On 12:09 Thu 23 May , Thomas Petazzoni wrote:
> Dear boris brezillon,
>
> On Thu, 23 May 2013 11:51:36 +0200, boris brezillon wrote:
> > On 23/05/2013 11:06, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > > On 11:05 Thu 23 May , Boris BREZILLON wrote:
> > >> Move arch/arm/mach-at91/irq.c to drivers/irqchip/irq-at91.c.
> > >> Move arch/arm/mach-at91/at91_aic.h to
> > >> arch/arm/mach-at91/include/mach/at91_aic.h to avoid ugly reference
> > >> to header file :
> > >> #include "../../arch/arm/mach-at91/at91_aic.h"
> > > no we are going to drop arch/arm/mach-at91/include/mach for multi support
> > >
> > > arch/arm/mach-at91/include/mach is for removal
> > What should be done ?
> >
> > include ../../arch/arm/mach-at91/at91_aic.h from irq-at91.c
> >
> > or
> >
> > move at91_aic.h into drivers/irqchips and patch every non dt board to
> > include ../../../drivers/irqchip/at91_aic.h
>
> You can put it in <linux/irqchip/at91-aic.h>, there are already a few
> headers there. But I think it's not the right solution.
>
> *However*, ideally, this header should disappear completely. All the
> register defines should go directly into the driver C file. The only
> usage of the AIC defines outside of the IRQ driver are for debug prints
> in the pm.c code, which I think you could get rid of.
>
> Ditto for the at91_aic_{read,write} macros.
>
> The NR_AIC5_IRQS can also move to the driver itself.
>
> The only remaining one would be NR_AIC_IRQS, you can't get rid of it,
> because it's used for the IRQ priority arrays. But I believe keeping
> this one in <linux/irqchip/at91-aic.h> is reasonable.
do want either as we expose the register acces for pm & co
until this is cleanup NACK
Best Regards,
J.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
