Re: [PATCH bpf-next 1/2] bpf: Change error code when ops is NULL
From: Dan Carpenter
Date: Wed Apr 22 2020 - 05:34:05 EST
On Wed, Apr 22, 2020 at 04:30:09PM +0800, Mao Wenan wrote:
> There is one error printed when use type
> BPF_MAP_TYPE_SOCKMAP to create map:
> libbpf: failed to create map (name: 'sock_map'): Invalid argument(-22)
>
> This is because CONFIG_BPF_STREAM_PARSER is not set, and
> bpf_map_types[type] return invalid ops. It is not clear
> to show the cause of config missing with return code -EINVAL,
> so add pr_warn() and change error code to describe the reason.
Since you're going to have to redo the commit any way, maybe you should
put the line breaks at 72 characters. I think you're using 65 character
line breaks, but that's only for the Subject.
>
> Signed-off-by: Mao Wenan <maowenan@xxxxxxxxxx>
> ---
> kernel/bpf/syscall.c | 8 +++++---
> 1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
> index d85f37239540..f67bc063bf75 100644
> --- a/kernel/bpf/syscall.c
> +++ b/kernel/bpf/syscall.c
> @@ -112,9 +112,11 @@ static struct bpf_map *find_and_alloc_map(union bpf_attr *attr)
> return ERR_PTR(-EINVAL);
> type = array_index_nospec(type, ARRAY_SIZE(bpf_map_types));
> ops = bpf_map_types[type];
> - if (!ops)
> - return ERR_PTR(-EINVAL);
> -
> + if (!ops) {
> + pr_warn("map type %d not supported or
> + kernel config not opened\n", type);
This pr_warn() will be badly formatted in dmesg because of the new line
and the tabs. I tried to add a checkpatch.pl warning for this but maybe
it only works with -f?...
regards,
dan carpenter