linux-kernel - Re: [PATCH v5] printk: Userspace format enumeration support

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: <YEi4qwBAd/O+sXyq@kroah.com>
Date:   Wed, 10 Mar 2021 13:16:43 +0100
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     Chris Down <chris@...isdown.name>
Cc:     linux-kernel@...r.kernel.org, Petr Mladek <pmladek@...e.com>,
        Sergey Senozhatsky <sergey.senozhatsky@...il.com>,
        John Ogness <john.ogness@...utronix.de>,
        Johannes Weiner <hannes@...xchg.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Steven Rostedt <rostedt@...dmis.org>,
        Kees Cook <keescook@...omium.org>, kernel-team@...com
Subject: Re: [PATCH v5] printk: Userspace format enumeration support

On Wed, Mar 10, 2021 at 12:12:57PM +0000, Chris Down wrote:
> Greg Kroah-Hartman writes:
> > On Wed, Mar 10, 2021 at 02:30:31AM +0000, Chris Down wrote:
> > > +	ps->file = debugfs_create_file(pi_get_module_name(mod), 0444, dfs_index,
> > > +				       ps, &dfs_index_fops);
> > > +
> > > +	if (IS_ERR(ps->file)) {
> > > +		pi_sec_remove(mod);
> > > +		return;
> > > +	}
> > 
> > No need to check this and try to clean up if there is a problem, just
> > save the pointer off and call debugfs_remove() when you want to clean
> > up.
> 
> Petr, what are your thoughts on this, since you requested the cleanup on
> debugfs failure? :-)

There is nothing to "clean up" if there is a debugfs failure here so I
don't see the need.

> > Or better yet, no need to save anything, you can always look it up when
> > you want to remove it, that will save you one pointer per module.
> 
> That's a good point, and with that maybe we can even do away with the pi_sec
> entirely then since that only leaves start/end pointers which we can
> calculate on demand from existing data.

Please do, that makes the code simpler overall.

thanks,

greg k-h