Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[Shirley Ma]

[-- Attachment #1: Type: text/plain, Size: 1151 bytes --]





Agree.

But if no applications use this value, and nothing is boken. It's not
harmful to change the kernel to provide MIBs in as-is format.

Thanks
Shirley Ma
IBM Linux Technology Center
15300 SW Koll Parkway
Beaverton, OR 97006-6063
Phone: (503) 578-7638
FAX:      (503) 578-3228



YOSHIFUJI Hideaki / 吉藤英明
<yoshfuji@linux-ipv6.org>@cerberus.hongo.wide.ad.jp> on 01/15/2004 10:36:22
AM

Sent by:    "Hideaki YOSHIFUJI" <yoshfuji@cerberus.hongo.wide.ad.jp>


To:    Shirley Ma/Beaverton/IBM@IBMUS
cc:    mashirle@us.ltcfwd.linux.ibm.com, davem@redhat.com,
       kuznet@ms2.inr.ac.ru, netdev@oss.sgi.com, yoshfuji@linux-ipv6.org
Subject:    Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable


In article <OF87A9D639.F6A8D26B-ON88256E1C.0064AD99@us.ibm.com> (at Thu, 15
Jan 2004 10:26:21 -0800), Shirley Ma <xma@us.ibm.com> says:

> This is a bug fix for ipDefaultRouter Table. According to the new IP
MIBs,
> the DefaultRouterLifeTime(expire) entry in this table is unsigned 32. It
> measures the remaining length of time in seconds. If it's expire, the
value
> should be 0.

Kernel does not have to provide MIB in as-is format.

--yoshfuji


[-- Attachment #2: Type: text/html, Size: 1499 bytes --]

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[YOSHIFUJI Hideaki / 吉藤英明]

In article <OFC6D64AE0.CEE7DA5F-ON87256E1C.0068A56A@us.ibm.com> (at Thu, 15 Jan 2004 11:09:47 -0800), Shirley Ma <xma@us.ibm.com> says:

> But if no applications use this value, and nothing is boken. It's not
> harmful to change the kernel to provide MIBs in as-is format.

Shirley, nobody can prove that nothing will be broken. :-)

--yoshfuji

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[David S. Miller]

On Fri, 16 Jan 2004 04:19:50 +0900 (JST)
YOSHIFUJI Hideaki / _$B5HF#1QL@ <yoshfuji@linux-ipv6.org> wrote:

> In article <OFC6D64AE0.CEE7DA5F-ON87256E1C.0068A56A@us.ibm.com> (at Thu, 15 Jan 2004 11:09:47 -0800), Shirley Ma <xma@us.ibm.com> says:
> 
> > But if no applications use this value, and nothing is boken. It's not
> > harmful to change the kernel to provide MIBs in as-is format.
> 
> Shirley, nobody can prove that nothing will be broken. :-)

And also, the current situation is OK as long as this value we present now
can be converted into the desired value.  Is this the case?

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[YOSHIFUJI Hideaki / 吉藤英明]

In article <20040115113700.166ddfb0.davem@redhat.com> (at Thu, 15 Jan 2004 11:37:00 -0800), "David S. Miller" <davem@redhat.com> says:

> And also, the current situation is OK as long as this value we present now
> can be converted into the desired value.  Is this the case?

I think so for the routing table AFAIK.
For neighbours, "jiffies" is required to be exported.

--yoshfuji

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[Shirley Ma]

[-- Attachment #1: Type: text/plain, Size: 1082 bytes --]





OK, I agree with both of you. Then this patch is not needed. Yoshifuji, you
can sleep in peace. ;-)

Thank you all!
Shirley Ma
IBM Linux Technology Center
15300 SW Koll Parkway
Beaverton, OR 97006-6063
Phone: (503) 578-7638
FAX:      (503) 578-3228



YOSHIFUJI Hideaki / 吉藤英明 <yoshfuji@linux-ipv6.org>@oss.sgi.com on
01/15/2004 11:53:41 AM

