Re: [patch] fix i386 mutex fastpath on FRAME_POINTER && !DEBUG_MUTEXES

[Ingo Molnar <mingo@redhat.com>]

On Wed, 11 Jan 2006, Chuck Ebbert wrote:

> In-Reply-To: <20060110210744.GA8850@elte.hu>
> 
> On Tue, 10 Jan 2006 at 22:07:44 +0100, Ingo Molnar wrote:
> 
> > --- linux.orig/include/asm-i386/mutex.h
> > +++ linux/include/asm-i386/mutex.h
> > @@ -28,7 +28,13 @@ do {                                                                       \
> >                                                                       \
> >       __asm__ __volatile__(                                           \
> >               LOCK    "   decl (%%eax)        \n"                     \
> > -                     "   js "#fail_fn"       \n"                     \
> > +                     "   js 2f               \n"                     \
> > +                     "1:                     \n"                     \
> > +                                                                     \
> > +             LOCK_SECTION_START("")                                  \
> > +                     "2: call "#fail_fn"     \n"                     \
> > +                     "   jmp 1b              \n"                     \
> > +             LOCK_SECTION_END                                        \
> >                                                                       \
> >               :"=a" (dummy)                                           \
> >               : "a" (count)                                           \
> 
> 
> But now it's inefficient again.
> 
> Why not this:
> 
>                 LOCK    "   decl (%%eax)        \n"                     \
>                         "   jns 1f              \n"                     \
>                         "   call "#fail_fn"     \n"                     \
>                         "1:                     \n"                     \
>                                                                         \
>                 :"=a" (dummy)                                           \
>                 : "a" (count)                                           \
> 
> 
> Will the extra taken forward conditional jump in the fastpath cause much
> of a slowdown?

yeah - the fastpath is much more common than the slowpath.

	Ingo
-
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/