Jan Kiszka wrote:
> According to Linux and the Intel spec, this prefix is not needed.
>

Obviously, it's not, since the whole purpose of xchg() is to guarantee bus
locking for memory operands anyway. Please merge.

> ---
>  include/asm-x86/atomic_32.h |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> Index: b/include/asm-x86/atomic_32.h
> ===================================================================
> --- a/include/asm-x86/atomic_32.h
> +++ b/include/asm-x86/atomic_32.h
> @@ -64,7 +64,7 @@ struct __xeno_xchg_dummy { unsigned long
>  static inline unsigned long xnarch_atomic_xchg (volatile void *ptr,
>                                               unsigned long x)
>  {
> -     __asm__ __volatile__(LOCK_PREFIX "xchgl %0,%1"
> +     __asm__ __volatile__("xchgl %0,%1"
>                            :"=r" (x)
>                            :"m" (*__xeno_xg(ptr)), "0" (x)
>                            :"memory");
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Xenomai-core mailing list
> Xenomai-core@gna.org
> https://mail.gna.org/listinfo/xenomai-core


-- 
Philippe.

_______________________________________________
Xenomai-core mailing list
Xenomai-core@gna.org
https://mail.gna.org/listinfo/xenomai-core

Reply via email to