linux-kernel - Re: [RFC PATCH] sched/cputime: drop local_irq_safe() in vtime_init

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: <570F9DFD.5080807@linutronix.de>
Date:	Thu, 14 Apr 2016 15:41:17 +0200
From:	Sebastian Andrzej Siewior <bigeasy@...utronix.de>
To:	Frederic Weisbecker <fweisbec@...il.com>
Cc:	Ingo Molnar <mingo@...hat.com>,
	Peter Zijlstra <peterz@...radead.org>,
	linux-kernel@...r.kernel.org,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Rik van Riel <riel@...hat.com>,
	Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [RFC PATCH] sched/cputime: drop local_irq_safe() in
 vtime_init_idle()

On 04/14/2016 03:18 PM, Frederic Weisbecker wrote:
> I think I tried this but I got lockdep warnings because the other updaters
> happen on IRQ.

I booted and played a little but lockdep didn't scream here so far.

> In practive we can't be interrupted since the idle task hasn't even started.
> 
> Now we can probably get rid of the write_seqcount_stuff here because what
> is initialized there is supposed to be visible by the idle task once it runs.
> 
> But I'm more worried about readers.

Most idle_thread_get() callers happen at boottime. Then I see two
during CPU-up (cpu-hotplug) at runtime. One from _cpu_up() via
idle_thread_get() and the second via bringup_cpu() and the second in
bringup_cpu().

Sebastian