Re: [PATCH 2/3 htlb-fault] Demand faulting for huge pages

[Andrew Morton <akpm@osdl.org>]

Adam Litke <agl@us.ibm.com> wrote:
>
> +static struct page *find_get_huge_page(struct address_space *mapping,
>  +			unsigned long idx)
>  +{
>  +	struct page *page = NULL;
>  +
>  +retry:
>  +	page = find_get_page(mapping, idx);
>  +	if (page)
>  +		goto out;
>  +
>  +	if (hugetlb_get_quota(mapping))
>  +		goto out;
>  +	page = alloc_huge_page();
>  +	if (!page) {
>  +		hugetlb_put_quota(mapping);
>  +		goto out;
>  +	}
>  +
>  +	if (add_to_page_cache(page, mapping, idx, GFP_ATOMIC)) {
>  +		put_page(page);
>  +		hugetlb_put_quota(mapping);
>  +		goto retry;

If add_to_page_cache() fails due to failure in radix_tree_preload(), this
code will lock up.

A lame fix is to check for -ENOMEM and bale.  A better fix would be to use
GFP_KERNEL.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/