On Thu, Oct 14, 2021 at 08:16:07AM -0700, Shakeel Butt wrote:
> The commit 5c1f4e690eec ("mm/vmalloc: switch to bulk allocator in
> __vmalloc_area_node()") switched to bulk page allocator for order 0
> allocation backing vmalloc. However bulk page allocator does not support
> __GFP_ACCOUNT allocations and there are several users of
> kvmalloc(__GFP_ACCOUNT).
>
> For now make __GFP_ACCOUNT allocations bypass bulk page allocator. In
> future if there is workload that can be significantly improved with the
> bulk page allocator with __GFP_ACCCOUNT support, we can revisit the
> decision.
>
> Fixes: 5c1f4e690eec ("mm/vmalloc: switch to bulk allocator in __vmalloc_area_node()")
> Signed-off-by: Shakeel Butt <shakeelb@xxxxxxxxxx>
> ---
> Changes since v1:
> - do fallback allocation instead of failure, suggested by Michal Hocko.
Acked-by: Roman Gushchin <guro@xxxxxx>
This looks indeed better! Thanks!
>
> mm/page_alloc.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 668edb16446a..9ca871dc8602 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -5230,6 +5230,10 @@ unsigned long __alloc_pages_bulk(gfp_t gfp, int preferred_nid,
> if (unlikely(page_array && nr_pages - nr_populated == 0))
> goto out;
>
> + /* Bulk allocator does not support memcg accounting. */
> + if (memcg_kmem_enabled() && (gfp & __GFP_ACCOUNT))
> + goto failed;
> +
> /* Use the single page allocator for one page. */
> if (nr_pages - nr_populated == 1)
> goto failed;
> --
> 2.33.0.882.g93a45727a2-goog
>
