linux-kernel - Re: [PATCH] irqchip/i8259: fix shutdown order by moving syscore

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:   Thu, 7 Feb 2019 22:58:12 +0200
From:   Aaro Koskinen <aaro.koskinen@....fi>
To:     Marc Zyngier <marc.zyngier@....com>
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        Jason Cooper <jason@...edaemon.net>,
        linux-kernel@...r.kernel.org, tomli@...li.me,
        "Rafael J. Wysocki" <rafael@...nel.org>,
        Ralf Baechle <ralf@...ux-mips.org>,
        Paul Burton <paul.burton@...s.com>,
        James Hogan <jhogan@...nel.org>
Subject: Re: [PATCH] irqchip/i8259: fix shutdown order by moving syscore_ops
 registration

Hi,

On Thu, Feb 07, 2019 at 08:56:37AM +0000, Marc Zyngier wrote:
> On 06/02/2019 21:26, Aaro Koskinen wrote:
> >  static void init_8259A(int auto_eoi)
> >  {
> >  	unsigned long flags;
> > @@ -332,6 +324,7 @@ struct irq_domain * __init __init_i8259_irqs(struct device_node *node)
> >  		panic("Failed to add i8259 IRQ domain");
> >  
> >  	setup_irq(I8259A_IRQ_BASE + PIC_CASCADE_IR, &irq2);
> > +	register_syscore_ops(&i8259_syscore_ops);
> >  	return domain;
> >  }
> >  
> > 
> 
> Given that this is a change of behaviour that is likely to affect other
> platforms (I see at least another 6 MIPS machines using the i8259),
> could someone make sure that this doesn't cause any regression? This is
> unlikely to affect the SGI boxes, as they predate any notion of power
> management, but something like Malta could potentially be affected.

For shutdown, I don't think there are many syscore_ops users on these
platforms. Actually I could find only two that I think could be used:
	- cpufreq (issue fixed by this patch, and Loongson is the only user
	  anyway)
	- leds-trigger

Then suspend/resume: i8259 doesn't implement suspend, so there is no
change in behaviour. In resume it does PIC re-init, but syscore_resume()
is done with interrupts disabled so the order shouldn't matter.

A.