add mask options to fwmark masking code
Subsystems (4)
PATCH 2/2 add mask options to fwmark masking code
[PATCH] This patch introduces a mask to the fwmark test cases in the advanced routing. This let's one test individual bits of the fwmark to determine how things should be routed (pick a routing table). This patch retains compatibility with tests that do not set the mask by assuming a mask of 0 is equivalent to a mask of 0xffffffff.
---
include/linux/rtnetlink.h | 1 + net/ipv4/fib_rules.c | 19 +++++++++++++++++-- 2 files changed, 18 insertions(+), 2 deletions(-)
bcdda64a16d4dfda6d95452bbf8541999121831a
include/linux/rtnetlink.h +1 -0
@@ -266,6 +266,7 @@ enum rtattr_type_t };#define RTA_FWMARK RTA_PROTOINFO +#define RTA_FWMARK_MASK RTA_CACHEINFO#define RTA_MAX (__RTA_MAX - 1)
net/ipv4/fib_rules.c +17 -2
@@ -68,6 +68,7 @@ struct fib_ruleu8 r_tos;#ifdef CONFIG_IP_ROUTE_FWMARK u32 r_fwmark; +u32 r_fwmark_mask;#endif int r_ifindex;#ifdef CONFIG_NET_CLS_ROUTE
@@ -117,6 +118,7 @@ int inet_rtm_delrule(struct sk_buff *skb rtm->rtm_tos == r->r_tos && #ifdef CONFIG_IP_ROUTE_FWMARK (!rta[RTA_FWMARK -1 ] ||memcmp (RTA_DATA (rta[RTA_FWMARK -1 ]), &r->r_fwmark,4 ) ==0 ) && + (!rta[RTA_FWMARK_MASK -1 ] ||memcmp (RTA_DATA (rta[RTA_FWMARK_MASK -1 ]), &r->r_fwmark_mask,4 ) ==0 ) && #endif (!rtm->rtm_type || rtm->rtm_type == r->r_action) && (!rta[RTA_PRIORITY-1] || memcmp(RTA_DATA(rta[RTA_PRIORITY-1]), &r->r_preference, 4) == 0) &&
@@ -202,6 +204,17 @@ int inet_rtm_newrule(struct sk_buff *skb#ifdef CONFIG_IP_ROUTE_FWMARK if (rta[RTA_FWMARK -1 ])memcpy (&new_r->r_fwmark,RTA_DATA (rta[RTA_FWMARK -1 ]),4 ); +if (rta[RTA_FWMARK_MASK -1 ]) +memcpy (&new_r->r_fwmark_mask,RTA_DATA (rta[RTA_FWMARK_MASK -1 ]),4 ); +/* +* if the user doesn't set a mask, then set it to care about +* all bits. This retains compatibility. Note it is impossible +* to match SOMETHING & 0xMASK == 0, because fwmark==0 means +* do not match fwmark at all. +*/ +if (new_r->r_fwmark_mask ==0 ) { + new_r->r_fwmark_mask =0xffffffff ; + }#endif new_r->r_action = rtm->rtm_type; new_r->r_flags = rtm->rtm_flags;
@@ -298,7 +311,7 @@ FRprintk("Lookup: %u.%u.%u.%u <- %u.%u.% ((daddr^r->r_dst) & r->r_dstmask) || (r->r_tos && r->r_tos != flp->fl4_tos) || #ifdef CONFIG_IP_ROUTE_FWMARK - (r->r_fwmark && r->r_fwmark != flp->fl4_fwmark) || + (r->r_fwmark && r->r_fwmark != (flp->fl4_fwmark & r->r_fwmark_mask)) || #endif (r->r_ifindex && r->r_ifindex != flp->iif)) continue;
@@ -382,8 +395,10 @@ static __inline__ int inet_fill_rule(str rtm->rtm_src_len = r->r_src_len; rtm->rtm_tos = r->r_tos;#ifdef CONFIG_IP_ROUTE_FWMARK -if (r->r_fwmark) +if (r->r_fwmark) {RTA_PUT (skb,RTA_FWMARK ,4 , &r->r_fwmark); +RTA_PUT (skb,RTA_FWMARK_MASK ,4 , &r->r_fwmark_mask); + }#endif rtm->rtm_table = r->r_table; rtm->rtm_protocol =0 ;
Quoted 13 lines · Michael Richardson
[PATCH] This patch introduces a mask to the fwmark test cases in the advanced routing. This let's one test individual bits of the fwmark to determine how things should be routed (pick a routing table). This patch retains compatibility with tests that do not set the mask by assuming a mask of 0 is equivalent to a mask of
0xffffffff.
