Re: [PATCH 2/3] Add hard_irq_disable()

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

> So you're saying that this mechanism forces the arch (that really
> wants hard_irq_disable) to _#define_ hard_irq_disable (as a macro),
> and if it implements it as an inline function, for example, then we're
> screwed?

No. The idea is to do like we did for a few other things already
(according to Linus request in fact), which is to write

static inline void hard_irq_disable(void)
{
	.../...
}
#define hard_irq_disable hard_irq_disable

This is nicer than having an ARCH_HAS_xxx

> 1. Introduce some CONFIG_WANTS_HARD_IRQ_DISABLE that is #defined (or
> left undefined) by the arch/.../defconfig (depending upon whether or
> not that arch implements a hard_irq_disable() for itself or not)
>
> 2. Then pull-in that code into include/linux/interrupt.h somehow
> (through some known / fixed header file, or through asm/system.h, or
> anyhow -- it doesn't really matter)
> 
> 3. And:
> 
> #ifndef CONFIG_WANTS_HARD_IRQ_DISABLE
> #define hard_irq_disable() do { } while(0)
> #endif

Well, last time I tried that, Linus NACKed it in favor of what I
described above.

> We don't need to standardize on some particular arch-specific header
> filename in this case.

True, that's my main problem here. Though really only the archs who
actually implement something special here need to be careful.

Ben.