Re: [bpf-next] bpf: use kvmemdup_bpfptr helper

[sdf@google.com]

On 09/09, Wang Yufen wrote:
> Use kvmemdup_bpfptr helper instead of open-coding to
> simplify the code.

> Signed-off-by: Wang Yufen <wangyufen@huawei.com>
> ---
>   kernel/bpf/syscall.c | 14 ++++----------
>   1 file changed, 4 insertions(+), 10 deletions(-)

> diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
> index 4e9d4622aef7..13ce28081982 100644
> --- a/kernel/bpf/syscall.c
> +++ b/kernel/bpf/syscall.c
> @@ -1413,20 +1413,14 @@ static int map_update_elem(union bpf_attr *attr,  
> bpfptr_t uattr)
>   	}

>   	value_size = bpf_map_value_size(map);
> -
> -	err = -ENOMEM;
> -	value = kvmalloc(value_size, GFP_USER | __GFP_NOWARN);
> -	if (!value)
> +	value = kvmemdup_bpfptr(uvalue, value_size);
> +	if (IS_ERR(value)) {
> +		err = PTR_ERR(value);
>   		goto free_key;
> -
> -	err = -EFAULT;
> -	if (copy_from_bpfptr(value, uvalue, value_size) != 0)
> -		goto free_value;
> +	}

>   	err = bpf_map_update_value(map, f, key, value, attr->flags);


[..]

> -free_value:
> -	kvfree(value);

And here you leak the value. We need to free it after update regardless
of error/success. That's why it is coded like that.

>   free_key:
>   	kvfree(key);
>   err_put:
> --
> 2.25.1