linux-kernel - Re: [PATCH 1/2] f2fs: add sysfs symbolic link to kobject with volume name

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:   Wed, 22 Jul 2020 09:43:56 -0700
From:   Jaegeuk Kim <jaegeuk@...nel.org>
To:     Greg KH <gregkh@...uxfoundation.org>
Cc:     Daeho Jeong <daeho43@...il.com>, linux-kernel@...r.kernel.org,
        linux-f2fs-devel@...ts.sourceforge.net, kernel-team@...roid.com,
        Daeho Jeong <daehojeong@...gle.com>
Subject: Re: [PATCH 1/2] f2fs: add sysfs symbolic link to kobject with volume
 name

On 07/19, Greg KH wrote:
> On Sun, Jul 19, 2020 at 02:44:08PM +0900, Daeho Jeong wrote:
> > From: Daeho Jeong <daehojeong@...gle.com>
> > 
> > Added a symbolic link directory pointing to its device name
> > directory using the volume name of the partition in sysfs.
> > (i.e., /sys/fs/f2fs/vol_#x -> /sys/fs/f2fs/sda1)
> 
> No, please no.
> 
> That is already created today for you in /dev/disk/  The kernel does not
> need to do this again.
> 
> If your distro/system/whatever does not provide you with /dev/disk/ and
> all of the symlinks in there, then work with your distro/system/whatever
> to do so.

I don't get the point, since /dev/disk points device node, not any sysfs entry.
Do you mean we need to create symlink to /sys/fs/f2fs/dm-X in /dev/disk?

> 
> Again, no need to do this on a per-filesystem-basis when we already have
> this around for all filesystems, and have had it for 15+ years now.

Could you point out where we can get this? And, the label support depends
on per-filesystem design. I'm not sure how this can be generic enough.

> 
> thanks,
> 
> greg k-h