2008-02-15 14:10:56

by Hennerich, Michael

[permalink] [raw]
Subject: linux/kernel/irq/chip.c IRQ disable, shutdown bug

free_irq() does not disable/mask the irq, in case disable or shutdown in struct irq_chip is left uninitilazied.

/**
* struct irq_chip - hardware interrupt chip descriptor
*
* @name: name for /proc/interrupts
* @startup: start up the interrupt (defaults to ->enable if NULL)
* @shutdown: shut down the interrupt (defaults to ->disable if NULL)
* @enable: enable the interrupt (defaults to chip->unmask if NULL)
* @disable: disable the interrupt (defaults to chip->mask if NULL)


According to linux/irq.h struct irq_chip information,
chip->disable should default to chip->mask if NULL.
However irq_chip_set_defaults(struct irq_chip *chip) will set it to default_disable a empty function.


Looking through various architectures, it's pretty common that disable and shutdown is NULL.
So this bug affects many architectures.

This patch fixes the issue.

Best regards,
Michael

Index: irq/chip.c
===================================================================
--- irq/chip.c (revision 4276)
+++ irq/chip.c (working copy)
@@ -233,6 +233,10 @@
*/
static void default_disable(unsigned int irq)
{
+ struct irq_desc *desc = irq_desc + irq;
+
+ desc->chip->mask(irq);
+ desc->status |= IRQ_MASKED;
}

/*


------------------------------------------------------------------
********* Analog Devices GmbH [email protected]
** ***** Systems Engineering
** ** Wilhelm-Wagenfeld-Strasse 6
** ***** D-80807 Munich
********* Germany
Registergericht M?nchen HRB 40368, Gesch?ftsf?hrer: Thomas Wessel, Vincent Roche, Joseph E. McDonough


2008-02-23 08:25:17

by Andrew Morton

[permalink] [raw]
Subject: Re: linux/kernel/irq/chip.c IRQ disable, shutdown bug

On Fri, 15 Feb 2008 14:10:18 -0000 "Hennerich, Michael" <[email protected]> wrote:

> free_irq() does not disable/mask the irq, in case disable or shutdown in struct irq_chip is left uninitilazied.
>
> /**
> * struct irq_chip - hardware interrupt chip descriptor
> *
> * @name: name for /proc/interrupts
> * @startup: start up the interrupt (defaults to ->enable if NULL)
> * @shutdown: shut down the interrupt (defaults to ->disable if NULL)
> * @enable: enable the interrupt (defaults to chip->unmask if NULL)
> * @disable: disable the interrupt (defaults to chip->mask if NULL)
>
>
> According to linux/irq.h struct irq_chip information,
> chip->disable should default to chip->mask if NULL.
> However irq_chip_set_defaults(struct irq_chip *chip) will set it to default_disable a empty function.
>
>
> Looking through various architectures, it's pretty common that disable and shutdown is NULL.
> So this bug affects many architectures.
>
> This patch fixes the issue.
>

Please send a Signed-off-by: for this work, as per
Documentation/SubmittingPatches, thanks.


> Index: irq/chip.c
> ===================================================================
> --- irq/chip.c (revision 4276)
> +++ irq/chip.c (working copy)

Please prepare future patches in `patch -p1' form. This should have been

> --- a/kernel/irq/chip.c
> +++ a/kernel/irq/chip.c


> @@ -233,6 +233,10 @@
> */
> static void default_disable(unsigned int irq)
> {
> + struct irq_desc *desc = irq_desc + irq;
> +
> + desc->chip->mask(irq);
> + desc->status |= IRQ_MASKED;
> }
>
> /*

Ingo, Thomas: ack?

2008-02-23 10:13:36

by Ingo Molnar

[permalink] [raw]
Subject: Re: linux/kernel/irq/chip.c IRQ disable, shutdown bug


* Andrew Morton <[email protected]> wrote:

> > + desc->chip->mask(irq);
> > + desc->status |= IRQ_MASKED;
> > }
> >
> > /*
>
> Ingo, Thomas: ack?

we've already queued it in up the genirq tree. (in a slightly different
form)

Ingo

2008-02-23 17:52:25

by Thomas Gleixner

[permalink] [raw]
Subject: Re: linux/kernel/irq/chip.c IRQ disable, shutdown bug

On Sat, 23 Feb 2008, Ingo Molnar wrote:

> * Andrew Morton <[email protected]> wrote:
>
> > > + desc->chip->mask(irq);
> > > + desc->status |= IRQ_MASKED;
> > > }
> > >
> > > /*
> >
> > Ingo, Thomas: ack?
>
> we've already queued it in up the genirq tree. (in a slightly different
> form)

Already mainline: 89d694b9dbe769ca1004e01db0ca43964806a611

Thanks,

tglx