Re: [PATCH 8/8] Inline 3 functions

[Adrian Bunk <bunk@stusta.de>]

On Thu, Nov 10, 2005 at 08:22:52PM +0100, Oliver Neukum wrote:
> Am Donnerstag, 10. November 2005 19:20 schrieb Adrian Bunk:
> > On Thu, Nov 10, 2005 at 07:04:22PM +0100, Oliver Neukum wrote:
> > > Am Donnerstag, 10. November 2005 18:38 schrieb Adrian Bunk:
> > > > > So are you suggesting that we don't mark these functions 'inline', or are
> > > > > you just pointing out that we'll need to drop the 'inline' if there is ever
> > > > > another caller?
> > > > 
> > > > I'd suggest to not mark them 'inline'.
> > > 
> > > It seems you have found one more use for sparse. How about a tag
> > > like __single_inline that will cause a warning if a function having it
> > > is called from more than one place?
> > 
> > Why should such a function be manually marked "inline" at all?
> > 
> > If a static function is called exactly once it is the job of the 
> > compiler to inline the function.
> 
> It should indeed. This documentation says it does:
> http://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html
> That makes me wonder what is the problem.

On i386, we have the problem that we are using -fno-unit-at-a-time to 
avoid stack usage problems.

But the proper solution will be to remove -fno-unit-at-a-time from the 
CFLAGS for gcc >= 4.1 or >= 4.2 and check whether this will cause any 
new stack usage problems.

> 	Puzzeled
> 		Oliver

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

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/