From: "Bernard Metzler" <BMT@zurich.ibm.com>
To: "linux-rdma" <linux-rdma@vger.kernel.org>
Cc: "Bob Pearson" <rpearsonhpe@gmail.com>,
"Jason Gunthorpe" <jgg@nvidia.com>,
"Bart Van Assche" <bvanassche@acm.org>
Subject: Issus with blktest/srp on 5.15-rc1 and rdma_rxe
Date: Thu, 16 Sep 2021 19:36:43 +0000 [thread overview]
Message-ID: <OF54F8428F.7EA7E570-ON00258752.006AB21B-00258752.006BBB93@ibm.com> (raw)
In-Reply-To:
Hi,
if I run the complete srp test series from the blktests suite,
the dmesg log contains many rdma_rxe messages of type:
rdma_rxe: rxe_invalidate_mr: rkey (n) doesn't match mr->ibmr.rkey (n + 1)
where 'n' is the current key. I expect this is not intended
behavior.
I am at commit 1b789bd4dbd48a92f5427d9c37a72a8f6ca17754
Furthermore, running ./check -q srp/005 sometimes I get this:
[ 308.903330] sd 11:0:0:1: [sde] Attached SCSI disk
[ 308.917293] scsi 11:0:0:1: alua: Detached
[ 308.918191] BUG: kernel NULL pointer dereference, address: 0000000000000000
[ 308.918223] #PF: supervisor instruction fetch in kernel mode
[ 308.918242] #PF: error_code(0x0010) - not-present page
[ 308.918259] PGD 0 P4D 0
[ 308.918271] Oops: 0010 [#1] SMP PTI
[ 308.918285] CPU: 1 PID: 4214 Comm: kworker/1:255 Not tainted 5.15.0-rc1+ #4
[ 308.918309] Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./Z77 Extreme6, BIOS P2.80 07/01/2013
[ 308.918338] Workqueue: srp_remove srp_remove_work [ib_srp]
[ 308.918362] RIP: 0010:0x0
[ 308.918375] Code: Unable to access opcode bytes at RIP 0xffffffffffffffd6.
[ 308.918397] RSP: 0018:ffffb6124b9a3b68 EFLAGS: 00010086
[ 308.918414] RAX: 0000000000000001 RBX: ffffb6124b9a3ce0 RCX: 0000000000000000
[ 308.918437] RDX: 0000000000000000 RSI: ffffb6124b9a3c50 RDI: ffff966063a27a00
[ 308.918459] RBP: ffffb6124b9a3bb0 R08: ffff966067481c00 R09: ffffeb578489b808
[ 308.918481] R10: ffff966043c0f200 R11: ffffb6124b9a3d00 R12: ffff966063a27a00
[ 308.918503] R13: 0000000000000004 R14: 0000000000000000 R15: ffffb6124b9a3c50
[ 308.918524] FS: 0000000000000000(0000) GS:ffff966157680000(0000) knlGS:0000000000000000
[ 308.918550] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 308.918568] CR2: ffffffffffffffd6 CR3: 000000005060a004 CR4: 00000000001706e0
[ 308.918590] Call Trace:
[ 308.918601] __ib_process_cq+0x89/0x150 [ib_core]
[ 308.918640] ib_process_cq_direct+0x30/0x50 [ib_core]
[ 308.918669] ? xas_store+0x331/0x640
[ 308.918684] ? free_unref_page_commit.isra.135+0x91/0x140
[ 308.918705] ? free_unref_page+0x6e/0xd0
[ 308.918719] ? __free_pages+0xa3/0xc0
[ 308.918733] ? kfree+0x32f/0x3b0
[ 308.918748] srp_destroy_qp+0x24/0x40 [ib_srp]
[ 308.918767] srp_free_ch_ib+0x77/0x180 [ib_srp]
[ 308.918784] srp_remove_work+0xde/0x1a0 [ib_srp]
[ 308.918801] process_one_work+0x1d0/0x380
[ 308.918817] worker_thread+0x37/0x390
[ 308.918831] ? process_one_work+0x380/0x380
[ 308.918846] kthread+0x12f/0x150
[ 308.918859] ? set_kthread_struct+0x40/0x40
[ 308.918874] ret_from_fork+0x22/0x30
With ./check -q srp/008 I sometimes get something similar:
[ 1772.149274] sd 11:0:0:1: [sde] Attached SCSI disk
[ 1772.150096] scsi 11:0:0:2: alua: Detached
[ 1772.150184] blk_update_request: I/O error, dev sde, sector 8 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 0
[ 1772.151653] blk_update_request: I/O error, dev sde, sector 8 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
[ 1772.153080] Buffer I/O error on dev sde, logical block 1, async page read
[ 1772.169139] scsi 11:0:0:1: alua: Detached
[ 1772.169446] BUG: kernel NULL pointer dereference, address: 0000000000000000
[ 1772.170881] #PF: supervisor instruction fetch in kernel mode
[ 1772.172297] #PF: error_code(0x0010) - not-present page
[ 1772.173751] PGD 0 P4D 0
[ 1772.175165] Oops: 0010 [#1] SMP PTI
[ 1772.176575] CPU: 3 PID: 8654 Comm: kworker/3:60 Not tainted 5.15.0-rc1+ #4
[ 1772.177995] Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./Z77 Extreme6, BIOS P2.80 07/01/2013
[ 1772.179430] Workqueue: srp_remove srp_remove_work [ib_srp]
[ 1772.180859] RIP: 0010:0x0
[ 1772.182276] Code: Unable to access opcode bytes at RIP 0xffffffffffffffd6.
