From: Jens Axboe <[email protected]>
To: Guangwu Zhang <[email protected]>,
[email protected], Ming Lei <[email protected]>,
Jeff Moyer <[email protected]>,
[email protected]
Subject: Re: [bug report] BUG: KASAN: out-of-bounds in io_req_local_work_add+0x3b1/0x4a0
Date: Wed, 21 Jun 2023 08:04:30 -0600 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>
On 6/21/23 7:40?AM, Jens Axboe wrote:
> On 6/21/23 1:38?AM, Guangwu Zhang wrote:
>> HI,
>> Found the io_req_local_work_add error when run liburing testing.
>>
>> kernel repo :
>> Merge branch 'for-6.5/block' into for-next
>> * for-6.5/block:
>> reiserfs: fix blkdev_put() warning from release_journal_dev()
>>
>> [ 1733.389012] BUG: KASAN: out-of-bounds in io_req_local_work_add+0x3b1/0x4a0
>> [ 1733.395900] Read of size 4 at addr ffff888133320458 by task
>> iou-wrk-97057/97138
>> [ 1733.403205]
>> [ 1733.404706] CPU: 4 PID: 97138 Comm: iou-wrk-97057 Kdump: loaded Not
>> tainted 6.4.0-rc3.kasan+ #1
>> [ 1733.413404] Hardware name: Dell Inc. PowerEdge R740/06WXJT, BIOS
>> 2.13.3 12/13/2021
>> [ 1733.420972] Call Trace:
>> [ 1733.423425] <TASK>
>> [ 1733.425533] dump_stack_lvl+0x33/0x50
>> [ 1733.429207] print_address_description.constprop.0+0x2c/0x3e0
>> [ 1733.434959] print_report+0xb5/0x270
>> [ 1733.438539] ? kasan_addr_to_slab+0x9/0xa0
>> [ 1733.442639] ? io_req_local_work_add+0x3b1/0x4a0
>> [ 1733.447258] kasan_report+0xcf/0x100
>> [ 1733.450839] ? io_req_local_work_add+0x3b1/0x4a0
>> [ 1733.455456] io_req_local_work_add+0x3b1/0x4a0
>> [ 1733.459903] ? __pfx_io_req_local_work_add+0x10/0x10
>> [ 1733.464871] ? __schedule+0x616/0x1530
>> [ 1733.468622] __io_req_task_work_add+0x1bc/0x270
>> [ 1733.473156] io_issue_sqe+0x55a/0xe80
>> [ 1733.476831] io_wq_submit_work+0x23e/0xa00
>> [ 1733.480930] io_worker_handle_work+0x2f5/0xa80
>> [ 1733.485384] io_wq_worker+0x6c5/0x9d0
>> [ 1733.489051] ? __pfx_io_wq_worker+0x10/0x10
>> [ 1733.493246] ? _raw_spin_lock_irq+0x82/0xe0
>> [ 1733.497430] ? __pfx_io_wq_worker+0x10/0x10
>> [ 1733.501616] ret_from_fork+0x29/0x50
>> [ 1733.505204] </TASK>
>> [ 1733.507396]
>> [ 1733.508894] Allocated by task 97057:
>> [ 1733.512475] kasan_save_stack+0x1e/0x40
>> [ 1733.516313] kasan_set_track+0x21/0x30
>> [ 1733.520068] __kasan_slab_alloc+0x83/0x90
>> [ 1733.524080] kmem_cache_alloc_bulk+0x13a/0x1e0
>> [ 1733.528526] __io_alloc_req_refill+0x238/0x510
>> [ 1733.532971] io_submit_sqes+0x65a/0xcd0
>> [ 1733.536810] __do_sys_io_uring_enter+0x4e9/0x830
>> [ 1733.541430] do_syscall_64+0x59/0x90
>> [ 1733.545010] entry_SYSCALL_64_after_hwframe+0x72/0xdc
>> [ 1733.550071]
>> [ 1733.551571] The buggy address belongs to the object at ffff8881333203c0
>> [ 1733.551571] which belongs to the cache io_kiocb of size 224
>> [ 1733.563816] The buggy address is located 152 bytes inside of
>> [ 1733.563816] 224-byte region [ffff8881333203c0, ffff8881333204a0)
>> [ 1733.575544]
>> [ 1733.577042] The buggy address belongs to the physical page:
>> [ 1733.582617] page:00000000edbe178c refcount:1 mapcount:0
>> mapping:0000000000000000 index:0x0 pfn:0x133320
>> [ 1733.592011] head:00000000edbe178c order:1 entire_mapcount:0
>> nr_pages_mapped:0 pincount:0
>> [ 1733.600096] memcg:ffff88810cd49001
>> [ 1733.603501] flags:
>> 0x17ffffc0010200(slab|head|node=0|zone=2|lastcpupid=0x1fffff)
>> [ 1733.610896] page_type: 0xffffffff()
>> [ 1733.614390] raw: 0017ffffc0010200 ffff888101222280 ffffea0004473900
>> 0000000000000002
>> [ 1733.622128] raw: 0000000000000000 0000000000190019 00000001ffffffff
>> ffff88810cd49001
>> [ 1733.629866] page dumped because: kasan: bad access detected
>> [ 1733.635439]
>> [ 1733.636938] Memory state around the buggy address:
>> [ 1733.641731] ffff888133320300: 00 00 00 00 00 00 00 00 00 00 00 00
>> fc fc fc fc
>> [ 1733.648952] ffff888133320380: fc fc fc fc fc fc fc fc 00 00 00 00
>> 00 00 00 00
>> [ 1733.656169] >ffff888133320400: 00 00 00 00 00 00 00 00 00 00 00 00
>> 00 00 00 00
>> [ 1733.663389] ^
>> [ 1733.669743] ffff888133320480: 00 00 00 00 fc fc fc fc fc fc fc fc
>> fc fc fc fc
>> [ 1733.676961] ffff888133320500: 00 00 00 00 00 00 00 00 00 00 00 00
>> 00 00 00 00
>
> I appreciate you running tests and sending in failures, but can you
> please be more specific about what exactly was run? We seem to need to
> do this dance every time, which is just wasting time. So:
>
> 1) What test triggered this?
> 2) Was it invoked with any arguments?
>
> In general, a good bug report should include exactly HOW you ended
> up there.
Another thing that's really handy in bug reports like this is a
disassemble of where the bad memory access occured. We have this:
BUG: KASAN: out-of-bounds in io_req_local_work_add+0x3b1/0x4a0
and you can look that up in gdb, for example.
--
Jens Axboe
prev parent reply other threads:[~2023-06-21 14:05 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-06-21 7:38 [bug report] BUG: KASAN: out-of-bounds in io_req_local_work_add+0x3b1/0x4a0 Guangwu Zhang
2023-06-21 13:40 ` Jens Axboe
2023-06-21 14:04 ` Ming Lei
2023-06-21 15:56 ` Jens Axboe
2023-06-23 5:51 ` Guangwu Zhang
2023-06-23 12:14 ` Jeff Moyer
2023-06-23 12:30 ` Pavel Begunkov
2023-06-21 14:04 ` Jens Axboe [this message]
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 \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/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