Re: [PATCH net] bpf: cpumap: use GFP_KERNEL instead of GFP_ATOMIC in __cpu_map_entry_alloc()

[Michal Hocko]

On Wed 14-02-18 22:17:34, Jason Wang wrote:
> There're several implications after commit 0bf7800f1799 ("ptr_ring:
> try vmalloc() when kmalloc() fails") with the using of vmalloc() since
> can't allow GFP_ATOMIC but mandate GFP_KERNEL. This will lead a WARN
> since cpumap try to call with GFP_ATOMIC. Fortunately, entry
> allocation of cpumap can only be done through syscall path which means
> GFP_ATOMIC is not necessary, so fixing this by replacing GFP_ATOMIC
> with GFP_KERNEL.

map_update_elem does the following. Unless I am missing something and
the callback doesn't call cpu_map_update_elem there then we are in a
non-preemptible context there and GFP_WAIT would blow up.
		rcu_read_lock();
		err = map->ops->map_update_elem(map, key, value, attr->flags);
		rcu_read_unlock();

> Reported-by: syzbot+1a240cdb1f4cc88819df@xxxxxxxxxxxxxxxxxxxxxxxxx
> Fixes: 0bf7800f1799 ("ptr_ring: try vmalloc() when kmalloc() fails")
> Cc: Michal Hocko <mhocko@xxxxxxxxxx>
> Cc: Daniel Borkmann <daniel@xxxxxxxxxxxxx>
> Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
> Cc: Jesper Dangaard Brouer <brouer@xxxxxxxxxx>
> Cc: akpm@xxxxxxxxxxxxxxxxxxxx
> Cc: dhowells@xxxxxxxxxx
> Cc: hannes@xxxxxxxxxxx
> Signed-off-by: Jason Wang <jasowang@xxxxxxxxxx>
> ---
>  kernel/bpf/cpumap.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
> index fbfdada6..a4bb0b3 100644
> --- a/kernel/bpf/cpumap.c
> +++ b/kernel/bpf/cpumap.c
> @@ -334,7 +334,7 @@ static int cpu_map_kthread_run(void *data)
>  static struct bpf_cpu_map_entry *__cpu_map_entry_alloc(u32 qsize, u32 cpu,
>  						       int map_id)
>  {
> -	gfp_t gfp = GFP_ATOMIC|__GFP_NOWARN;
> +	gfp_t gfp = GFP_KERNEL | __GFP_NOWARN;
>  	struct bpf_cpu_map_entry *rcpu;
>  	int numa, err;
>  
> -- 
> 2.7.4

-- 
Michal Hocko
SUSE Labs