Re: [PATCH] [V2] infiniband: fix a missing check of nla_put

[Leon Romanovsky]

On Sat, Jan 05, 2019 at 10:18:07AM -0600, Aditya Pakki wrote:
> nla_put() may fail. The fix adds a check for its return value, and
> returns -EMSGSIZE if it fails, post canceling netlink msg.
>
> Signed-off-by: Aditya Pakki <pakki001@xxxxxxx>
> ---
>  drivers/infiniband/core/addr.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c
> index 0dce94e3c495..e847b34bacd1 100644
> --- a/drivers/infiniband/core/addr.c
> +++ b/drivers/infiniband/core/addr.c
> @@ -178,7 +178,10 @@ static int ib_nl_ip_send_msg(struct rdma_dev_addr *dev_addr,
>  	/* Construct the family header first */
>  	header = skb_put(skb, NLMSG_ALIGN(sizeof(*header)));
>  	header->ifindex = dev_addr->bound_dev_if;
> -	nla_put(skb, attrtype, size, daddr);
> +	if (nla_put(skb, attrtype, size, daddr)) {
> +		nlmsg_cancel(skb, nlh);

It should be nlmsg_free() and not nlmsg_cancel()

> +		return -EMSGSIZE;
> +	}
>
>  	/* Repair the nlmsg header length */
>  	nlmsg_end(skb, nlh);
> --
> 2.17.1
>

Attachment: signature.asc
Description: PGP signature