Sent by:    netdev-bounce@oss.sgi.com


To:    davem@redhat.com
cc:    Shirley Ma/Beaverton/IBM@IBMUS, mashirle@us.ltcfwd.linux.ibm.com,
       kuznet@ms2.inr.ac.ru, netdev@oss.sgi.com, yoshfuji@linux-ipv6.org
Subject:    Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable


In article <20040115113840.4b7eb5d8.davem@redhat.com> (at Thu, 15 Jan 2004
11:38:40 -0800), "David S. Miller" <davem@redhat.com> says:

> We can create new files to present information in new ways and formats,
> but I totally prefer using netlink based information gathering over
making
> new procfs files.

Although this discussion is about rtnetlink message,
problem will be solved if we provide jiffies for neigbours
IMHO.

--yoshfuji @ let me sleep...



[-- Attachment #2: Type: text/html, Size: 1435 bytes --]

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[Shirley Ma]

[-- Attachment #1: Type: text/plain, Size: 1360 bytes --]






Hi, Yoshifuji,

This rule would mean that any new kernel development *could* break
something somewhere else? ;-)

I don't see any application uses these values for now, so I change them.
It's good for kernel to provide the real MIBs information. If later on,
some applications start to use these values, they can get the correct
numbers without converting. Unless you can give me an broken example. :-)

Thanks
Shirley Ma
IBM Linux Technology Center
15300 SW Koll Parkway
Beaverton, OR 97006-6063
Phone: (503) 578-7638
FAX:      (503) 578-3228



YOSHIFUJI Hideaki / 吉藤英明
<yoshfuji@linux-ipv6.org>@cerberus.hongo.wide.ad.jp> on 01/15/2004 11:19:50
AM

Sent by:    "Hideaki YOSHIFUJI" <yoshfuji@cerberus.hongo.wide.ad.jp>


To:    Shirley Ma/Beaverton/IBM@IBMUS
cc:    mashirle@us.ltcfwd.linux.ibm.com, davem@redhat.com,
       kuznet@ms2.inr.ac.ru, netdev@oss.sgi.com, yoshfuji@linux-ipv6.org
Subject:    Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable


In article <OFC6D64AE0.CEE7DA5F-ON87256E1C.0068A56A@us.ibm.com> (at Thu, 15
Jan 2004 11:09:47 -0800), Shirley Ma <xma@us.ibm.com> says:

> But if no applications use this value, and nothing is boken. It's not
> harmful to change the kernel to provide MIBs in as-is format.

Shirley, nobody can prove that nothing will be broken. :-)

--yoshfuji


[-- Attachment #2: Type: text/html, Size: 1685 bytes --]

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[YOSHIFUJI Hideaki / 吉藤英明]

In article <OF6C177E72.62FA8BF9-ON87256E1C.006A8953@us.ibm.com> (at Thu, 15 Jan 2004 11:30:25 -0800), Shirley Ma <xma@us.ibm.com> says:

> I don't see any application uses these values for now, so I change them.
> It's good for kernel to provide the real MIBs information. If later on,
> some applications start to use these values, they can get the correct
> numbers without converting. Unless you can give me an broken example. :-)

ip -6 route ?

--yoshfuji

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[David S. Miller]

On Thu, 15 Jan 2004 11:30:25 -0800
Shirley Ma <xma@us.ibm.com> wrote:

> I don't see any application uses these values for now, so I change them.
> It's good for kernel to provide the real MIBs information. If later on,
> some applications start to use these values, they can get the correct
> numbers without converting. Unless you can give me an broken example. :-)

Format and layout of procfs files have to stay as they are, this is just the
nature of the game.

We can create new files to present information in new ways and formats,
but I totally prefer using netlink based information gathering over making
new procfs files.

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[YOSHIFUJI Hideaki / 吉藤英明]

In article <20040115113840.4b7eb5d8.davem@redhat.com> (at Thu, 15 Jan 2004 11:38:40 -0800), "David S. Miller" <davem@redhat.com> says:

> We can create new files to present information in new ways and formats,
> but I totally prefer using netlink based information gathering over making
> new procfs files.

