Skip to contentSkip to conversation

ipv6: clear RTF_EXPIRES when call ip6_rt_copy

PATCH In review [email protected] 5 participants 23 messages Dec 16, 2013 – Dec 19, 2013

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 ipv6: clear RTF_EXPIRES when call ip6_rt_copy

From: Li RongQing <[email protected]>

The commit ecd9883724b [ipv6: fix race condition regarding dst->expires and dst->from.] removed rt6_clean_expires in ip6_rt_copy, which causes the TAHI test case v6LC.4.1.4 { Reduce PMTU On-link } failed, since the newly generated rt maybe always expired {with RTF_EXPIRES flag, dst.expires is 0, and dst.from is NULL}.

Fix it by clearing RTF_EXPIRES as before 1716a96101[ipv6: fix problem with expired dst cache]

---
 net/ipv6/route.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/route.c +1 -1
index a1a5752..e3d7f21 100644
@@ -1908,7 +1908,7 @@ static struct rt6_info *ip6_rt_copy(struct rt6_info *ort,
 			rt->rt6i_gateway = ort->rt6i_gateway;
 		else
 			rt->rt6i_gateway = *dest;
-		rt->rt6i_flags = ort->rt6i_flags;
+		rt->rt6i_flags = ort->rt6i_flags & ~RTF_EXPIRES;
 		if ((ort->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) ==
 		    (RTF_DEFAULT | RTF_ADDRCONF))
 			rt6_set_from(rt, ort);
Signature · 1 line
1.7.10.4
#1 lore raw
Quoted 10 lines · On Mon, 2013-12-16 at 16:31 +0800, [email protected] wrote:

From: Li RongQing <[email protected]>

The commit ecd9883724b [ipv6: fix race condition regarding dst->expires and dst->from.] removed rt6_clean_expires in ip6_rt_copy, which causes the TAHI test case v6LC.4.1.4 { Reduce PMTU On-link } failed, since the newly generated rt maybe always expired {with RTF_EXPIRES flag, dst.expires is 0, and dst.from is NULL}.

Fix it by clearing RTF_EXPIRES as before 1716a96101[ipv6: fix problem with expired dst cache]

Any particular reason you didnt CC the author of the commit ?

Thanks

#2 lore raw
In reply to [email protected]
Quoted 22 lines · On 12/16/2013 04:31 PM, [email protected] wrote:

From: Li RongQing <[email protected]>

The commit ecd9883724b [ipv6: fix race condition regarding dst->expires and dst->from.] removed rt6_clean_expires in ip6_rt_copy, which causes the TAHI test case v6LC.4.1.4 { Reduce PMTU On-link } failed, since the newly generated rt maybe always expired {with RTF_EXPIRES flag, dst.expires is 0, and dst.from is NULL}.

Fix it by clearing RTF_EXPIRES as before 1716a96101[ipv6: fix problem with expired dst cache]

---
 net/ipv6/route.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/route.c +0 -0
index a1a5752..e3d7f21 100644

net/ipv6/route.c static struct rt6_info *ip6_rt_copy(struct rt6_info *ort, · line 1914

Quoted 8 lines
net/ipv6/route.c +1 -1
@@ -1908,7 +1908,7 @@ static struct rt6_info *ip6_rt_copy(struct rt6_info *ort,
 			rt->rt6i_gateway = ort->rt6i_gateway;
 		else
 			rt->rt6i_gateway = *dest;
-		rt->rt6i_flags = ort->rt6i_flags;
+		rt->rt6i_flags = ort->rt6i_flags & ~RTF_EXPIRES;
 		if ((ort->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) ==
 		    (RTF_DEFAULT | RTF_ADDRCONF))
 			rt6_set_from(rt, ort);

I don't understand, rt6_set_from already clears the RTF_EXPIRES from rt->rt6i_flags.

#3 lore raw

If the ort->rt6i_flags is RTF_EXPIRES|RTF_ADDRCONF, then rt6_set_from will not be called, and new created rt will have RTF_EXPIRES, but dst.expires is 0, and dst.from is NULL

-Roy

2013/12/17 Gao feng <[email protected]>:

Quoted 34 lines · On 12/16/2013 04:31 PM, [email protected] wrote:
Quoted 22 lines

From: Li RongQing <[email protected]>

The commit ecd9883724b [ipv6: fix race condition regarding dst->expires and dst->from.] removed rt6_clean_expires in ip6_rt_copy, which causes the TAHI test case v6LC.4.1.4 { Reduce PMTU On-link } failed, since the newly generated rt maybe always expired {with RTF_EXPIRES flag, dst.expires is 0, and dst.from is NULL}.

