Re: [cpuops cmpxchg double V3 4/5] Lockless (and preemptless)fastpaths for slub

[Christoph Lameter]

On Fri, 25 Feb 2011, Mathieu Desnoyers wrote:

> > Index: linux-2.6/include/linux/slub_def.h
> > ===================================================================
> > --- linux-2.6.orig/include/linux/slub_def.h	2011-02-25 10:45:49.000000000 -0600
> > +++ linux-2.6/include/linux/slub_def.h	2011-02-25 10:46:19.000000000 -0600
> > @@ -35,7 +35,10 @@ enum stat_item {
> >  	NR_SLUB_STAT_ITEMS };
> >
> >  struct kmem_cache_cpu {
> > -	void **freelist;	/* Pointer to first free per cpu object */
> > +	void **freelist;	/* Pointer to next available object */
> > +#ifdef CONFIG_CMPXCHG_LOCAL
> > +	unsigned long tid;	/* Globally unique transaction id */
> > +#endif
>
> There seem to be no strong guarantee that freelist is double-word aligned here.

The struct kmem_cache_cpu allocation via alloc_percpu() specifies double
word alignment. See the remainder of the code quoted by you:

> > +#ifdef CONFIG_CMPXCHG_LOCAL
> > +	/*
> > +	 * Must align to double word boundary for the double cmpxchg instructions
> > +	 * to work.
> > +	 */
> > +	s->cpu_slab = __alloc_percpu(sizeof(struct kmem_cache_cpu), 2 * sizeof(void *));
> > +#else
> > +	/* Regular alignment is sufficient */
> >  	s->cpu_slab = alloc_percpu(struct kmem_cache_cpu);
> > +#endif
> > +
> > +	if (!s->cpu_slab)
> > +		return 0;
> >
> > -	return s->cpu_slab != NULL;
> > +	init_kmem_cache_cpus(s);
> > +
> > +	return 1;
> >  }
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/