Although this discussion is about rtnetlink message,
problem will be solved if we provide jiffies for neigbours 
IMHO.

--yoshfuji @ let me sleep...

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[YOSHIFUJI Hideaki / 吉藤英明]

In article <OF6C177E72.62FA8BF9-ON87256E1C.006A8953@us.ibm.com> (at Thu, 15 Jan 2004 11:30:25 -0800), Shirley Ma <xma@us.ibm.com> says:

> This rule would mean that any new kernel development *could* break
> something somewhere else? ;-)

Please try to avoid changing the interface as long as you can.
Please add something rather change it.

In these case, we can avoid changing the interface,
so let's avoid breaking interface.

For ipv6inetNetToMeiaTable case, it is ok to add "jiffies," 
but it wrong to change the value's context.
If you really want to "change" that, please change the name.

Thanks.

-- 
Hideaki YOSHIFUJI @ USAGI Project <yoshfuji@linux-ipv6.org>
GPG FP: 9022 65EB 1ECF 3AD1 0BDF  80D8 4807 F894 E062 0EEA

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[Shirley Ma]

[-- Attachment #1: Type: text/plain, Size: 992 bytes --]





This is a bug fix for ipDefaultRouter Table. According to the new IP MIBs,
the DefaultRouterLifeTime(expire) entry in this table is unsigned 32. It
measures the remaining length of time in seconds. If it's expire, the value
should be 0.

Thanks
Shirley Ma
IBM Linux Technology Center
15300 SW Koll Parkway
Beaverton, OR 97006-6063
Phone: (503) 578-7638
FAX:      (503) 578-3228



YOSHIFUJI Hideaki / 吉藤英明
<yoshfuji@linux-ipv6.org>@cerberus.hongo.wide.ad.jp> on 01/15/2004 01:03:15
AM

Sent by:    "Hideaki YOSHIFUJI" <yoshfuji@cerberus.hongo.wide.ad.jp>


To:    mashirle@us.ltcfwd.linux.ibm.com
cc:    davem@redhat.com, kuznet@ms2.inr.ac.ru, netdev@oss.sgi.com, Shirley
       Ma/Beaverton/IBM@IBMUS
Subject:    Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable


In article <200401141622.09299.mashirle@us.ibm.com> (at Wed, 14 Jan 2004
16:22:09 -0800), Shirley Ma <mashirle@us.ibm.com> says:

> This patch is agaist 2.6.1 kernel.

Wrong. Do not change user interface.

--yoshfuji


[-- Attachment #2: Type: text/html, Size: 1309 bytes --]

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[YOSHIFUJI Hideaki / 吉藤英明]

In article <OF87A9D639.F6A8D26B-ON88256E1C.0064AD99@us.ibm.com> (at Thu, 15 Jan 2004 10:26:21 -0800), Shirley Ma <xma@us.ibm.com> says:

> This is a bug fix for ipDefaultRouter Table. According to the new IP MIBs,
> the DefaultRouterLifeTime(expire) entry in this table is unsigned 32. It
> measures the remaining length of time in seconds. If it's expire, the value
> should be 0.

Kernel does not have to provide MIB in as-is format.

--yoshfuji

IPv6 MIB:ipv6PrefixTable implementation

[Shirley Ma]

Hi, Alexy,

A while ago, we had a discussion about IPv6 Prefix Table implementation. 
I agree with you it's not necessary to process offlink prefix.

In each ipv6PrefixTable, there are 9 objects, which are Ifindex, Type, Prefix, 
Length, Origin(addrconf, manually, dhcp, others), OnLinkFlag, AutonomoueFlag, 
AdvPreferredLiftTime and ValidLifeTime. 

As we know, the routing table has all on-link prefix routes. But only Prefix,
Length, Ifindex are saved in routing table, the rest objects info are not saved
anywhere.

