Re: [PATCH Resend] - SN: validate smp_affinity mask on intr

[Andrew Morton <akpm@linux-foundation.org>]

On Tue, 8 May 2007 13:14:26 -0700
"Luck, Tony" <tony.luck@intel.com> wrote:

> > It had a dopey little bug:
> >
> > -#define is_affinity_mask_valid() 1
> > +#define is_affinity_mask_valid(val) 1
> 
> That would fix warnings on non-ia64 systems (which is
> a step in the right direction).  But on ia64 I have
> the
> 
> #define is_affinity_mask_valid is_affinity_mask_valid
> 
> in play at that point, so I have a real function call
> which doesn't have an in-scope declaration, so I get:
> 
> kernel/irq/proc.c:49: warning: implicit declaration of function `is_affinity_mask_valid'
> 

hm, I wonder if John sent the right patch.

The obvious fix fixes it up for me:

--- a/include/asm-ia64/irq.h~sn-validate-smp_affinity-mask-on-intr-redirect-fix-2
+++ a/include/asm-ia64/irq.h
@@ -11,6 +11,9 @@
  * 02/29/00     D.Mosberger	moved most things into hw_irq.h
  */
 
+#include <linux/types.h>
+#include <linux/cpumask.h>
+
 #define NR_IRQS		256
 #define NR_IRQ_VECTORS	NR_IRQS
 
@@ -29,6 +32,7 @@ extern void disable_irq (unsigned int);
 extern void disable_irq_nosync (unsigned int);
 extern void enable_irq (unsigned int);
 extern void set_irq_affinity_info (unsigned int irq, int dest, int redir);
+bool is_affinity_mask_valid(cpumask_t cpumask);
 
 #define is_affinity_mask_valid is_affinity_mask_valid
 
_

but the new includes are always a worry.

We need the cpumask.h include because we went and created a stupid typedef.
 I think at the time we did this because cpumask_t could optionally be a
plain old "long".  However I don't think we ended up doing that so we could
now do:

-typedef struct { DECLARE_BITMAP(bits, NR_CPUS); } cpumask_t;
+typedef struct cpumask { DECLARE_BITMAP(bits, NR_CPUS); } cpumask_t;

and then we can forward-declare `struct cpumask' in include/asm-ia64/irq.h
rather than adding the nested include.