Re: [PATCH] x86: unify current 3 similar ways of saving IRQ info

[Thomas Gleixner]

On Mon, 17 May 2010, Feng Tang wrote:
> On Mon, 17 May 2010 16:16:40 +0800
> Thomas Gleixner <tglx@xxxxxxxxxxxxx> wrote:
> 
> > On Mon, 17 May 2010, Feng Tang wrote:
> > 
> > > On Mon, 17 May 2010 16:07:05 +0800
> > > Thomas Gleixner <tglx@xxxxxxxxxxxxx> wrote:
> > > 
> > > > On Mon, 17 May 2010, Feng Tang wrote:
> > > > 
> > > > >  {
> > > > >  	if (mp_irq->dstapic != m->dstapic)
> > > > > @@ -211,7 +211,8 @@ static int __init
> > > > > mp_irq_mpc_intsrc_cmp(struct mpc_intsrc *mp_irq, return 0;
> > > > >  }
> > > > >  
> > > > > -static void __init MP_intsrc_info(struct mpc_intsrc *m)
> > > > > +/* Will also be called in acpi/sfi related code */
> > > > > +void mp_save_irq(struct mpc_intsrc *m)
> > > > 
> > > >   Again, this code is called from __init context only and should
> > > > stay __init.
> > > 
> > > Hi Thomas,
> > > 
> > > The reason I removed the __init is it will be called by non __init
> > > functions from acpi/boot.c and mrst.c
> > 
> > Why should that happen ?
> 
> The place calling this API from mrst.c should be changed to __init, for acpi,
> it will be called by acpi_register_gsi(), which is exported out and used by
> several places in drivers/

Ok, missed the acpi_register_gsi() context.
 
Thanks,

	tglx
--
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/