Fix it by clearing RTF_EXPIRES as before 1716a96101[ipv6: fix problem with expired dst cache]

---
 net/ipv6/route.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/route.c +0 -0
index a1a5752..e3d7f21 100644

net/ipv6/route.c static struct rt6_info *ip6_rt_copy(struct rt6_info *ort, · line 1914

Quoted 8 lines
net/ipv6/route.c +1 -1
@@ -1908,7 +1908,7 @@ static struct rt6_info *ip6_rt_copy(struct rt6_info *ort,
                      rt->rt6i_gateway = ort->rt6i_gateway;
              else
                      rt->rt6i_gateway = *dest;
-             rt->rt6i_flags = ort->rt6i_flags;
+             rt->rt6i_flags = ort->rt6i_flags & ~RTF_EXPIRES;
              if ((ort->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) ==
                  (RTF_DEFAULT | RTF_ADDRCONF))
                      rt6_set_from(rt, ort);

I don't understand, rt6_set_from already clears the RTF_EXPIRES from rt->rt6i_flags.

#4 lore raw
Quoted 3 lines · RongQing Li

If the ort->rt6i_flags is RTF_EXPIRES|RTF_ADDRCONF, then rt6_set_from will not be called, and new created rt will have RTF_EXPIRES, but dst.expires is 0, and dst.from is NULL

Ok, but I think you need to add more detail/test purpose of the test case v6LC.4.1.4 { Reduce PMTU On-link }. just the number of test case is not good for people to know what's the real problem.

Quoted 39 lines

-Roy

2013/12/17 Gao feng <[email protected]>:

Quoted 34 lines · On 12/16/2013 04:31 PM, [email protected] wrote:
Quoted 22 lines

From: Li RongQing <[email protected]>

The commit ecd9883724b [ipv6: fix race condition regarding dst->expires and dst->from.] removed rt6_clean_expires in ip6_rt_copy, which causes the TAHI test case v6LC.4.1.4 { Reduce PMTU On-link } failed, since the newly generated rt maybe always expired {with RTF_EXPIRES flag, dst.expires is 0, and dst.from is NULL}.

Fix it by clearing RTF_EXPIRES as before 1716a96101[ipv6: fix problem with expired dst cache]

---
 net/ipv6/route.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/route.c +0 -0
index a1a5752..e3d7f21 100644

net/ipv6/route.c static struct rt6_info *ip6_rt_copy(struct rt6_info *ort, · line 1914

Quoted 8 lines
net/ipv6/route.c +1 -1
@@ -1908,7 +1908,7 @@ static struct rt6_info *ip6_rt_copy(struct rt6_info *ort,
                      rt->rt6i_gateway = ort->rt6i_gateway;
              else
                      rt->rt6i_gateway = *dest;
-             rt->rt6i_flags = ort->rt6i_flags;
+             rt->rt6i_flags = ort->rt6i_flags & ~RTF_EXPIRES;
              if ((ort->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) ==
                  (RTF_DEFAULT | RTF_ADDRCONF))
                      rt6_set_from(rt, ort);

I don't understand, rt6_set_from already clears the RTF_EXPIRES from rt->rt6i_flags.

#5 lore raw
Quoted 7 lines · Gao feng
Quoted 5 lines · RongQing Li

If the ort->rt6i_flags is RTF_EXPIRES|RTF_ADDRCONF, then rt6_set_from will not
be called, and new created rt will have RTF_EXPIRES, but dst.expires is 0, and
dst.from is NULL

Quoted 6 lines

Ok, but I think you need to add more detail/test purpose of the test case v6LC.4.1.4
{ Reduce PMTU On-link }. just the number of test case is not good for people to know
what's the real problem.

I have a question, why does we set dst.from only when the ort has flag RTF_ADDRCONF
and RTF_DEFAULT?

