netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Hangbin Liu <liuhangbin@gmail.com>
To: David Ahern <dsahern@gmail.com>
Cc: netdev@vger.kernel.org, Cong Wang <xiyou.wangcong@gmail.com>,
	Roopa Prabhu <roopa@cumulusnetworks.com>
Subject: Re: [PATCHv3 net] ipv6: no need to return rt->dst.error if it is prohibit entry
Date: Thu, 27 Jul 2017 21:48:23 +0800	[thread overview]
Message-ID: <20170727134823.GD5465@leo.usersys.redhat.com> (raw)
In-Reply-To: <07c66e6f-5b78-3317-18c6-bd2f955d5f90@gmail.com>

On Wed, Jul 26, 2017 at 11:09:39AM -0600, David Ahern wrote:
> On 7/26/17 3:20 AM, Hangbin Liu wrote:
> > After commit 18c3a61c4264 ("net: ipv6: RTM_GETROUTE: return matched fib
> > result when requested"). When we get a prohibit ertry, we will return
> > -EACCES directly.
> > 
> > Before:
> 
> Do you mean "Before commit 18c3a61c4264?"
> 
> > + ip netns exec client ip -6 route get 2003::1
> > prohibit 2003::1 dev lo table unspec proto kernel src 2001::1 metric
> > 4294967295 error -13
> > 
> > After:
> 
> And "After commit 18c3a61c4264?"

Yes. Sorry I didn't make it clear.

> 
> > + ip netns exec server ip -6 route get 2002::1
> > RTNETLINK answers: Permission denied
> > 
> > Fix this by add prohibit and blk hole check.
> > 
> > At the same time, after commit
> > 2f460933f58e ("ipv6: initialize route null entry in addrconf_init()") and
> > 242d3a49a2a1 ("ipv6: reorder ip6_route_dev_notifier after ipv6_dev_notf")
> > We will init rt6i_idev correctly. So we could dump ip6_null_entry
> > (unreachable route entry) safely now.
> > 
> > Fixes: 18c3a61c4264 ("net: ipv6: RTM_GETROUTE: return matched fib...")
> > Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
> 
> This is what I see with your patch:
> 
> # ip -6 ro ls vrf red
> 2001:db8:1::/120 dev eth1 proto kernel metric 256  pref medium
> prohibit 5000::/120 dev lo metric 1024  error -13 pref medium
> fe80::/64 dev eth1 proto kernel metric 256  pref medium
> ff00::/8 dev eth1 metric 256  pref medium
> unreachable default dev lo metric 8192  error -113 pref medium
> 
> ie., I added a prohibit route for 5000:/120
> 
> and then running:
> # ip -6 ro get vrf red 5000::1
> RTNETLINK answers: Permission denied
> 
> Which is the behavior without your patch.
> 
> Now if I delete just the first bit:
> 
> diff --git a/net/ipv6/route.c b/net/ipv6/route.c
> index 4d30c96a819d..8fc52de40175 100644
> --- a/net/ipv6/route.c
> +++ b/net/ipv6/route.c
> @@ -3637,12 +3637,6 @@ static int inet6_rtm_getroute(struct sk_buff
> *in_skb, struct nlmsghdr *nlh,
>                 dst = ip6_route_lookup(net, &fl6, 0);
> 
>         rt = container_of(dst, struct rt6_info, dst);
> -       if (rt->dst.error) {
> -               err = rt->dst.error;
> -               ip6_rt_put(rt);
> -               goto errout;
> -       }
> -
>         if (rt == net->ipv6.ip6_null_entry) {
>                 err = rt->dst.error;
>                 ip6_rt_put(rt);
> 
> Then I get:
> 
> # ip -6 ro get vrf red 5000::1
> prohibit 5000::1 from :: dev lo table red src 2001:db8::2 metric 1024
> error -13 pref medium
> 
> which seems to be your objective.