One implementation detail question, do you think I need to save all the other 
Prefix Objects: Type, Origin(addrconf, manually, dhcp, others), OnLinkFlag, 
AutonomoueFlag, AdvPreferredLiftTime and ValidLifeTime in routing table also?
If so, sounds like I add a big structure in each prefix route entry in the routing table.

Thanks
Shirley Ma

IBM Linux Technology Center
=======================



List:     linux-netdev
Subject:  Re: [Issues] Implementation for IPv6 MIB:ipv6PrefixTable
From:     kuznet () ms2 ! inr ! ac ! ru
Date:     2003-10-24 17:59:27

Hello!

> First, do you think it's a good idea to save all prefix objects in routing
> table?

Those which are on-link, of course.


> Second, where is the best location for the off-link prefix, on-link flag 0
> prefix to be saved?

Do you really think off-link prefixes are to be saved? Prefix in use is onlink
by defintion, so just return 1 in these bit and ignore off-link prefixes
completely, they are meaningless, are not they?


> Third, do you think it's a good idea to implement a prefix table/per
> interface in the kernel?

Well, if SNMP is going to be used as a tool to snoop ndisc promiscuously
(which seems to be the case if it shows inforrmation about offlink prefixes),
then it is definitely not a kernel task.

Alexey

[PATCH] IPv6 MIB:ipv6Prefix netlink notification

[Shirley Ma]

[-- Attachment #1: Type: text/plain, Size: 224 bytes --]

Once receiving a router advertisement message for prefix info, 
a netlink notification event will be created.

This patch has been tested against linux-2.6.0-test11. 

Thanks
Shirley Ma
IBM Linux Technology Center


[-- Attachment #2: linux-2.6.0-test11-ipv6mib3.patch --]
[-- Type: text/x-diff, Size: 4590 bytes --]

diff -urN linux-2.6.0-test11/include/linux/rtnetlink.h linux-2.6.0-test11-ipv6mib3/include/linux/rtnetlink.h
--- linux-2.6.0-test11/include/linux/rtnetlink.h	2003-11-26 12:45:11.000000000 -0800
+++ linux-2.6.0-test11-ipv6mib3/include/linux/rtnetlink.h	2003-12-05 12:20:11.000000000 -0800
@@ -44,7 +44,10 @@
 #define	RTM_DELTFILTER	(RTM_BASE+29)
 #define	RTM_GETTFILTER	(RTM_BASE+30)
 
-#define	RTM_MAX		(RTM_BASE+31)
+#define RTM_NEWPREFIX	(RTM_BASE+36)
+#define RTM_GETPREFIX	(RTM_BASE+38)
+
+#define	RTM_MAX		(RTM_BASE+39)
 
 /* 
    Generic structure for encapsulation of optional route information.
@@ -458,6 +461,34 @@
 	unsigned	ifi_change;		/* IFF_* change mask */
 };
 
+/********************************************************************
+ *		prefix information 
+ ****/
+
+struct prefixmsg
+{
+	unsigned char	prefix_family;
+	int		prefix_ifindex;
+	unsigned char	prefix_type;
+	unsigned char	prefix_len;
+	unsigned char	prefix_flags;
+};
+
+enum 
+{
+	PREFIX_UNSPEC,
+	PREFIX_ADDRESS,
+	PREFIX_CACHEINFO,
+};
+
+#define PREFIX_MAX	PREFIX_CACHEINFO
+
+struct prefix_cacheinfo
+{
+	__u32	preferred_time;
+	__u32	valid_time;
+};
+
 /* The struct should be in sync with struct net_device_stats */
 struct rtnl_link_stats
 {
@@ -614,6 +645,8 @@
 #define RTMGRP_DECnet_IFADDR    0x1000
 #define RTMGRP_DECnet_ROUTE     0x4000
 
+#define RTMGRP_IPV6_PREFIX	0x20000
+
 /* End of information exported to user level */
 