-Roy

Quoted 48 lines
Quoted 46 lines

-Roy

2013/12/17 Gao feng <[email protected]>:

Quoted 41 lines · On 12/16/2013 04:31 PM, [email protected] wrote:
Quoted 34 lines

From: Li RongQing <[email protected]>

The commit ecd9883724b [ipv6: fix race condition regarding dst->expires and
dst->from.] removed rt6_clean_expires in ip6_rt_copy, which causes the TAHI
test case v6LC.4.1.4 { Reduce PMTU On-link } failed, since the newly generated
rt maybe always expired {with RTF_EXPIRES flag, dst.expires is 0, and dst.from
is NULL}.

Fix it by clearing RTF_EXPIRES as before 1716a96101[ipv6: fix problem with
expired dst cache]

---
 net/ipv6/route.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/route.c +1 -1
index a1a5752..e3d7f21 100644
@@ -1908,7 +1908,7 @@ static struct rt6_info *ip6_rt_copy(struct
 rt6_info *ort,
                      rt->rt6i_gateway = ort->rt6i_gateway;
              else
                      rt->rt6i_gateway = *dest;
-             rt->rt6i_flags = ort->rt6i_flags;
+             rt->rt6i_flags = ort->rt6i_flags & ~RTF_EXPIRES;
              if ((ort->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) ==
                  (RTF_DEFAULT | RTF_ADDRCONF))
Quoted 2 lines
                      rt6_set_from(rt, ort);

I don't understand, rt6_set_from already clears the RTF_EXPIRES from rt->rt6i_flags.

#6 lore raw
Quoted 15 lines · RongQing Li
Quoted 7 lines · Gao feng
Quoted 5 lines · RongQing Li

If the ort->rt6i_flags is RTF_EXPIRES|RTF_ADDRCONF, then rt6_set_from will not
be called, and new created rt will have RTF_EXPIRES, but dst.expires is 0, and
dst.from is NULL

Quoted 6 lines

Ok, but I think you need to add more detail/test purpose of the test case v6LC.4.1.4
{ Reduce PMTU On-link }. just the number of test case is not good for people to know
what's the real problem.

Quoted 3 lines

I have a question, why does we set dst.from only when the ort has flag RTF_ADDRCONF
and RTF_DEFAULT?

Good question. ;)

I wonder, too. In the past from and expires were a union and either from or expires was allowed to be used. In the commit you referred to this union was split into seperate fields.

It somehow worked in the past because routes normally have longer timeouts and the routes will get evicted from the route cache eventually. My guess is that we can set from unconditionally or copy over the expires value.

This patch already needed quite a long time to review for me and I am still unsure. :/

Gao, do you still remember why you used RTF_ADDRCONF|RTF_DEFAULT?

Greetings,

Hannes

#7 lore raw
Quoted 26 lines · Hannes Frederic Sowa
Quoted 15 lines · RongQing Li
Quoted 7 lines · Gao feng
Quoted 5 lines · RongQing Li

If the ort->rt6i_flags is RTF_EXPIRES|RTF_ADDRCONF, then rt6_set_from will not
be called, and new created rt will have RTF_EXPIRES, but dst.expires is 0, and
dst.from is NULL

Quoted 6 lines

Ok, but I think you need to add more detail/test purpose of the test case v6LC.4.1.4
{ Reduce PMTU On-link }. just the number of test case is not good for people to know
what's the real problem.

Quoted 3 lines

I have a question, why does we set dst.from only when the ort has flag RTF_ADDRCONF
and RTF_DEFAULT?

Good question. ;)

I wonder, too. In the past from and expires were a union and either from or expires was allowed to be used. In the commit you referred to this union was split into seperate fields.

Quoted 9 lines

It somehow worked in the past because routes normally have longer timeouts and the routes will get evicted from the route cache eventually. My guess is that we can set from unconditionally or copy over the expires value.

This patch already needed quite a long time to review for me and I am still unsure. :/

Gao, do you still remember why you used RTF_ADDRCONF|RTF_DEFAULT?

It's a mystery, I noticed this problem when I wrote the codes.
http://lists.openwall.net/netdev/2012/03/19/7

I used the flags RTF_ADDRCONF|RTF_DEFAULT because they are exist in rt6_{get,add,purge}_dflt_router.

