Re: [PATCH 2/3] HPET: Enter hpet_interrupt_handler with interrupts disabled

[Takashi Iwai]

At Sun, 2 Nov 2008 22:23:13 +0000,
Matt Fleming wrote:
> 
> 2008/11/2 Will Newton <will.newton@xxxxxxxxx>:
> >
> > Combining IRQF_DISABLED and IRQF_SHARED does not reliably disable
> > interrupts in the handler. Perhaps IRQF_SHARED should be removed at
> > the same time?
> >
> 
> I didn't know that. Under what conditions is it unreliable?

The kernel checks IRQF_DISABLED bit of only the first handler
even if multiple handlers are assigned to a single IRQ line.
Thus, if another handler without IRQF_DISABLED has been already
registered before yours, your IRQF_DISABLED is ignored.

See kernel/irq/handle.c.


Takashi
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/