linux-kernel - Re: [PATCH] random: Initialize vsprintf's pointer hash once the random core is ready.

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: <YuO7ommLFTSLQQ6h@linutronix.de>
Date:   Fri, 29 Jul 2022 12:51:14 +0200
From:   Sebastian Andrzej Siewior <bigeasy@...utronix.de>
To:     "Jason A. Donenfeld" <Jason@...c4.com>
Cc:     linux-kernel@...r.kernel.org, Theodore Ts'o <tytso@....edu>,
        Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
        John Ogness <john.ogness@...utronix.de>,
        Mike Galbraith <efault@....de>, Petr Mladek <pmladek@...e.com>,
        Rasmus Villemoes <linux@...musvillemoes.dk>,
        Sergey Senozhatsky <senozhatsky@...omium.org>,
        Steven Rostedt <rostedt@...dmis.org>,
        Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [PATCH] random: Initialize vsprintf's pointer hash once the
 random core is ready.

On 2022-07-29 12:38:06 [+0200], Jason A. Donenfeld wrote:
> Hi Sebastian,
Hi Jason,

> On Fri, Jul 29, 2022 at 12:21:27PM +0200, Sebastian Andrzej Siewior wrote:
> > So launching a worker to obtain the random data? That would mean that
> > the first %p print won't have nothing, right? I could do it as part of
> 
> "First" isn't very meaningful here. If the rng isn't initialized by
> add_bootloader_randomness() or similar, then it'll almost miss some
> amount of %p anyway.

only if that printk happens during boot. But it could happen much later.
In that case !RT won't lose that pointer but RT will.

> But anyway, it sounds like you only need to hoist into a worker IF
> you're `IS_ENABLED(CONFIG_PREEMPT_RT) && in_hardirq()`, right? So just
> conditionalize it on that, and this should have pretty minimal impact.

I need always to hoist into a worker because there could warning in a
preempt-off region leading to this error.
Maybe I am putting too much importance into this. Let me do what you
suggest and always lose that first pointer and if someone complains than
maybe we think about something else…

> I don't think this patch will require touching random.c.
> 
> Jason

Sebastian