linux-kernel - Re: [PATCH 1/2] zram: fix crashes due to use of cpu hotplug multistate

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]

Message-ID: <YG66OWzum5DGcSTn@kroah.com>
Date:   Thu, 8 Apr 2021 10:09:29 +0200
From:   Greg KH <gregkh@...uxfoundation.org>
To:     Jiri Kosina <jikos@...nel.org>
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        Luis Chamberlain <mcgrof@...nel.org>,
        Minchan Kim <minchan@...nel.org>, keescook@...omium.org,
        dhowells@...hat.com, hch@...radead.org, mbenes@...e.com,
        ngupta@...are.org, sergey.senozhatsky.work@...il.com,
        axboe@...nel.dk, linux-block@...r.kernel.org,
        linux-kernel@...r.kernel.org, Jiri Kosina <jikos@...os.cz>
Subject: Re: [PATCH 1/2] zram: fix crashes due to use of cpu hotplug
 multistate

On Thu, Apr 08, 2021 at 10:01:23AM +0200, Jiri Kosina wrote:
> On Thu, 8 Apr 2021, Greg KH wrote:
> 
> > Removing a module from a system has always been "let's try it and see!"
> > type of operation for a very long time.  
> 
> Which part of it?
> 
> If there is a driver/subsystem code that can't handle the reverse 
> operation to modprobe, it clearly can't handle error handling during 
> modprobe (which, one would hope, is supported), and should be fixed.

Huh?  No, that's not the issue here, it's the issue of different
userspace code paths into the module at the same time that it is trying
to be unloaded.  That has nothing to do with loading the module the
first time as userspace is not touching those apis yet.

> If there is a particular issue in kernel dynamic linker that causes crash 
> on module removal, we'd better off fixing it. Is there one such that makes 
> you claim module removal unsupported?

The linker has nothing to do with this, it's userspace tasks touching
code paths.

thanks,

greg k-h