bcdda64a16d4dfda6d95452bbf8541999121831ainclude/linux/rtnetlink.h +0
-0
include/linux/rtnetlink.h enum rtattr_type_t · line 269
Quoted 4 lines
include/linux/rtnetlink.h +1
-0@@ -266,6 +266,7 @@ enum rtattr_type_t };#define RTA_FWMARK RTA_PROTOINFO +#define RTA_FWMARK_MASK RTA_CACHEINFO
Please introduce a new attribute for this instead of overloading RTA_CACHEINFO.
net/ipv4/fib_rules.c struct fib_rule · line 71
Quoted 9 lines
net/ipv4/fib_rules.c +1
-0@@ -68,6 +68,7 @@ struct fib_ruleu8 r_tos;#ifdef CONFIG_IP_ROUTE_FWMARK u32 r_fwmark; +u32 r_fwmark_mask;
Both patches have whitespace issues. You should also change decnet, which also supports routing by fwmark. Other than that the patch looks fine, in fact its nearly identical to a patch I wanted to send soon which does the same :)
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 3 lines
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
"Patrick" == Patrick McHardy <[email protected]> writes:
#define
RTA_FWMARKRTA_PROTOINFO+#defineRTA_FWMARK_MASKRTA_CACHEINFO
Patrick> Please introduce a new attribute for this instead of Patrick> overloading RTA_CACHEINFO.
I would be happy to do that.
Should I also un-overload FWMARK, with backwards compatibility?
Quoted 5 lines
Quoted 5 lines
net/ipv4/fib_rules.c index renamed from net/ipv4/fib_rules.c +0
-0de327b3..69eed89 100644 ---
a/net/ipv4/fib_rules.c+++b/net/ipv4/fib_rules.c@@ -68,6 +68,7 @@ structfib_ruleu8r_tos; #ifdefCONFIG_IP_ROUTE_FWMARKu32r_fwmark; + u32r_fwmark_mask;
Patrick> Both patches have whitespace issues. You should also change
uhm. okay.
I'm surprised, since I produced it with git-format-patch. Maybe there
are tabs that emacs screwed up.
- --
] ON HUMILITY: to err is human. To moo, bovine. | firewalls [ ] Michael Richardson, Xelerance Corporation, Ottawa, ON |net architect[ ] [email protected] http://www.sandelman.ottawa.on.ca/mcr/ |device driver[
] panic("Just another Debian GNU/Linux using, kernel hacking, security guy"); [
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Finger me for keys
iQEVAwUBQ/4O2ICLcPvd0N1lAQK/egf6A0iQ1hvecR4BeaCrQiu53beGZd6zHldk o6logfar94kPP/H/D/kMcNeAvL2a3cJ8wyfyP02Cav8gP1C3X+XV+yLtA9jHIrdK nqQ1gw7F4Cj2+v7du/jS8GxNMWevXhJ7f9hvnzh8+DHMUCjqiksgsuIgcRQYrqOQ vxYERvR5TojEIaJfg8kH/lJRn3sm/APuMphM6c6SAeqrWpAdijbZb4LSNpGH50ci nNhUp+FxoP8vVFTMTu7M1MK4fpCIWA/PxBkmy3YDhcQx1+mE2nrEqHdbKfx9uY+t 0mxR8UC5sthhn94/VCjcqWOoHe3S/Gi+WWoPtwN1sFe5BujwU7Vcfw== =yKIA
-----END PGP SIGNATURE-----
Quoted 4 lines · Michael Richardson
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
"Patrick" == Patrick McHardy <[email protected]> writes:
Quoted 8 lines
Quoted 2 lines
Quoted 2 lines
#define
RTA_FWMARKRTA_PROTOINFO+#defineRTA_FWMARK_MASKRTA_CACHEINFOPatrick> Please introduce a new attribute for this instead of Patrick> overloading
RTA_CACHEINFO.I would be happy to do that.
Should I also un-overload FWMARK, with backwards compatibility?
No, that one is fine since it doesn't already have a different meaning.
Quoted 5 lines · Michael Richardson
[PATCH] This patch introduces a mask to the fwmark test cases in the advanced routing. This let's one test individual bits of the fwmark to determine how things should be routed (pick a routing table). This patch retains compatibility with tests that do not set the mask by assuming a mask of 0 is equivalent to a mask of
0xffffffff.
One last comment: instead of looking at the value you should check if the mask attribute is present to decide when to use the compat mask.
Quoted 5 lines · Michael Richardson
[PATCH] This patch introduces a mask to the fwmark test cases in the advanced routing. This let's one test individual bits of the fwmark to determine how things should be routed (pick a routing table). This patch retains compatibility with tests that do not set the mask by assuming a mask of 0 is equivalent to a mask of
0xffffffff.
Sorry if I misunderstood the intention of your patch, but isn't similar code already in mainline?
linux-2.6.16-rc3/net/sched/cls_u32.c:146
#ifdef CONFIG_CLS_U32_MARK
if ((skb->nfmark & n->mark.mask) != n->mark.val) {Regards,
Carl-Daniel
Signature · 1 line
http://www.hailfinger.org/