Yes
>
>> I don't understand why you are focused on the built-in null and prohibit
>> route entries.
>
>I see. You are using fib rules for the prohibit entry; I am using an
>explicit route entry.

Yes, This is my fault. I should put my steps in the commit.

# ip -6 rule add to 2003::1/64 table 100 prohibit
# ip -6 route get 2003::1
RTNETLINK answers: Permission denied

>
>If I run 'ip ro get fibmatch' for the latter I want to see that route
>entry since it is a route in the FIB:
>
># ip -6 ro get fibmatch vrf red 5000::1
>prohibit 5000::/120 dev lo table red metric 1024 error -13 pref medium
>
>So there are multiple cases to verify.

Yeah... there are more stuff I need to learn.

Thanks
Hangbin

  parent reply	other threads:[~2017-07-27 13:48 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-07-20 14:51 [PATCH net] ipv6: no need to return rt->dst.error if it is not null entry Hangbin Liu
2017-07-20 15:06 ` Hangbin Liu
2017-07-20 15:23   ` Hangbin Liu
2017-07-21 15:53     ` David Ahern
2017-07-21 18:42     ` Cong Wang
2017-07-21 21:53       ` Roopa Prabhu
2017-07-23  4:54         ` Roopa Prabhu
2017-07-24  3:09       ` Hangbin Liu
2017-07-24 19:57         ` Cong Wang
2017-07-25  0:08           ` Hangbin Liu
2017-07-25  3:28             ` David Ahern
2017-07-25  7:32               ` Hangbin Liu
2017-07-26 17:18                 ` David Ahern
2017-07-26 18:27                   ` Roopa Prabhu
2017-07-26 18:49                     ` David Ahern
2017-07-26 18:55                       ` Roopa Prabhu
2017-07-26 19:00                         ` David Ahern
2017-07-26 19:38                           ` Roopa Prabhu
2017-07-27 16:08                           ` Hangbin Liu
2017-07-28  4:56                       ` Cong Wang
2017-07-28 11:04                         ` Hangbin Liu
2017-07-28 15:10                         ` David Ahern
2017-07-28 17:13                           ` Roopa Prabhu
2017-07-28 17:39                             ` David Ahern
2017-07-28 19:52                               ` Roopa Prabhu
2017-07-29 14:41                                 ` David Ahern
2017-07-31 18:37                               ` Cong Wang
2017-07-31 18:40                                 ` David Ahern
2017-07-25 17:49             ` Cong Wang
2017-07-26  9:18               ` Hangbin Liu
2017-07-21  3:47 ` [PATCHv2 net] ipv6: should not return rt->dst.error if it is prohibit or blk hole entry Hangbin Liu
2017-07-21 15:29   ` kbuild test robot
2017-07-21 16:34   ` kbuild test robot
2017-07-23  4:55 ` [PATCH net] ipv6: no need to return rt->dst.error if it is not null entry Roopa Prabhu
2017-07-24  2:28   ` Hangbin Liu
2017-07-26  9:20 ` [PATCHv3 net] ipv6: no need to return rt->dst.error if it is prohibit entry Hangbin Liu
2017-07-26 17:09   ` David Ahern
2017-07-26 18:48     ` David Ahern
2017-07-27 13:48     ` Hangbin Liu [this message]
2017-07-27 16:25 ` [PATCHv4 net] ipv6: no need to check rt->dst.error when get route info Hangbin Liu
2017-07-27 18:03   ` David Ahern
2017-07-28 17:23     ` David Ahern
2017-07-27 19:52   ` Roopa Prabhu
2017-07-31 23:22   ` David Miller
2017-07-31 23:34     ` David Ahern
2017-07-31 23:39       ` David Miller

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170727134823.GD5465@leo.usersys.redhat.com \
    --to=liuhangbin@gmail.com \
    --cc=dsahern@gmail.com \
    --cc=netdev@vger.kernel.org \
    --cc=roopa@cumulusnetworks.com \
    --cc=xiyou.wangcong@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).