Re: Get rid of IRQF_DISABLED - (was [PATCH] genirq: warn about IRQF_SHARED|IRQF_DISABLED)

[Thomas Gleixner <tglx@linutronix.de>]

On Mon, 30 Nov 2009, Peter Zijlstra wrote:
> On Mon, 2009-11-30 at 14:54 +0100, Thomas Gleixner wrote:
> > The correct solution IMNSHO is to get rid of IRQF_DISABLED and run
> > interrupt handlers always with interrupts disabled and require them
> > not to reenable interrupts themself.
> > 
> > Thoughts ?
> 
> I'm all for removing that brain damage:
> 
>   http://lkml.org/lkml/2009/3/2/33

Darn, I knew that we discussed that before, but my memory tricked me
into believing that it was years ago :)

> We should convert the broken hardware PIO and 3com interrupt things to
> threaded interrupts, and simply mandate all IRQ handlers run short and
> with IRQs disabled.

The ide stuff seems to be the only one which uses
local_irq_enable_in_hardirq() so that at least it tripped over lockdep
already.

What's the 3com wreckage about ?

> Except I guess that will upset some of the IRQ priority folks, like
> power, where they (iirc) have a stack per irq prio level.

Is that actually used ? Ben ?
 
> But its not like the core kernel knows about these nesting rules and can
> actually track any of that muck.

True.

	tglx