[ 1772.183705] RSP: 0018:ffffa9710a2f7b68 EFLAGS: 00010086
[ 1772.185129] RAX: 0000000000000001 RBX: ffffa9710a2f7c50 RCX: 0000000000000000
[ 1772.186566] RDX: 0000000000000000 RSI: ffffa9710a2f7c08 RDI: ffff91a703825a00
[ 1772.187994] RBP: ffffa9710a2f7bb0 R08: ffff91a7184f8300 R09: 0000000000000000
[ 1772.189425] R10: ffff91a7869ce000 R11: ffffa9710a2f7d00 R12: ffff91a703825a00
[ 1772.190848] R13: 0000000000000002 R14: 0000000000000000 R15: ffffa9710a2f7c08
[ 1772.192270] FS: 0000000000000000(0000) GS:ffff91a817780000(0000) knlGS:0000000000000000
[ 1772.193689] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1772.195118] CR2: ffffffffffffffd6 CR3: 0000000111458006 CR4: 00000000001706e0
[ 1772.196558] Call Trace:
[ 1772.197975] __ib_process_cq+0x89/0x150 [ib_core]
[ 1772.199402] ib_process_cq_direct+0x30/0x50 [ib_core]
[ 1772.200830] ? put_cpu_partial+0x98/0xb0
[ 1772.202260] ? __slab_free+0x226/0x3c0
[ 1772.203664] ? __slab_free+0x226/0x3c0
[ 1772.205038] ? xas_store+0x331/0x640
[ 1772.206404] ? rxe_elem_release+0x4f/0x60 [rdma_rxe]
[ 1772.207763] ? kfree+0x372/0x3b0
[ 1772.209101] ? srp_destroy_fr_pool+0x43/0x50 [ib_srp]
[ 1772.210439] srp_destroy_qp+0x24/0x40 [ib_srp]
[ 1772.211759] srp_free_ch_ib+0x77/0x180 [ib_srp]
[ 1772.213093] srp_remove_work+0xde/0x1a0 [ib_srp]
[ 1772.214417] process_one_work+0x1d0/0x380
[ 1772.215756] worker_thread+0x37/0x390
[ 1772.217082] ? process_one_work+0x380/0x380
[ 1772.218422] kthread+0x12f/0x150
[ 1772.219744] ? set_kthread_struct+0x40/0x40
[ 1772.221059] ret_from_fork+0x22/0x30
Many thanks,
Bernard.
next reply other threads:[~2021-09-16 19:36 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CAHj4cs9Rzte5zbgy7o158m7JA8dbSEpxy5oR-+K0NQCK1gxG=Q@mail.gmail.com>
2021-09-16 19:36 ` Bernard Metzler [this message]
2021-09-16 22:21 ` Issus with blktest/srp on 5.15-rc1 and rdma_rxe Robert Pearson
2021-09-17 8:23 ` Bernard Metzler
2021-09-18 2:56 ` Bart Van Assche
2021-09-18 12:15 ` Yi Zhang
2021-09-21 20:08 ` Still issues with blktest/srp on 5.15-rc1 and software rdma providers Bernard Metzler
2021-09-21 20:16 ` Bart Van Assche
2021-09-17 8:15 ` Re: Issus with blktest/srp on 5.15-rc1 and rdma_rxe Bernard Metzler
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=OF54F8428F.7EA7E570-ON00258752.006AB21B-00258752.006BBB93@ibm.com \
--to=bmt@zurich.ibm.com \
--cc=bvanassche@acm.org \
--cc=jgg@nvidia.com \
--cc=linux-rdma@vger.kernel.org \
--cc=rpearsonhpe@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).