2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* IP Virtual Server
|
|
|
|
* data structure and functionality definitions
|
|
|
|
*/
|
|
|
|
|
2008-07-31 21:45:24 -06:00
|
|
|
#ifndef _NET_IP_VS_H
|
|
|
|
#define _NET_IP_VS_H
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-07-31 21:45:24 -06:00
|
|
|
#include <linux/ip_vs.h> /* definitions shared with userland */
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-07-31 21:45:24 -06:00
|
|
|
#include <asm/types.h> /* for __uXX types */
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/list.h> /* for struct list_head */
|
|
|
|
#include <linux/spinlock.h> /* for struct rwlock_t */
|
2011-07-26 17:09:06 -06:00
|
|
|
#include <linux/atomic.h> /* for struct atomic_t */
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/compiler.h>
|
2005-12-26 21:43:12 -07:00
|
|
|
#include <linux/timer.h>
|
2011-11-23 18:12:59 -07:00
|
|
|
#include <linux/bug.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2005-12-26 21:43:12 -07:00
|
|
|
#include <net/checksum.h>
|
2008-09-02 07:55:33 -06:00
|
|
|
#include <linux/netfilter.h> /* for union nf_inet_addr */
|
2008-10-01 08:33:10 -06:00
|
|
|
#include <linux/ip.h>
|
2008-09-02 07:55:33 -06:00
|
|
|
#include <linux/ipv6.h> /* for struct ipv6hdr */
|
2011-11-20 20:39:03 -07:00
|
|
|
#include <net/ipv6.h>
|
2012-10-22 13:22:45 -06:00
|
|
|
#if IS_ENABLED(CONFIG_IP_VS_IPV6)
|
ipvs: Fix faulty IPv6 extension header handling in IPVS
IPv6 packets can contain extension headers, thus its wrong to assume
that the transport/upper-layer header, starts right after (struct
ipv6hdr) the IPv6 header. IPVS uses this false assumption, and will
write SNAT & DNAT modifications at a fixed pos which will corrupt the
message.
To fix this, proper header position must be found before modifying
packets. Introducing ip_vs_fill_iph_skb(), which uses ipv6_find_hdr()
to skip the exthdrs. It finds (1) the transport header offset, (2) the
protocol, and (3) detects if the packet is a fragment.
Note, that fragments in IPv6 is represented via an exthdr. Thus, this
is detected while skipping through the exthdrs.
This patch depends on commit 84018f55a:
"netfilter: ip6_tables: add flags parameter to ipv6_find_hdr()"
This also adds a dependency to ip6_tables.
Originally based on patch from: Hans Schillstrom
kABI notes:
Changing struct ip_vs_iphdr is a potential minor kABI breaker,
because external modules can be compiled with another version of
this struct. This should not matter, as they would most-likely
be using a compiled-in version of ip_vs_fill_iphdr(). When
recompiled, they will notice ip_vs_fill_iphdr() no longer exists,
and they have to used ip_vs_fill_iph_skb() instead.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:06:41 -06:00
|
|
|
#include <linux/netfilter_ipv6/ip6_tables.h>
|
|
|
|
#endif
|
2012-09-26 06:06:20 -06:00
|
|
|
#if IS_ENABLED(CONFIG_NF_CONNTRACK)
|
2010-09-21 09:35:41 -06:00
|
|
|
#include <net/netfilter/nf_conntrack.h>
|
|
|
|
#endif
|
2011-01-03 06:44:42 -07:00
|
|
|
#include <net/net_namespace.h> /* Netw namespace */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Generic access of ipvs struct
|
|
|
|
*/
|
|
|
|
static inline struct netns_ipvs *net_ipvs(struct net* net)
|
|
|
|
{
|
|
|
|
return net->ipvs;
|
|
|
|
}
|
2011-01-03 06:44:43 -07:00
|
|
|
/*
|
|
|
|
* Get net ptr from skb in traffic cases
|
|
|
|
* use skb_sknet when call is from userland (ioctl or netlink)
|
|
|
|
*/
|
2011-01-03 06:44:58 -07:00
|
|
|
static inline struct net *skb_net(const struct sk_buff *skb)
|
2011-01-03 06:44:43 -07:00
|
|
|
{
|
|
|
|
#ifdef CONFIG_NET_NS
|
|
|
|
#ifdef CONFIG_IP_VS_DEBUG
|
|
|
|
/*
|
|
|
|
* This is used for debug only.
