Re: [WARNING] at kernel/irq/manage.c:421 __enable_irq

[Steven Rostedt <rostedt@goodmis.org>]

On Wed, 2012-05-16 at 13:27 +0200, Thomas Gleixner wrote:
> On Tue, 15 May 2012, Steven Rostedt wrote:
> > The code in ide_probe_port() does the following:
> 
> Right, you just forgot to show the most interesting gem of that code:
> 
> 	/*
> 	 * We must always disable IRQ, as probe_for_drive will assert IRQ, but
> 	 * we'll install our IRQ driver much later...
> 	 */

Heh, I would have kept that, but I was more interested in cut and
pasting code than comments. I'll make sure to keep comments in emails
too.


> >         if (irqd)
> >                 enable_irq(irqd);
> 
> So this code is wrong and unsafe in various aspects. I wonder why it
> didn't blow up before. Maybe init ordering changed ....
> 

Or it always did. This is on my main box that I reboot once every 100
days or so. I'm not usually present when the machine comes online, nor
do I search dmesg for warnings. Although, yesterday I did. This could
have been complaining from day one.


> I have an idea how to fix that proper. Will send out patches later.

I'll test them, but it's a bit of a pain, as it would require me
shutting down evolution, xchat, firefox, chrome and all my emacs
windows :-)  This is why I don't reboot this box often.

-- Steve