linux-kernel - Re: [PATCH] block: fix deadlock between sd_remove & sd

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <331c036b-322e-0658-ff88-4cce080c5aba@huaweicloud.com>
Date: Fri, 19 Jul 2024 11:15:22 +0800
From: Yu Kuai <yukuai1@...weicloud.com>
To: Bart Van Assche <bvanassche@....org>, YangYang <yang.yang@...o.com>,
 Yu Kuai <yukuai1@...weicloud.com>
Cc: Jens Axboe <axboe@...nel.dk>, linux-block@...r.kernel.org,
 linux-kernel@...r.kernel.org, "yukuai (C)" <yukuai3@...wei.com>
Subject: Re: [PATCH] block: fix deadlock between sd_remove & sd_release



在 2024/07/18 0:23, Bart Van Assche 写道:
> On 7/17/24 3:15 AM, YangYang wrote:
>> These sysfs nodes are in different directories, the scsi node located
>> at /sys/bus/scsi/devices/0:0:0:0 and the gendisk node located at
>> /sys/block/sda. Would it be necessary to wait for the completion of
>> the scsi sysfs nodes' read/write operations before removing the
>> gendisk sysfs node?
> 
> No. sysfs_remove_files() waits for pending read and write operations to
> complete.
> 

So I check this and actually gendisk kobject is a child of scsi disk
kobject, not the other way. sda/device is just a symbol link.

sda/device -> ../../../14:2:0:0.

So, the sysfs api for gendisk itself doesn't issue IO, and as long as
related driver doesn't create new such api under gendisk, this won't
be a problem.

Thanks,
Kuai
> Bart.
> 
> .
>