|
|
|
|
* Start with the most likely hit
|
|
|
|
* End with BUG
|
|
|
|
*/
|
|
|
|
if (likely(skb->dev && skb->dev->nd_net))
|
|
|
|
return dev_net(skb->dev);
|
2011-04-04 07:25:18 -06:00
|
|
|
if (skb_dst(skb) && skb_dst(skb)->dev)
|
2011-01-03 06:44:43 -07:00
|
|
|
return dev_net(skb_dst(skb)->dev);
|
|
|
|
WARN(skb->sk, "Maybe skb_sknet should be used in %s() at line:%d\n",
|
|
|
|
__func__, __LINE__);
|
|
|
|
if (likely(skb->sk && skb->sk->sk_net))
|
|
|
|
return sock_net(skb->sk);
|
|
|
|
pr_err("There is no net ptr to find in the skb in %s() line:%d\n",
|
|
|
|
__func__, __LINE__);
|
|
|
|
BUG();
|
|
|
|
#else
|
|
|
|
return dev_net(skb->dev ? : skb_dst(skb)->dev);
|
|
|
|
#endif
|
|
|
|
#else
|
|
|
|
return &init_net;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2011-01-03 06:44:58 -07:00
|
|
|
static inline struct net *skb_sknet(const struct sk_buff *skb)
|
2011-01-03 06:44:43 -07:00
|
|
|
{
|
|
|
|
#ifdef CONFIG_NET_NS
|
|
|
|
#ifdef CONFIG_IP_VS_DEBUG
|
|
|
|
/* Start with the most likely hit */
|
|
|
|
if (likely(skb->sk && skb->sk->sk_net))
|
|
|
|
return sock_net(skb->sk);
|
|
|
|
WARN(skb->dev, "Maybe skb_net should be used instead in %s() line:%d\n",
|
|
|
|
__func__, __LINE__);
|
|
|
|
if (likely(skb->dev && skb->dev->nd_net))
|
|
|
|
return dev_net(skb->dev);
|
|
|
|
pr_err("There is no net ptr to find in the skb in %s() line:%d\n",
|
|
|
|
__func__, __LINE__);
|
|
|
|
BUG();
|
|
|
|
#else
|
|
|
|
return sock_net(skb->sk);
|
|
|
|
#endif
|
|
|
|
#else
|
|
|
|
return &init_net;
|
|
|
|
#endif
|
|
|
|
}
|
2011-01-03 06:44:56 -07:00
|
|
|
/*
|
|
|
|
* This one needed for single_open_net since net is stored directly in
|
2011-03-30 19:57:33 -06:00
|
|
|
* private not as a struct i.e. seq_file_net can't be used.
|
2011-01-03 06:44:56 -07:00
|
|
|
*/
|
|
|
|
static inline struct net *seq_file_single_net(struct seq_file *seq)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_NET_NS
|
|
|
|
return (struct net *)seq->private;
|
|
|
|
#else
|
|
|
|
return &init_net;
|
|
|
|
#endif
|
|
|
|
}
|
IPVS: Allow boot time change of hash size
I was very frustrated about the fact that I have to recompile the kernel
to change the hash size. So, I created this patch.
If IPVS is built-in you can append ip_vs.conn_tab_bits=?? to kernel
command line, or, if you built IPVS as modules, you can add
options ip_vs conn_tab_bits=??.
To keep everything backward compatible, you still can select the size at
compile time, and that will be used as default.
It has been about a year since this patch was originally posted
and subsequently dropped on the basis of insufficient test data.
Mark Bergsma has provided the following test results which seem
to strongly support the need for larger hash table sizes:
We do however run into the same problem with the default setting (212 =
4096 entries), as most of our LVS balancers handle around a million
connections/SLAB entries at any point in time (around 100-150 kpps
load). With only 4096 hash table entries this implies that each entry
consists of a linked list of 256 connections *on average*.
To provide some statistics, I did an oprofile run on an 2.6.31 kernel,
with both the default 4096 table size, and the same kernel recompiled
with IP_VS_CONN_TAB_BITS set to 18 (218 = 262144 entries). I built a
quick test setup with a part of Wikimedia/Wikipedia's live traffic
mirrored by the switch to the test host.
