Re: [S+Q Cleanup 2/6] slub: remove dynamic dma slab allocation

[Christoph Lameter <cl@linux-foundation.org>]

On Tue, 17 Aug 2010, David Rientjes wrote:

> > -	page = new_slab(kmalloc_caches, gfpflags, node);
> > +	page = new_slab(kmalloc_caches, GFP_KERNEL, node);
> >
> >  	BUG_ON(!page);
> >  	if (page_to_nid(page) != node) {
>
> early_kmem_cache_node_alloc() is called when we don't have a
> gfp_allowed_mask, so this is actually GFP_NOWAIT.

The page allocator will do the conversion anyway but I will update it. We
cannot do this consistenly for code sections that can be run both at boot
time and later though.

> > +	for (i = 1; i < SLUB_PAGE_SHIFT; i++) {
> > +		struct kmem_cache *s = &kmalloc_caches[i];
> > +
> > +		if (s->size) {
> > +			char *name = kasprintf(GFP_KERNEL,
> > +				 "dma-kmalloc-%d", s->objsize);
>
> Same for this, it's GFP_NOWAIT.
>
> There's no actual bug with either of those since the bits get masked off,
> but the code is clearer if the allocation context is known to be during
> early boot.

Ok.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>