The from of new cloned rt should not be set if it's impossible for the ort to be expired.

but seems we should set from if flags have RTF_ADDRCONF bit. RA package not only generate the default route.

#8 lore raw
Quoted 5 lines · Gao feng
Quoted 2 lines

Gao, do you still remember why you used RTF_ADDRCONF|RTF_DEFAULT?

It's a mystery, I noticed this problem when I wrote the codes.
http://lists.openwall.net/netdev/2012/03/19/7

I already found this thread, thanks. ;)

Quoted 2 lines

I used the flags RTF_ADDRCONF|RTF_DEFAULT because they are exist in rt6_{get,add,purge}_dflt_router.

I thought so, but we have to deal with !DEFAULT ADDRCONF routes, too.

Quoted 2 lines

The from of new cloned rt should not be set if it's impossible for the ort to be expired.

Ok.

Quoted 2 lines

but seems we should set from if flags have RTF_ADDRCONF bit. RA package not only generate the default route.

Exactly, but it is worse:

Prefix routes can be added with expiration time if user space installs a prefix with valid_lft != infinity. So I fear we already have permament route entries which expire.

Userspace router advertisment listener already use that.

I fear the flags don't have a well defined semantic any more. :(

As for the original patch in this thread, I would suggest to hold it back until this mess is understood. Ok?

Thanks,

Hannes

#9 lore raw
Quoted 7 lines · Hannes Frederic Sowa
Quoted 5 lines · Gao feng
Quoted 2 lines

Gao, do you still remember why you used RTF_ADDRCONF|RTF_DEFAULT?

It's a mystery, I noticed this problem when I wrote the codes.
http://lists.openwall.net/netdev/2012/03/19/7

Quoted 6 lines

I already found this thread, thanks. ;)

Quoted 2 lines

I used the flags RTF_ADDRCONF|RTF_DEFAULT because they are exist in rt6_{get,add,purge}_dflt_router.

I thought so, but we have to deal with !DEFAULT ADDRCONF routes, too.

Right.

Quoted 6 lines
Quoted 3 lines

The from of new cloned rt should not be set if it's impossible for the ort to be expired.

Ok.

Quoted 8 lines
Quoted 2 lines

but seems we should set from if flags have RTF_ADDRCONF bit. RA package not only generate the default route.

Exactly, but it is worse:

Prefix routes can be added with expiration time if user space installs a prefix with valid_lft != infinity. So I fear we already have permament route entries which expire.

Yes, this should happen..

Quoted 8 lines

Userspace router advertisment listener already use that.

I fear the flags don't have a well defined semantic any more. :(

As for the original patch in this thread, I would suggest to hold it back until this mess is understood. Ok?

I'm ok if we want the behave back to the commit 1716a96101[ipv6: fix problem with expired dst cache].

#10 lore raw
In reply to Gao feng
Quoted 2 lines · Gao feng

The from of new cloned rt should not be set if it's impossible for the ort to be expired.

Actually, why do you think so? What could go wrong?

Greetings,

Hannes

#11 lore raw
Quoted 6 lines · Hannes Frederic Sowa
Quoted 2 lines · Gao feng

The from of new cloned rt should not be set if it's impossible for the ort to be expired.

Actually, why do you think so? What could go wrong?

I just don't want rt6_check_expired to check some impossible expired routes.

#12 lore raw
Quoted 10 lines · Gao feng
Quoted 6 lines · Hannes Frederic Sowa
Quoted 2 lines · Gao feng

The from of new cloned rt should not be set if it's impossible for the ort to be expired.

Actually, why do you think so? What could go wrong?

I just don't want rt6_check_expired to check some impossible expired routes.

What is wrong if we set from for new cloned rt by checking if ort has RTF_EXPIRES flag?

-Roy

#13 lore raw
Quoted 6 lines · RongQing Li
Quoted 5 lines · Gao feng
Quoted 4 lines · Hannes Frederic Sowa
Quoted 2 lines · Gao feng

The from of new cloned rt should not be set if it's impossible for the ort to be expired.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

Actually, why do you think so? What could go wrong?

I just don't want rt6_check_expired to check some impossible expired routes.

What is wrong if we set from for new cloned rt by checking if ort has RTF_EXPIRES flag?