With the default setting, at ~ 120 kpps packet load we saw a typical %si
CPU usage of around 30-35%, and oprofile reported a hot spot in
ip_vs_conn_in_get:
samples % image name app name
symbol name
1719761 42.3741 ip_vs.ko ip_vs.ko ip_vs_conn_in_get
302577 7.4554 bnx2 bnx2 /bnx2
181984 4.4840 vmlinux vmlinux __ticket_spin_lock
128636 3.1695 vmlinux vmlinux ip_route_input
74345 1.8318 ip_vs.ko ip_vs.ko ip_vs_conn_out_get
68482 1.6874 vmlinux vmlinux mwait_idle
After loading the recompiled kernel with 218 entries, %si CPU usage
dropped in half to around 12-18%, and oprofile looks much healthier,
with only 7% spent in ip_vs_conn_in_get:
samples % image name app name
symbol name
265641 14.4616 bnx2 bnx2 /bnx2
143251 7.7986 vmlinux vmlinux __ticket_spin_lock
140661 7.6576 ip_vs.ko ip_vs.ko ip_vs_conn_in_get
94364 5.1372 vmlinux vmlinux mwait_idle
86267 4.6964 vmlinux vmlinux ip_route_input
[ horms@verge.net.au: trivial up-port and minor style fixes ]
Signed-off-by: Catalin(ux) M. BOIE <catab@embedromix.ro>
Cc: Mark Bergsma <mark@wikimedia.org>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Patrick McHardy <kaber@trash.net>
2010-01-04 21:50:24 -07:00
|
|
|
|
|
|
|
/* Connections' size value needed by ip_vs_ctl.c */
|
|
|
|
extern int ip_vs_conn_tab_size;
|
|
|
|
|
2008-09-02 07:55:34 -06:00
|
|
|
struct ip_vs_iphdr {
|
ipvs: Fix faulty IPv6 extension header handling in IPVS
IPv6 packets can contain extension headers, thus its wrong to assume
that the transport/upper-layer header, starts right after (struct
ipv6hdr) the IPv6 header. IPVS uses this false assumption, and will
write SNAT & DNAT modifications at a fixed pos which will corrupt the
message.
To fix this, proper header position must be found before modifying
packets. Introducing ip_vs_fill_iph_skb(), which uses ipv6_find_hdr()
to skip the exthdrs. It finds (1) the transport header offset, (2) the
protocol, and (3) detects if the packet is a fragment.
Note, that fragments in IPv6 is represented via an exthdr. Thus, this
is detected while skipping through the exthdrs.
This patch depends on commit 84018f55a:
"netfilter: ip6_tables: add flags parameter to ipv6_find_hdr()"
This also adds a dependency to ip6_tables.
Originally based on patch from: Hans Schillstrom
kABI notes:
Changing struct ip_vs_iphdr is a potential minor kABI breaker,
because external modules can be compiled with another version of
this struct. This should not matter, as they would most-likely
be using a compiled-in version of ip_vs_fill_iphdr(). When
recompiled, they will notice ip_vs_fill_iphdr() no longer exists,
and they have to used ip_vs_fill_iph_skb() instead.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:06:41 -06:00
|
|
|
__u32 len; /* IPv4 simply where L4 starts
|
|
|
|
IPv6 where L4 Transport Header starts */
|
2012-09-26 06:06:59 -06:00
|
|
|
__u32 thoff_reasm; /* Transport Header Offset in nfct_reasm skb */
|
ipvs: Fix faulty IPv6 extension header handling in IPVS
IPv6 packets can contain extension headers, thus its wrong to assume
that the transport/upper-layer header, starts right after (struct
ipv6hdr) the IPv6 header. IPVS uses this false assumption, and will
write SNAT & DNAT modifications at a fixed pos which will corrupt the
message.
To fix this, proper header position must be found before modifying
packets. Introducing ip_vs_fill_iph_skb(), which uses ipv6_find_hdr()
to skip the exthdrs. It finds (1) the transport header offset, (2) the
protocol, and (3) detects if the packet is a fragment.
Note, that fragments in IPv6 is represented via an exthdr. Thus, this
is detected while skipping through the exthdrs.
This patch depends on commit 84018f55a:
"netfilter: ip6_tables: add flags parameter to ipv6_find_hdr()"
This also adds a dependency to ip6_tables.
