/008 Chaitanya Kulkarni
@ 2018-07-25 16:56 ` Chaitanya Kulkarni
2018-07-25 16:59 ` blktests/nvme/008 Bart Van Assche
0 siblings, 1 reply; 4+ messages in thread
From: Chaitanya Kulkarni @ 2018-07-25 16:56 UTC (permalink / raw)
Hi Bart,
I ran the test on:-
commit d72e90f33aa4709ebecc5005562f52335e106a60 (HEAD, tag: v4.18-rc6)
Author: Linus Torvalds <torvalds at linux-foundation.org>
Date:? ?Sun Jul 22 14:12:20 2018 -0700
? ? Linux 4.18-rc6
Following are the results:-
[root at qemu blktests]# for i in `seq 10` ; do ./check tests/nvme/008; done?
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
? ? runtime? 0.740s? ...? 0.660s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
? ? runtime? 0.660s? ...? 0.664s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
? ? runtime? 0.664s? ...? 0.667s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
? ? runtime? 0.667s? ...? 0.649s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
? ? runtime? 0.649s? ...? 0.664s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
? ? runtime? 0.664s? ...? 0.655s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
? ? runtime? 0.655s? ...? 0.680s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
? ? runtime? 0.680s? ...? 0.757s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
? ? runtime? 0.757s? ...? 0.658s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
? ? runtime? 0.658s? ...? 0.648s
[root at qemu blktests]# uname -r
4.18.0-rc6LINUX
[root at qemu blktests]#
I'll try and reproduce the issue and look into the test for a bug.
Regards,
Chaitanya
From: Chaitanya Kulkarni
Sent: Wednesday, July 25, 2018 9:05 AM
To: Bart Van Assche; linux-nvme at lists.infradead.org
Subject: Re: blktests/nvme/008
?
I just ran the same test to check if the test is broken on the
different kernel nvme-4.19, and it did pass without any problem
multiple times.
[root at qemu blktests]# for i in `seq 10` ; do ./check tests/nvme/008; done
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
??? runtime? 0.541s? ...? 0.523s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
??? runtime? 0.523s? ...? 0.520s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
??? runtime? 0.520s? ...? 0.513s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
??? runtime? 0.513s? ...? 0.504s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
??? runtime? 0.504s? ...? 0.502s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
??? runtime? 0.502s? ...? 0.517s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
??? runtime? 0.517s? ...? 0.505s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
??? runtime? 0.505s? ...? 0.521s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
??? runtime? 0.521s? ...? 0.516s
nvme/008 (create an NVMeOF host with a block device-backed ns) [passed]
??? runtime? 0.516s? ...? 0.510s
[root at qemu blktests]#
I'll build the v4.18-rc6 and run it on that tree.
From: Bart Van Assche
Sent: Wednesday, July 25, 2018 8:56 AM
To: linux-nvme at lists.infradead.org
Cc: Chaitanya Kulkarni
Subject: blktests/nvme/008
?
?
Hello,
If I run blktests test nvme/008 against kernel v4.18-rc6 then the following
appears:
run blktests nvme/008 at 2018-07-25 08:38:04
nvmet: adding nsid 1 to subsystem blktests-subsystem-1
nvmet: creating controller 1 for subsystem blktests-subsystem-1 for NQN nqn.2014-08.org.nvmexpress:uuid:5a9f0ba8-37f1-48e9-b3f0-7715fc55c789.
nvme nvme0: creating 32 I/O queues.
nvme nvme0: new ctrl: "blktests-subsystem-1"
nvme nvme0: Removing ctrl: NQN "blktests-subsystem-1"
nvmet: ctrl 1 keep-alive timer (15 seconds) expired!
nvmet: ctrl 1 fatal error occurred!
INFO: task kworker/u64:4:375 blocked for more than 120 seconds.
????? Not tainted 4.18.0-rc6-dbg+ #15
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
kworker/u64:4?? D24720?? 375????? 2 0x80000000
Workqueue: nvme-delete-wq nvme_delete_ctrl_work [nvme_core]
Call Trace:
?__schedule+0x444/0xff0
?schedule+0x4e/0xe0
?blk_mq_freeze_queue_wait+0xca/0x180
?blk_freeze_queue+0x30/0x50
?blk_cleanup_queue+0x153/0x340
?nvme_ns_remove+0x20c/0x260 [nvme_core]
?nvme_remove_namespaces+0x18b/0x1e0 [nvme_core]
?nvme_delete_ctrl_work+0x7d/0xe0 [nvme_core]
?process_one_work+0x4fe/0xad0
?worker_thread+0x3a9/0x5a0
?kthread+0x1c1/0x1e0
?ret_from_fork+0x24/0x30
Is this a bug in the test script or in the kernel driver?
Thanks,
Bart.
???
^ permalink raw reply [flat|nested] 4+ messages in thread
* blktests/nvme/008
2018-07-25 16:56 ` blktests/nvme/008 Chaitanya Kulkarni
@ 2018-07-25 16:59 ` Bart Van Assche
0 siblings, 0 replies; 4+ messages in thread
From: Bart Van Assche @ 2018-07-25 16:59 UTC (permalink / raw)
On Wed, 2018-07-25@16:56 +0000, Chaitanya Kulkarni wrote:
> I'll try and reproduce the issue and look into the test for a bug.
Hello Chaitanya,
I do not expect you to spend more time on this. I will check first whether
this is due to one of the v4.19 for-rc branches that I had merged in my tree.
Bart.
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2018-07-25 16:59 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2018-07-25 15:56 blktests/nvme/008 Bart Van Assche
2018-07-25 16:05 ` blktests/nvme/008 Chaitanya Kulkarni
2018-07-25 16:56 ` blktests/nvme/008 Chaitanya Kulkarni
2018-07-25 16:59 ` blktests/nvme/008 Bart Van Assche
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).