The RTF_EXPIRES flag may be changed by router advertisment package, the ort may become expired after you hadn't set from for new cloned rt.

we should set from even this kind of ort doesn't have RTF_EXPIRES flag.

#14 lore raw
Quoted 13 lines · Gao feng
Quoted 12 lines · RongQing Li
Quoted 10 lines · Gao feng
Quoted 6 lines · Hannes Frederic Sowa
Quoted 2 lines · Gao feng

The from of new cloned rt should not be set if it's impossible for the ort to be expired.

Actually, why do you think so? What could go wrong?

I just don't want rt6_check_expired to check some impossible expired routes.

Quoted 8 lines
Quoted 2 lines

What is wrong if we set from for new cloned rt by checking if ort has RTF_EXPIRES flag?

The RTF_EXPIRES flag may be changed by router advertisment package, the ort may become expired after you hadn't set from for new cloned rt.

we should set from even this kind of ort doesn't have RTF_EXPIRES flag.

Thanks;

Do we want to set from only from RA route? if so, we should check ort with RTF_ADDRCONF|RTF_DEFAULT, or RTF_ADDRCONF | RTF_ROUTEINFO,

like in rt6_fill_node

        else if (rt->rt6i_flags & RTF_ADDRCONF) {
                if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ROUTEINFO))
                        rtm->rtm_protocol = RTPROT_RA;
                else
                        rtm->rtm_protocol = RTPROT_KERNEL;
        }

-R

#15 lore raw
Quoted 21 lines · RongQing Li
Quoted 20 lines · Gao feng
Quoted 6 lines · RongQing Li
Quoted 5 lines · Gao feng
Quoted 4 lines · Hannes Frederic Sowa
Quoted 2 lines · Gao feng

The from of new cloned rt should not be set if it's impossible for the ort to be expired.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

Actually, why do you think so? What could go wrong?

I just don't want rt6_check_expired to check some impossible expired routes.

What is wrong if we set from for new cloned rt by checking if ort has RTF_EXPIRES flag?

The RTF_EXPIRES flag may be changed by router advertisment package, the ort may become expired after you hadn't set from for new cloned rt.

Quoted 7 lines
Quoted 1 line

we should set from even this kind of ort doesn't have RTF_EXPIRES flag.

Thanks;

Do we want to set from only from RA route? if so, we should check ort with RTF_ADDRCONF|RTF_DEFAULT, or RTF_ADDRCONF | RTF_ROUTEINFO,

Nope, as I said, also prefix routes which did get installed by hand locally can have an expiration. I don't see any flag combination which ensure a potential from does never expire. IMHO we should always from.

Also, in case we overwrite a route and the route is already in there, we reset the expiry values: search for rt6_set_expires in fib6_add_rt2node.

We had the same problem with ECMP routes. We only wanted to add non-expiration routes but we could not find a combination which did ensure that. In the end we now also add non-expiration routes to an ecmp set (see commit 307f2fb95e9b96).

Greetings,

Hannes

#16 lore raw
Quoted 30 lines · Hannes Frederic Sowa
Quoted 21 lines · RongQing Li
Quoted 20 lines · Gao feng
Quoted 6 lines · RongQing Li
Quoted 5 lines · Gao feng
Quoted 4 lines · Hannes Frederic Sowa
Quoted 2 lines · Gao feng

The from of new cloned rt should not be set if it's impossible for the ort to be expired.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

Actually, why do you think so? What could go wrong?

I just don't want rt6_check_expired to check some impossible expired routes.

What is wrong if we set from for new cloned rt by checking if ort has RTF_EXPIRES flag?

The RTF_EXPIRES flag may be changed by router advertisment package, the ort may become expired after you hadn't set from for new cloned rt.

Quoted 7 lines
Quoted 1 line

we should set from even this kind of ort doesn't have RTF_EXPIRES flag.

Thanks;

Do we want to set from only from RA route? if so, we should check ort with RTF_ADDRCONF|RTF_DEFAULT, or RTF_ADDRCONF | RTF_ROUTEINFO,

Quoted 6 lines

Nope, as I said, also prefix routes which did get installed by hand locally can have an expiration. I don't see any flag combination which ensure a potential from does never expire. IMHO we should always from.

