From: Yanjun Zhu <yanjun.zhu@linux.dev>
To: Bart Van Assche <bvanassche@acm.org>,
Bob Pearson <rpearsonhpe@gmail.com>,
Zhu Yanjun <zyjzyj2000@gmail.com>,
"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
Bernard Metzler <bmt@zurich.ibm.com>,
Jason Gunthorpe <jgg@nvidia.com>
Subject: Re: Apparent regression in blktests since 5.18-rc1+
Date: Sat, 7 May 2022 08:29:31 +0800 [thread overview]
Message-ID: <4b0153c7-a8e9-98de-26ae-d421434a116d@linux.dev> (raw)
In-Reply-To: <cf8b9980-3965-a4f6-07e0-d4b25755b0db@acm.org>
在 2022/5/7 8:10, Bart Van Assche 写道:
> On 5/6/22 11:11, Bob Pearson wrote:
>> Before the most recent kernel update I had blktests running OK on
>> rdma_rxe. Since we went on to 5.18.0-rc1+
>> I have been experiencing hangs. All of this is with the 'revert
>> scsi-debug' patch which addressed the
>> 3 min timeout related to modprobe -r scsi-debug.
>>
>> You suggested checking with siw and I finally got around to this and
>> the behavior is exactly the same.
>>
>> Specifically here is a run and dmesgs from that run:
>>
>> root@u-22:/home/bob/src/blktests# use_siw=1 ./check srp
>>
>> srp/001 (Create and remove LUNs) [passed]
>>
>> runtime 3.388s ... 3.501s
>>
>> srp/002 (File I/O on top of multipath concurrently with logout and
>> login (mq))
>>
>> runtime 54.689s ...
>> <HANGS HERE>
>>
>> I had to reboot to recover.
>>
>> The dmesg output is attached in a long file called out.
>> The output looks normal until line 1875 where it hangs at an "Already
>> connected ..." message.
>> This is the same as the other hangs I have been seeing.
>> This is followed by a splat warning that a cpu has hung for 120 seconds.
>>
>> Since this is behaving the same for rxe and siw I am going to stop
>> chasing this bug since
>> it is most likely outside of the the rxe driver.
>
> Hi Bob,
>
> What I see on my test setup is that the SRP tests from the blktests
> suite pass with
> the SoftiWARP driver (kernel v5.18-rc5 / commit 4b97bac0756a):
>
> # (cd blktests && use_siw=1 ./check -q srp)
> srp/001 (Create and remove LUNs) [passed]
> runtime 5.781s ... 5.464s
> srp/002 (File I/O on top of multipath concurrently with logout and login
> (mq)) [passed]time 40.772s ...
> runtime 40.772s ... 42.039s
> srp/003 (File I/O on top of multipath concurrently with logout and login
> (sq)) [not run]
> legacy device mapper support is missing
> srp/004 (File I/O on top of multipath concurrently with logout and login
> (sq-on-srp/004 (File I/O on top of multipath concurrently with logout
> and login (sq-on-mq)) [not run]
> legacy device mapper support is missing
> srp/005 (Direct I/O with large transfer sizes, cmd_sg_entries=255 and
> bs=4M) [passed]untime 17.870s ...
> runtime 17.870s ... 17.016s
> srp/006 (Direct I/O with large transfer sizes, cmd_sg_entries=255 and
> bs=8M) [passed]untime 16.369s ...
> runtime 16.369s ... 17.315s
> srp/007 (Direct I/O with large transfer sizes, cmd_sg_entries=1 and
> bs=4M) [passed] runtime 16.729s ...
> runtime 16.729s ... 17.409s
> srp/008 (Direct I/O with large transfer sizes, cmd_sg_entries=1 and
> bs=8M) [passed] runtime 16.823s ...
> runtime 16.823s ... 16.453s
> srp/009 (Buffered I/O with large transfer sizes, cmd_sg_entries=255 and
> bs=4M) [passed]time 17.304s ...
> runtime 17.304s ... 17.838s
> srp/010 (Buffered I/O with large transfer sizes, cmd_sg_entries=255 and
> bs=8M) [passed]time 17.191s ...
> runtime 17.191s ... 17.117s
> srp/011 (Block I/O on top of multipath concurrently with logout and
> login) [passed] runtime 40.835s ...
> runtime 40.835s ... 38.728s
> srp/012 (dm-mpath on top of multiple I/O schedulers) [passed]
> runtime 23.703s ... 24.763s
> srp/013 (Direct I/O using a discontiguous buffer) [passed]
> runtime 11.279s ... 9.265s
> srp/014 (Run sg_reset while I/O is ongoing) [passed]
> runtime 39.110s ... 37.929s
> srp/015 (File I/O on top of multipath concurrently with logout and login
> (mq) ussrp/015
> (File I/O on top of multipath concurrently with logout and login
> (mq) using the SoftiWARP (siw) driver) [passed]
> runtime 40.027s ... 40.220s
>
> If I try to run the SRP test 002 with the soft-RoCE driver, the
> following appears:
>
> [ 749.901966] ================================
> [ 749.903638] WARNING: inconsistent lock state
> [ 749.905376] 5.18.0-rc5-dbg+ #1 Not tainted
> [ 749.907039] --------------------------------
> [ 749.908699] inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} usage.
