[PATCH v4 3/5] irqchip: Add DT binding doc for the virtual irq demuxer chip

[Rafael J. Wysocki]

On Monday, February 16, 2015 12:23:43 PM Mark Rutland wrote:
> [...]
> 
> > > The "suspend" part is kind of a distraction to me here, because that really
> > > only is about sharing an IRQ with a timer and the "your interrupt handler
> > > may be called when the device is suspended" part is just a consequence of that.
> > > 
> > > So IMO it's better to have "TIMER" in the names to avoid encouraging people to
> > > abuse this for other purposes not related to timers.
> > 
> > Sorry to be late to the bike-shed party, but what about:
> 
> [...]
> 
> > arch/arm/mach-omap2/mux.c:		omap_hwmod_mux_handle_irq, IRQF_SHARED | IRQF_NO_SUSPEND,
> > arch/arm/mach-omap2/pm34xx.c:		_prcm_int_handle_io, IRQF_SHARED | IRQF_NO_SUSPEND, "pm_io",
> > drivers/pinctrl/pinctrl-single.c:				  IRQF_SHARED | IRQF_NO_SUSPEND,
> 
> These are chained IRQ handlers. If any of these have a chained timer irq
> then the IRQF_NO_SUSPEND may be legitimate. I can't imagine why these
> would be shared, however.
> 
> It also looks like these abuse IRQF_NO_SUSPEND for wakeup interrupts.
> 
> > drivers/rtc/rtc-pl031.c:	.irqflags = IRQF_SHARED | IRQF_NO_SUSPEND,
> 
> This looks to be an abuse and should use {enable,disable}_irq_wake.
> 
> However, we'd then need to handle mismatch with wakeup interrupts (which
> is effectively the same problem as sharing with a timer).

IRQF_NO_SUSPEND and wakeup fundamentally don't match due to the way
wakeup is implemented in the IRQ core now.

Unless drivers with IRQF_NO_SUSPEND do the wakeup behind the core's back
which is just disgusting and should never happen.


-- 
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.