Re: irqflags.h changes

[Ingo Molnar]

* Alexey Dobriyan <adobriyan@xxxxxxxxx> wrote:

> Currently, linux/irqflags.h defines local_irq_enable() and friends
> through raw_* counterparts.
> 
> Sometimes it defines raw_* counterparts through local_irq_enable() and
> friends.
> 
> I'm going to end this subtle joke and make on all archs:
> 
> 	local_irq_enable()	via	raw_local_irq_enable()
> 	local_irq_disable()	via	raw_local_irq_disable()
> 	local_irq_save()	via	raw_local_irq_save()
> 	local_irq_restore()	via	raw_local_irq_restore() + stuff
> 	local_save_flags()	via	raw_local_save_flags()
> 	irqs_disabled()		via	raw_irqs_disabled_flags() + stuff
> 
> linux/irqflags.h will include asm/irqflags.h unconditionally.
> Arches that don't have asm/irqflags.h will get it through extraction
> from asm/system.h or whatever.
> 
> Users will just include <linux/irqflags.h>
> 
> Irq tracing will remain the same (it's nop depending on config option,
> after all).
> 
> Typechecking will be preserved.
> 
> Compilation will be fixed.
> 
> And possibly changing some defines into static inlines.
> 
> OK?

This is a much welcome cleanup!

Since if this causes a regression anywhere that's going to be lockdep,
can we pick this up into the locking tree - or would you like to carry
it in a separate tree? (the procfs tree isnt appropriate for it i guess)

	Ingo
--
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/