On 10/2/18 4:04 AM, Jiri Benc wrote:
> On Fri, 28 Sep 2018 12:28:41 -0700, David Ahern wrote:
>> --- a/net/core/rtnetlink.c
>> +++ b/net/core/rtnetlink.c
>> @@ -1898,10 +1898,8 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb,
>> struct netlink_callback *cb)
>> if (tb[IFLA_IF_NETNSID]) {
>> netnsid = nla_get_s32(tb[IFLA_IF_NETNSID]);
>> tgt_net = get_target_net(skb->sk, netnsid);
>> - if (IS_ERR(tgt_net)) {
>> - tgt_net = net;
>> - netnsid = -1;
>> - }
>> + if (IS_ERR(tgt_net))
>> + return PTR_ERR(tgt_net);
>> }
>>
>> if (tb[IFLA_EXT_MASK])
>
> Sorry for the late review, I see it has been applied.
>
> I intentionally chose the behavior to preserve the behavior of the
> older kernels: that attribute was silently ignored. Note that the
> IFLA_IF_NETNSID is not returned in such case, thus it's easy to
> distinguish that it was not applied. And the user space has to do such
> check anyway to support old kernels.
First, rtnl_dellink, rtnl_newlink and rtnl_getlink all fail if net
namespace id is invalid. Second, the user is requesting data from a
target namespace and the dump happily continued with the current
namespace which is not what the user requested. Hence, it makes no sense
for a dump to continue which is why I sent the patch.
>
> But you're right that there was no way to distinguish "the kernel does
> not support IFLA_IF_NETNSID" from "wrong IFLA_IF_NETNSID provided". I'm
> okay with the patch, I just don't think the "Fixes" tag is justified but
> whatever, can't be unapplied :-) (and it's my fault for not reviewing
> the patches timely).
The id was the commit that added the code to ignore the error.
