From: Sebastian Ott <sebott@linux.ibm.com>
To: Christoph Hellwig <hch@lst.de>
Cc: Ming Lei <ming.lei@redhat.com>, Hannes Reinecke <hare@suse.com>,
Jens Axboe <axboe@kernel.dk>,
linux-block@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: too large sg segments with commit 09324d32d2a08
Date: Wed, 5 Jun 2019 11:13:11 +0200 (CEST) [thread overview]
Message-ID: <alpine.LFD.2.21.1906051057200.2118@schleppi> (raw)
Hi,
this warning turned up on s390:
[ 7.041512] ------------[ cut here ]------------
[ 7.041518] DMA-API: nvme 0000:00:00.0: mapping sg segment longer than device claims to support [len=106496] [max=65536]
[ 7.041531] WARNING: CPU: 1 PID: 229 at kernel/dma/debug.c:1233 debug_dma_map_sg+0x21e/0x350
[ 7.041537] Modules linked in: scm_block(+) eadm_sch sch_fq_codel autofs4
[ 7.041547] CPU: 1 PID: 229 Comm: systemd-udevd Not tainted 5.2.0-rc3-00002-g112d38aa4733-dirty #146
[ 7.041552] Hardware name: IBM 3906 M03 703 (LPAR)
[ 7.041558] Krnl PSW : 0704d00180000000 00000000af580b6e (debug_dma_map_sg+0x21e/0x350)
[ 7.041566] R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:1 PM:0 RI:0 EA:3
[ 7.041572] Krnl GPRS: 0000000095969122 0000000080000000 000000000000006c 00000000af5624bc
[ 7.041578] 0000000000000007 0000000000000001 0000000000010000 0000000000000001
[ 7.041583] 00000000b081d278 00000000fea06888 000000008fa43400 00000000f255b418
[ 7.041589] 00000000f4a28100 ffffffff00000000 00000000af580b6a 000003e0004a36e0
[ 7.041599] Krnl Code: 00000000af580b5e: c02000566fd7 larl %r2,b004eb0c
00000000af580b64: c0e5fffad1fe brasl %r14,af4daf60
#00000000af580b6a: a7f40001 brc 15,af580b6c
>00000000af580b6e: c010005f45f5 larl %r1,b0169758
00000000af580b74: e31010000012 lt %r1,0(%r1)
00000000af580b7a: a774000f brc 7,af580b98
00000000af580b7e: c010005fac23 larl %r1,b01763c4
00000000af580b84: e31010000012 lt %r1,0(%r1)
[ 7.041620] Call Trace:
[ 7.041626] ([<00000000af580b6a>] debug_dma_map_sg+0x21a/0x350)
[ 7.041633] [<00000000afbe2152>] nvme_queue_rq+0x49a/0xd18
[ 7.041639] [<00000000afa178d0>] __blk_mq_try_issue_directly+0x108/0x1f0
[ 7.041645] [<00000000afa18e96>] blk_mq_request_issue_directly+0x4e/0x70
[ 7.041651] [<00000000afa18f42>] blk_mq_try_issue_list_directly+0x8a/0x118
[ 7.041657] [<00000000afa1e42e>] blk_mq_sched_insert_requests+0x1c6/0x350
[ 7.041663] [<00000000afa18e40>] blk_mq_flush_plug_list+0x4f8/0x500
[ 7.041669] [<00000000afa0bb3e>] blk_flush_plug_list+0x106/0x110
[ 7.041674] [<00000000afa0bb7c>] blk_finish_plug+0x34/0x50
[ 7.041680] [<00000000af6938c2>] read_pages+0x152/0x160
[ 7.041687] [<00000000af693b06>] __do_page_cache_readahead+0x236/0x268
[ 7.041693] [<00000000af694458>] force_page_cache_readahead+0x110/0x120
[ 7.041699] [<00000000af683fa4>] generic_file_buffered_read+0x144/0x968
[ 7.041706] [<00000000af74d14c>] new_sync_read+0x13c/0x1b8
[ 7.041712] [<00000000af74f9fa>] vfs_read+0x82/0x138
[ 7.041717] [<00000000af74fd92>] ksys_read+0x62/0xd8
[ 7.041724] [<00000000afe60e00>] system_call+0x2b0/0x2d0
[ 7.041729] 1 lock held by systemd-udevd/229:
[ 7.041734] #0: 00000000f715c4f3 (rcu_read_lock){....}, at: hctx_lock+0x28/0xf8
[ 7.041743] Last Breaking-Event-Address:
[ 7.041749] [<00000000af580b6a>] debug_dma_map_sg+0x21a/0x350
[ 7.041754] irq event stamp: 14457
[ 7.041760] hardirqs last enabled at (14465): [<00000000af560a2c>] console_unlock+0x63c/0x6a8
[ 7.041766] hardirqs last disabled at (14472): [<00000000af5604ba>] console_unlock+0xca/0x6a8
[ 7.041773] softirqs last enabled at (11488): [<00000000afa20ed4>] get_gendisk+0xf4/0x148
[ 7.041779] softirqs last disabled at (11486): [<00000000afa20e48>] get_gendisk+0x68/0x148
[ 7.041784] ---[ end trace 9142fc6f63a22c6e ]---
The length of the sg entry created by blk_rq_map_sg is indeed largen than
the dma max_segment_size.
Bisect points to 09324d32d2a0 ("block: force an unlimited segment size on queues with a virt boundary")
Regards,
Sebastian
next reply other threads:[~2019-06-05 9:13 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-05 9:13 Sebastian Ott [this message]
2019-06-05 10:09 ` too large sg segments with commit 09324d32d2a08 Christoph Hellwig
2019-06-05 13:30 ` Christoph Hellwig
2019-06-05 13:57 ` Sebastian Ott
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=alpine.LFD.2.21.1906051057200.2118@schleppi \
--to=sebott@linux.ibm.com \
--cc=axboe@kernel.dk \
--cc=hare@suse.com \
--cc=hch@lst.de \
--cc=linux-block@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=ming.lei@redhat.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