[2.6 patch] include/asm-mips/processor.h: "extern inline" -> "static inline"

[2.6 patch] include/asm-mips/processor.h: "extern inline" -> "static inline"

[Adrian Bunk]

"extern inline" will have different semantics with gcc 4.3,
and "static inline" is correct here.

Signed-off-by: Adrian Bunk <bunk@stusta.de>

---
--- linux-2.6.22-rc6-mm1/include/asm-mips/processor.h.old	2007-07-07 01:06:35.000000000 +0200
+++ linux-2.6.22-rc6-mm1/include/asm-mips/processor.h	2007-07-07 01:06:57.000000000 +0200
@@ -237,7 +237,7 @@
 
 #define ARCH_HAS_PREFETCH
 
-extern inline void prefetch(const void *addr)
+static inline void prefetch(const void *addr)
 {
 	__asm__ __volatile__(
 	"	.set	mips4		\n"

Re: [2.6 patch] include/asm-mips/processor.h: "extern inline" -> "static inline"

[Ralf Baechle]

On Sat, Jul 07, 2007 at 03:03:30AM +0200, Adrian Bunk wrote:

> "extern inline" will have different semantics with gcc 4.3,
> and "static inline" is correct here.

The idea was to have a linker error in case gcc should deciede for some
reason not to inline this function which as I understand will continue
to be the behaviour of gcc 4.3?

  Ralf

Re: [2.6 patch] include/asm-mips/processor.h: "extern inline" -> "static inline"

[Franck Bui-Huu]

Hi,

On 7/9/07, Ralf Baechle <ralf@linux-mips.org> wrote:
> On Sat, Jul 07, 2007 at 03:03:30AM +0200, Adrian Bunk wrote:
>
> > "extern inline" will have different semantics with gcc 4.3,
> > and "static inline" is correct here.
>
> The idea was to have a linker error in case gcc should deciede for some
> reason not to inline this function which as I understand will continue
> to be the behaviour of gcc 4.3?
>

I don't know for this peculiar case but it usually is a good thing to
let gcc decide if the function needs to be inlined or not.

If we really want this function to be inlined in all cases, maybe we
should use __always_inline attribute instead ?

-- 
               Franck

Re: [2.6 patch] include/asm-mips/processor.h: "extern inline" -> "static inline"

[Adrian Bunk]

On Mon, Jul 09, 2007 at 11:27:55AM +0100, Ralf Baechle wrote:
> On Sat, Jul 07, 2007 at 03:03:30AM +0200, Adrian Bunk wrote:
> 
> > "extern inline" will have different semantics with gcc 4.3,
> > and "static inline" is correct here.
> 
> The idea was to have a linker error in case gcc should deciede for some
> reason not to inline this function

If that's the intention, please use __always_inline instead.

> which as I understand will continue
> to be the behaviour of gcc 4.3?

In C99 (and therefore in gcc >= 4.3), "extern inline" means that the 
function should be compiled inline where the inline definition is seen, 
and that the compiler should also emit a copy of the function body with 
an externally visible symbol.

You don't want this.

>   Ralf

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: [2.6 patch] include/asm-mips/processor.h: "extern inline" -> "static inline"

[Ralf Baechle]

On Mon, Jul 09, 2007 at 02:22:40PM +0200, Adrian Bunk wrote:

> In C99 (and therefore in gcc >= 4.3), "extern inline" means that the 
> function should be compiled inline where the inline definition is seen, 
> and that the compiler should also emit a copy of the function body with 
> an externally visible symbol.

Uh, indeed.  Thanks for the reminder.  Patch applied.

  Ralf