linux-kernel - Re: [PATCH] scsi: sd: call device_del() if device_add

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

Date:   Thu, 31 Mar 2022 18:11:52 +0300
From:   Dan Carpenter <dan.carpenter@...cle.com>
To:     James Bottomley <jejb@...ux.ibm.com>
Cc:     Wenchao Hao <haowenchao@...wei.com>,
        syzkaller-bugs@...glegroups.com, fmdefrancesco@...il.com,
        axboe@...nel.dk, linux-block@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-scsi@...r.kernel.org,
        martin.petersen@...cle.com,
        syzbot+f08c77040fa163a75a46@...kaller.appspotmail.com,
        linfeilong@...wei.com
Subject: Re: [PATCH] scsi: sd: call device_del() if device_add_disk() fails

On Thu, Mar 31, 2022 at 10:19:57AM -0400, James Bottomley wrote:
> On Thu, 2022-03-31 at 16:42 +0300, Dan Carpenter wrote:
> [...]
> > Also, I don't really understand why we don't have to call
> > put_device(&sdkp->disk_dev) at the end of sd_remove().
> 
> That's because the final put is done by the gendisk ->free_disk()
> function which is scsi_disk_free_disk().  Most of the gendisk functions
> we provide convert a gendisk to a scsi_disk (via the gendisk
> private_data), so the sdkp has to live as long as the gendisk.
> 
> James

Thanks James.

Ah...  And scsi_disk_free_disk() will not be called unless
device_add_disk() succeeds.  So Wenchao Hao's patch is correct.

And after that there is just a missing device_del() and also I see
another problem where if device_add() fails then we need to call
put_disk(gd); on that error path.

regards,
dan carpenter