On Thu, 3 Dec 2015, Jesper Dangaard Brouer wrote:
> + size_t i;
> +
> + flags &= gfp_allowed_mask;
> + lockdep_trace_alloc(flags);
> +
> + if (slab_should_failslab(s, flags))
> + return 0;
Ok here is an overlap with slub;'s pre_alloc_hook() and that stuff is
really not allocator specific. Could make it generic and move the hook
calls into slab_common.c/slab.h? That also gives you the opportunity to
get the array option in there.
> + s = memcg_kmem_get_cache(s, flags);
> +
> + cache_alloc_debugcheck_before(s, flags);
> +
> + local_irq_disable();
> + for (i = 0; i < size; i++) {
> + void *objp = __do_cache_alloc(s, flags);
> +
> + // this call could be done outside IRQ disabled section
> + objp = cache_alloc_debugcheck_after(s, flags, objp, _RET_IP_);
> +
> + if (unlikely(!objp))
> + goto error;
> +
> + prefetchw(objp);
Is the prefetch really useful here? Only if these objects are immediately
used I would think.
> + p[i] = objp;
> + }
> + local_irq_enable();
> +
> + /* Kmemleak and kmemcheck outside IRQ disabled section */
> + for (i = 0; i < size; i++) {
> + void *x = p[i];
> +
> + kmemleak_alloc_recursive(x, s->object_size, 1, s->flags, flags);
> + kmemcheck_slab_alloc(s, flags, x, s->object_size);
> + }
> +
> + /* Clear memory outside IRQ disabled section */
> + if (unlikely(flags & __GFP_ZERO))
> + for (i = 0; i < size; i++)
> + memset(p[i], 0, s->object_size);
Maybe make this one loop instead of two?
> +// FIXME: Trace call missing... should we create a bulk variant?
> +/* Like:
> + trace_kmem_cache_alloc(_RET_IP_, ret, s->object_size, s->size, flags);
> +*/
That trace call could be created when you do the genericization of the
hooks() which also involve debugging stuff.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>