 #ifdef __KERNEL__
diff -urN linux-2.6.0-test11/include/net/if_inet6.h linux-2.6.0-test11-ipv6mib3/include/net/if_inet6.h
--- linux-2.6.0-test11/include/net/if_inet6.h	2003-11-26 12:45:45.000000000 -0800
+++ linux-2.6.0-test11-ipv6mib3/include/net/if_inet6.h	2003-12-05 12:20:11.000000000 -0800
@@ -25,6 +25,10 @@
 #define IF_RA_RCVD	0x20
 #define IF_RS_SENT	0x10
 
+/* prefix flags */
+#define IF_PREFIX_ONLINK	0x01
+#define IF_PREFIX_AUTOCONF	0x02
+
 #ifdef __KERNEL__
 
 struct inet6_ifaddr 
diff -urN linux-2.6.0-test11/net/ipv6/addrconf.c linux-2.6.0-test11-ipv6mib3/net/ipv6/addrconf.c
--- linux-2.6.0-test11/net/ipv6/addrconf.c	2003-11-26 12:45:37.000000000 -0800
+++ linux-2.6.0-test11-ipv6mib3/net/ipv6/addrconf.c	2003-12-05 12:21:59.000000000 -0800
@@ -138,6 +138,8 @@
 static void addrconf_rs_timer(unsigned long data);
 static void ipv6_ifa_notify(int event, struct inet6_ifaddr *ifa);
 
+static void inet6_prefix_notify(int event, struct inet6_dev *idev, 
+				struct prefix_info *pinfo);
 static int ipv6_chk_same_addr(const struct in6_addr *addr, struct net_device *dev);
 
 static struct notifier_block *inet6addr_chain;
@@ -1491,6 +1493,7 @@
 			addrconf_verify(0);
 		}
 	}
+	inet6_prefix_notify(RTM_NEWPREFIX, in6_dev, pinfo);
 	in6_dev_put(in6_dev);
 }
 
@@ -2751,6 +2754,66 @@
 	return skb->len;
 }
 
+static int inet6_fill_prefix(struct sk_buff *skb, struct inet6_dev *idev,
+			struct prefix_info *pinfo, u32 pid, u32 seq, int event)
+{
+	struct prefixmsg	*pmsg;
+	struct nlmsghdr 	*nlh;
+	unsigned char		*b = skb->tail;
+	struct prefix_cacheinfo	ci;
+
+	nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*pmsg));
+	
+	if (pid) 
+		nlh->nlmsg_flags |= NLM_F_MULTI;
+	
+	pmsg = NLMSG_DATA(nlh);
+	pmsg->prefix_family = AF_INET6;
+	pmsg->prefix_ifindex = idev->dev->ifindex;
+	pmsg->prefix_len = pinfo->prefix_len;
+	pmsg->prefix_type = pinfo->type;
+	
+	pmsg->prefix_flags = 0;
+	if (pinfo->onlink)
+		pmsg->prefix_flags |= IF_PREFIX_ONLINK;
+	if (pinfo->autoconf)
+		pmsg->prefix_flags |= IF_PREFIX_AUTOCONF;
+
+	RTA_PUT(skb, PREFIX_ADDRESS, sizeof(pinfo->prefix), &pinfo->prefix);
+
+	ci.preferred_time = ntohl(pinfo->prefered);
+	ci.valid_time = ntohl(pinfo->valid);
+	RTA_PUT(skb, PREFIX_CACHEINFO, sizeof(ci), &ci);
+
+	nlh->nlmsg_len = skb->tail - b;
+	return skb->len;
+
+nlmsg_failure:
+rtattr_failure:
+	skb_trim(skb, b - skb->data);
+	return -1;
+}
+
+static void inet6_prefix_notify(int event, struct inet6_dev *idev, 
+			 struct prefix_info *pinfo)
+{
+	struct sk_buff *skb;
+	int size = NLMSG_SPACE(sizeof(struct prefixmsg)+128);
+
+	skb = alloc_skb(size, GFP_ATOMIC);
+	if (!skb) {
+		netlink_set_err(rtnl, 0, RTMGRP_IPV6_PREFIX, ENOBUFS);
+		return;
+	}
+	if (inet6_fill_prefix(skb, idev, pinfo, 0, 0, event) < 0) {
+		kfree_skb(skb);
+		netlink_set_err(rtnl, 0, RTMGRP_IPV6_PREFIX, EINVAL);
+		return;
+	}
+	NETLINK_CB(skb).dst_groups = RTMGRP_IPV6_PREFIX;
+	netlink_broadcast(rtnl, skb, 0, RTMGRP_IPV6_PREFIX, GFP_ATOMIC);
+}
+
 static struct rtnetlink_link inet6_rtnetlink_table[RTM_MAX - RTM_BASE + 1] = {
 	[RTM_GETLINK - RTM_BASE] = { .dumpit	= inet6_dump_ifinfo, },
 	[RTM_NEWADDR - RTM_BASE] = { .doit	= inet6_rtm_newaddr, },