Also, in case we overwrite a route and the route is already in there, we reset the expiry values: search for rt6_set_expires in fib6_add_rt2node.

Seem like user can add expired route by manually. rtmsg_to_fib6_config set the cfg->fc_expires. I'm ok if you want to set from always now.

Thanks

#17 lore raw
Quoted 2 lines · Gao feng

Seem like user can add expired route by manually. rtmsg_to_fib6_config set the cfg->fc_expires. I'm ok if you want to set from always now.

Roy, can you send an updated patch now?

Thanks,

Hannes

#18 lore raw
Quoted 10 lines · Hannes Frederic Sowa
Quoted 2 lines · Gao feng

Seem like user can add expired route by manually. rtmsg_to_fib6_config set the cfg->fc_expires. I'm ok if you want to set from always now.

Roy, can you send an updated patch now?

Thanks,

Hannes

Ok, Thanks

-Roy

#19 lore raw

PATCH v2 1/1 ipv6: always set the new created dst's from in ip6_rt_copy

In reply to Hannes Frederic Sowa

From: Li RongQing <[email protected]>

ip6_rt_copy only sets dst.from if ort has flag RTF_ADDRCONF and RTF_DEFAULT. but the prefix routes which did get installed by hand locally can have an expiration, and no any flag combination which can ensure a potential from does never expire, so we should always set the new created dst's from.

This also fixes the new created dst is always expired since the ort, which is created by RA, maybe has RTF_EXPIRES and RTF_ADDRCONF, but no RTF_DEFAULT.

---
 net/ipv6/route.c |    4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
net/ipv6/route.c +1 -3
index a1a5752..5da9fb1 100644
@@ -1909,9 +1909,7 @@ static struct rt6_info *ip6_rt_copy(struct rt6_info *ort,
 		else
 			rt->rt6i_gateway = *dest;
 		rt->rt6i_flags = ort->rt6i_flags;
-		if ((ort->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) ==
-		    (RTF_DEFAULT | RTF_ADDRCONF))
-			rt6_set_from(rt, ort);
+		rt6_set_from(rt, ort);
 		rt->rt6i_metric = 0;
 
 #ifdef CONFIG_IPV6_SUBTREES
Signature · 1 line
1.7.10.4
Acked-by on v2 1/1
#20 lore raw

Re: [PATCH v2] ipv6: always set the new created dst's from in ip6_rt_copy

Quoted 7 lines · On Thu, Dec 19, 2013 at 12:40:26PM +0800, [email protected] wrote:

From: Li RongQing <[email protected]>

ip6_rt_copy only sets dst.from if ort has flag RTF_ADDRCONF and RTF_DEFAULT. but the prefix routes which did get installed by hand locally can have an expiration, and no any flag combination which can ensure a potential from does never expire, so we should always set the new created dst's from.

Quoted 6 lines

This also fixes the new created dst is always expired since the ort, which is created by RA, maybe has RTF_EXPIRES and RTF_ADDRCONF, but no RTF_DEFAULT.

Thanks!

maintainer
#21 lore raw

Re: [PATCH v2] ipv6: always set the new created dst's from in ip6_rt_copy

In reply to [email protected]

From: [email protected]
Date: Thu, 19 Dec 2013 12:40:26 +0800

Quoted 7 lines

From: Li RongQing <[email protected]>

ip6_rt_copy only sets dst.from if ort has flag RTF_ADDRCONF and RTF_DEFAULT. but the prefix routes which did get installed by hand locally can have an expiration, and no any flag combination which can ensure a potential from does never expire, so we should always set the new created dst's from.

Quoted 6 lines

This also fixes the new created dst is always expired since the ort, which is created by RA, maybe has RTF_EXPIRES and RTF_ADDRCONF, but no RTF_DEFAULT.

Applied and queued up for -stable, thanks.

#22 lore raw
In reply to Gao feng
Quoted 9 lines · Gao feng
Quoted 6 lines · Hannes Frederic Sowa
Quoted 2 lines · Gao feng

The from of new cloned rt should not be set if it's impossible for the ort to be expired.

Actually, why do you think so? What could go wrong?

I just don't want rt6_check_expired to check some impossible expired routes.

Only for performance? It does not seem to do any harm?