> [ 749.910646] ksoftirqd/5/40 [HC0[0]:SC1[1]:HE0:SE0] takes:
> [ 749.912499] ffff88818244d350 (&xa->xa_lock#14){+.?.}-{2:2}, at:
> rxe_pool_get_index+0x73/0x170 [rdma_rxe]
> [ 749.914691] {SOFTIRQ-ON-W} state was registered at:
> [ 749.916648] __lock_acquire+0x45b/0xce0
> [ 749.918599] lock_acquire+0x18a/0x450
> [ 749.920480] _raw_spin_lock+0x34/0x50
> [ 749.922580] __rxe_add_to_pool+0xcc/0x140 [rdma_rxe]
> [ 749.924583] rxe_alloc_pd+0x2d/0x40 [rdma_rxe]
> [ 749.926394] __ib_alloc_pd+0xa3/0x270 [ib_core]
> [ 749.928579] ib_mad_port_open+0x44a/0x790 [ib_core]
> [ 749.930640] ib_mad_init_device+0x8e/0x110 [ib_core]
> [ 749.932495] add_client_context+0x26a/0x330 [ib_core]
> [ 749.934302] enable_device_and_get+0x169/0x2b0 [ib_core]
> [ 749.936217] ib_register_device+0x26f/0x330 [ib_core]
> [ 749.938020] rxe_register_device+0x1b4/0x1d0 [rdma_rxe]
> [ 749.939794] rxe_add+0x8c/0xc0 [rdma_rxe]
> [ 749.941552] rxe_net_add+0x5b/0x90 [rdma_rxe]
> [ 749.943356] rxe_newlink+0x71/0x80 [rdma_rxe]
> [ 749.945182] nldev_newlink+0x21e/0x370 [ib_core]
> [ 749.946917] rdma_nl_rcv_msg+0x200/0x410 [ib_core]
> [ 749.948657] rdma_nl_rcv+0x140/0x220 [ib_core]
> [ 749.950373] netlink_unicast+0x307/0x460
> [ 749.952063] netlink_sendmsg+0x422/0x750
> [ 749.953672] __sys_sendto+0x1c2/0x250
> [ 749.955281] __x64_sys_sendto+0x7f/0x90
> [ 749.956849] do_syscall_64+0x35/0x80
> [ 749.958353] entry_SYSCALL_64_after_hwframe+0x44/0xae
> [ 749.959942] irq event stamp: 1411849
> [ 749.961517] hardirqs last enabled at (1411848): [<ffffffff810cdb28>]
> __local_bh_enable_ip+0x88/0xf0
> [ 749.963338] hardirqs last disabled at (1411849): [<ffffffff81ebf24d>]
> _raw_spin_lock_irqsave+0x5d/0x60
> [ 749.965214] softirqs last enabled at (1411838): [<ffffffff82200467>]
> __do_softirq+0x467/0x6e1
> [ 749.967027] softirqs last disabled at (1411843): [<ffffffff810cd947>]
> run_ksoftirqd+0x37/0x60
To this, Please use this patch series
news://nntp.lore.kernel.org:119/20220422194416.983549-1-yanjun.zhu@linux.dev
Zhu Yanjun
>
> I think the above is strong evidence that there is something wrong with the
> soft-RoCE driver.
>
> Thanks,
>
> Bart.
next prev parent reply other threads:[~2022-05-07 0:29 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-06 18:11 Apparent regression in blktests since 5.18-rc1+ Bob Pearson
2022-05-07 0:10 ` Bart Van Assche
2022-05-07 0:29 ` Yanjun Zhu [this message]
2022-05-07 1:29 ` Jason Gunthorpe
2022-05-07 1:55 ` Yanjun Zhu
2022-05-07 13:43 ` Bob Pearson
2022-05-08 4:13 ` Bart Van Assche
2022-05-10 15:24 ` Pearson, Robert B
2022-05-12 21:57 ` Bob Pearson
2022-05-12 22:25 ` Bart Van Assche
2022-05-13 0:41 ` Bob Pearson
2022-05-13 3:40 ` Bart Van Assche
2022-05-17 15:21 ` Bob Pearson
2022-05-17 20:44 ` Bart Van Assche
2022-05-17 20:54 ` Bob Pearson
2022-05-17 20:59 ` Bob Pearson
2022-05-08 8:43 ` Yanjun Zhu
2022-05-09 8:01 ` Zhu Yanjun
2022-05-09 11:52 ` Jason Gunthorpe
2022-05-09 12:31 ` Yanjun Zhu
2022-05-09 12:33 ` Jason Gunthorpe
2022-05-09 12:42 ` Yanjun Zhu
2022-05-07 13:40 ` Bob Pearson
2022-05-09 6:56 ` Thorsten Leemhuis
2022-05-10 3:53 ` Bart Van Assche
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4b0153c7-a8e9-98de-26ae-d421434a116d@linux.dev \
--to=yanjun.zhu@linux.dev \
--cc=bmt@zurich.ibm.com \
--cc=bvanassche@acm.org \
--cc=jgg@nvidia.com \
--cc=linux-rdma@vger.kernel.org \
--cc=rpearsonhpe@gmail.com \
--cc=zyjzyj2000@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).