Re: [PATCH] IPv6 MIB:ipv6Prefix netlink notification

[David S. Miller]

On Fri, 5 Dec 2003 13:51:47 -0800
Shirley Ma <mashirle@us.ibm.com> wrote:

> Once receiving a router advertisement message for prefix info, 
> a netlink notification event will be created.
> 
> This patch has been tested against linux-2.6.0-test11. 

Let's queue this up for 2.6.1, please resend it after 2.6.0 is
released.

[PATCH] IPv6 MIB:ipv6DefaultRouterTable

[Shirley Ma]

[-- Attachment #1: Type: text/plain, Size: 91 bytes --]

This patch is agaist 2.6.1 kernel.

Thanks
Shirley Ma
IBM Linux Technology Center



[-- Attachment #2: linux-2.6.1-ipv6mib7.patch --]
[-- Type: text/x-diff, Size: 1121 bytes --]

diff -urN linux-2.6.1/include/linux/rtnetlink.h linux-2.6.1-ipv6mib7/include/linux/rtnetlink.h
--- linux-2.6.1/include/linux/rtnetlink.h	2004-01-08 22:59:55.000000000 -0800
+++ linux-2.6.1-ipv6mib7/include/linux/rtnetlink.h	2004-01-13 10:41:55.000000000 -0800
@@ -247,7 +247,7 @@
 {
 	__u32	rta_clntref;
 	__u32	rta_lastuse;
-	__s32	rta_expires;
+	__u32	rta_expires;	/* seconds */
 	__u32	rta_error;
 	__u32	rta_used;
 
diff -urN linux-2.6.1/net/ipv6/route.c linux-2.6.1-ipv6mib7/net/ipv6/route.c
--- linux-2.6.1/net/ipv6/route.c	2004-01-08 22:59:48.000000000 -0800
+++ linux-2.6.1-ipv6mib7/net/ipv6/route.c	2004-01-13 10:41:55.000000000 -0800
@@ -1535,8 +1535,8 @@
 		RTA_PUT(skb, RTA_OIF, sizeof(int), &rt->rt6i_dev->ifindex);
 	RTA_PUT(skb, RTA_PRIORITY, 4, &rt->rt6i_metric);
 	ci.rta_lastuse = jiffies_to_clock_t(jiffies - rt->u.dst.lastuse);
-	if (rt->rt6i_expires)
-		ci.rta_expires = jiffies_to_clock_t(rt->rt6i_expires - jiffies);
+	if (rt->rt6i_expires && time_after(rt->rt6i_expires, jiffies))
+		ci.rta_expires = (rt->rt6i_expires - jiffies)/HZ;
 	else
 		ci.rta_expires = 0;
 	ci.rta_used = rt->u.dst.__use;

Re: [PATCH] IPv6 MIB:ipv6DefaultRouterTable

[YOSHIFUJI Hideaki / 吉藤英明]

In article <200401141622.09299.mashirle@us.ibm.com> (at Wed, 14 Jan 2004 16:22:09 -0800), Shirley Ma <mashirle@us.ibm.com> says:

> This patch is agaist 2.6.1 kernel.

Wrong. Do not change user interface.

--yoshfuji