2015-06-29 15:52:23

by Michal Hocko

[permalink] [raw]
Subject: Re: [PATCH] mm:Return proper error code return if call to kzalloc_node falis in the function alloc_mem_cgroup_per_zone_info

On Mon 29-06-15 11:46:53, Nicholas Krause wrote:
> This changes us returning the value of one to -ENOMEM when the call
> for allocating memory with the function kzalloc_node fails in order
> to better comply with kernel coding pratices of returning this
> particular error code when memory allocations that are unrecoverable
> occur.

I do not see any point in such a patch. Let me repeat, and hopefully for
the last time, the patch has to make _sense_ and the changelog should
provide a _justification_ for the change. None of this is true for this
patch.

> Signed-off-by: Nicholas Krause <[email protected]>

I am not interested in changes like this in the code I maintain.

> ---
> mm/memcontrol.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index acb93c5..4e80811 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -4442,7 +4442,7 @@ static int alloc_mem_cgroup_per_zone_info(struct mem_cgroup *memcg, int node)
> tmp = -1;
> pn = kzalloc_node(sizeof(*pn), GFP_KERNEL, tmp);
> if (!pn)
> - return 1;
> + return -ENOMEM;
>
> for (zone = 0; zone < MAX_NR_ZONES; zone++) {
> mz = &pn->zoneinfo[zone];
> --
> 2.1.4
>

--
Michal Hocko
SUSE Labs