Re: [PATCH] take 2 of the tr-based current

[Jakub Jelinek <jakub@redhat.com>]

On Thu, Nov 08, 2001 at 09:11:43PM -0500, Benjamin LaHaise wrote:
> On Thu, Nov 08, 2001 at 07:05:46PM -0500, Benjamin LaHaise wrote:
> > If other people could bang on this a bit and post any problems, I'd 
> > appreciate it.  tia,
> 
> Ooops.  A slight typo is fixed below.
> 
> 		-ben (who shouldn't edit patches before hitting send)
> 
> 
> diff -ur ./v2.4.13-ac8/include/asm-i386/smp.h ../toomuch-v2.4.13-ac8+tr/include/asm-i386/smp.h
> --- ./v2.4.13-ac8/include/asm-i386/smp.h	Thu Nov  8 21:07:47 2001
> +++ toomuch-v2.4.13-ac8+tr/include/asm-i386/smp.h	Thu Nov  8 21:06:25 2001
> @@ -102,7 +102,8 @@
>   * so this is correct in the x86 case.
>   */
>  
> -static unsigned get_TR(void) __attribute__ ((pure))
> +static unsigned get_TR(void) __attribute__ ((pure));
> +static unsigned get_TR(void)
>  {
>  	unsigned tr;
>  	__asm__("str %w0" : "=g" (tr));

Why not
static inline unsigned __attribute__ ((const)) get_TR(void)
{
}
?
If TR register only ever changes during cpu_init, I don't see why you
cannot use const. Using pure would mean if you do get_TR, then store
something into global memory and do get_TR again, it will be done twice.
Also, I wonder why you don't inline it.

	Jakub