[patch 2/2] powerpc: optimise smp_wmb

Nick Piggin npiggin at suse.de
Thu May 22 01:34:20 EST 2008


On Wed, May 21, 2008 at 11:26:32AM -0400, Benjamin Herrenschmidt wrote:
> 
> On Wed, 2008-05-21 at 16:12 +0200, Nick Piggin wrote:
> > lwsync is the recommended method of store/store ordering on caching enabled
> > memory. For those subarchs which have lwsync, use it rather than eieio for
> > smp_wmb.
> 
> Yuck... existence of lwsync depends on the processor at boot time...

Not according to the __stringify(LWSYNC) that I just removed. At least,
presumably it is always present on 64 bit processors, and 32 bit ones
will be no worse off as they'll continue just using eieio.

> 
> Ben.
> 
> > Signed-off-by: Nick Piggin <npiggin at suse.de>
> > ---
> > Index: linux-2.6/include/asm-powerpc/system.h
> > ===================================================================
> > --- linux-2.6.orig/include/asm-powerpc/system.h
> > +++ linux-2.6/include/asm-powerpc/system.h
> > @@ -30,8 +30,8 @@
> >   *
> >   * For wmb(), we use sync since wmb is used in drivers to order
> >   * stores to system memory with respect to writes to the device.
> > - * However, smp_wmb() can be a lighter-weight eieio barrier on
> > - * SMP since it is only used to order updates to system memory.
> > + * However, smp_wmb() can be a lighter-weight lwsync or eieio barrier
> > + * on SMP since it is only used to order updates to system memory.
> >   */
> >  #define mb()   __asm__ __volatile__ ("sync" : : : "memory")
> >  #define rmb()  __asm__ __volatile__ ("sync" : : : "memory")
> > @@ -43,9 +43,16 @@
> >  #ifdef __KERNEL__
> >  #define AT_VECTOR_SIZE_ARCH 6 /* entries in ARCH_DLINFO */
> >  #ifdef CONFIG_SMP
> > +
> > +#ifdef __SUBARCH_HAS_LWSYNC
> > +#    define SMPWMB      lwsync
> > +#else
> > +#    define SMPWMB      eieio
> > +#endif
> > +
> >  #define smp_mb()	mb()
> >  #define smp_rmb()	rmb()
> > -#define smp_wmb()	eieio()
> > +#define smp_wmb()	__asm__ __volatile__ (__stringify(SMPWMB) : : :"memory")
> >  #define smp_read_barrier_depends()	read_barrier_depends()
> >  #else
> >  #define smp_mb()	barrier()
> > _______________________________________________
> > Linuxppc-dev mailing list
> > Linuxppc-dev at ozlabs.org
> > https://ozlabs.org/mailman/listinfo/linuxppc-dev



More information about the Linuxppc-dev mailing list