Originally based on patch from: Hans Schillstrom
kABI notes:
Changing struct ip_vs_iphdr is a potential minor kABI breaker,
because external modules can be compiled with another version of
this struct. This should not matter, as they would most-likely
be using a compiled-in version of ip_vs_fill_iphdr(). When
recompiled, they will notice ip_vs_fill_iphdr() no longer exists,
and they have to used ip_vs_fill_iph_skb() instead.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:06:41 -06:00
|
|
|
__u16 fragoffs; /* IPv6 fragment offset, 0 if first frag (or not frag)*/
|
|
|
|
__s16 protocol;
|
|
|
|
__s32 flags;
|
2008-09-02 07:55:34 -06:00
|
|
|
union nf_inet_addr saddr;
|
|
|
|
union nf_inet_addr daddr;
|
|
|
|
};
|
|
|
|
|
2012-09-26 06:06:59 -06:00
|
|
|
/* Dependency to module: nf_defrag_ipv6 */
|
|
|
|
#if defined(CONFIG_NF_DEFRAG_IPV6) || defined(CONFIG_NF_DEFRAG_IPV6_MODULE)
|
|
|
|
static inline struct sk_buff *skb_nfct_reasm(const struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
return skb->nfct_reasm;
|
|
|
|
}
|
|
|
|
static inline void *frag_safe_skb_hp(const struct sk_buff *skb, int offset,
|
|
|
|
int len, void *buffer,
|
|
|
|
const struct ip_vs_iphdr *ipvsh)
|
|
|
|
{
|
|
|
|
if (unlikely(ipvsh->fragoffs && skb_nfct_reasm(skb)))
|
|
|
|
return skb_header_pointer(skb_nfct_reasm(skb),
|
|
|
|
ipvsh->thoff_reasm, len, buffer);
|
|
|
|
|
|
|
|
return skb_header_pointer(skb, offset, len, buffer);
|
|
|
|
}
|
|
|
|
#else
|
|
|
|
static inline struct sk_buff *skb_nfct_reasm(const struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
static inline void *frag_safe_skb_hp(const struct sk_buff *skb, int offset,
|
|
|
|
int len, void *buffer,
|
|
|
|
const struct ip_vs_iphdr *ipvsh)
|
|
|
|
{
|
|
|
|
return skb_header_pointer(skb, offset, len, buffer);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2008-09-02 07:55:34 -06:00
|
|
|
static inline void
|
ipvs: Fix faulty IPv6 extension header handling in IPVS
IPv6 packets can contain extension headers, thus its wrong to assume
that the transport/upper-layer header, starts right after (struct
ipv6hdr) the IPv6 header. IPVS uses this false assumption, and will
write SNAT & DNAT modifications at a fixed pos which will corrupt the
message.
To fix this, proper header position must be found before modifying
packets. Introducing ip_vs_fill_iph_skb(), which uses ipv6_find_hdr()
to skip the exthdrs. It finds (1) the transport header offset, (2) the
protocol, and (3) detects if the packet is a fragment.
Note, that fragments in IPv6 is represented via an exthdr. Thus, this
is detected while skipping through the exthdrs.
This patch depends on commit 84018f55a:
"netfilter: ip6_tables: add flags parameter to ipv6_find_hdr()"
This also adds a dependency to ip6_tables.
Originally based on patch from: Hans Schillstrom
kABI notes:
Changing struct ip_vs_iphdr is a potential minor kABI breaker,
because external modules can be compiled with another version of
this struct. This should not matter, as they would most-likely
be using a compiled-in version of ip_vs_fill_iphdr(). When
recompiled, they will notice ip_vs_fill_iphdr() no longer exists,
and they have to used ip_vs_fill_iph_skb() instead.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:06:41 -06:00
|
|
|
ip_vs_fill_ip4hdr(const void *nh, struct ip_vs_iphdr *iphdr)
|
|
|
|
{
|
|
|
|
const struct iphdr *iph = nh;
|
|
|
|
|
|
|
|
iphdr->len = iph->ihl * 4;
|
|
|
|
iphdr->fragoffs = 0;
|
|
|
|
iphdr->protocol = iph->protocol;
|
|
|
|
iphdr->saddr.ip = iph->saddr;
|
|
|
|
iphdr->daddr.ip = iph->daddr;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* This function handles filling *ip_vs_iphdr, both for IPv4 and IPv6.
|
|
|
|
* IPv6 requires some extra work, as finding proper header position,
|
|
|
|
* depend on the IPv6 extension headers.
|
|
|
|
*/
|
|
|
|
static inline void
|
|
|
|
ip_vs_fill_iph_skb(int af, const struct sk_buff *skb, struct ip_vs_iphdr *iphdr)
|
2008-09-02 07:55:34 -06:00
|
|
|
{
|
|
|
|
#ifdef CONFIG_IP_VS_IPV6
|
|
|
|
if (af == AF_INET6) {
|
ipvs: Fix faulty IPv6 extension header handling in IPVS
IPv6 packets can contain extension headers, thus its wrong to assume
that the transport/upper-layer header, starts right after (struct
ipv6hdr) the IPv6 header. IPVS uses this false assumption, and will
write SNAT & DNAT modifications at a fixed pos which will corrupt the
message.
To fix this, proper header position must be found before modifying
packets. Introducing ip_vs_fill_iph_skb(), which uses ipv6_find_hdr()
to skip the exthdrs. It finds (1) the transport header offset, (2) the
protocol, and (3) detects if the packet is a fragment.
Note, that fragments in IPv6 is represented via an exthdr. Thus, this
is detected while skipping through the exthdrs.
This patch depends on commit 84018f55a:
"netfilter: ip6_tables: add flags parameter to ipv6_find_hdr()"
This also adds a dependency to ip6_tables.
Originally based on patch from: Hans Schillstrom
kABI notes:
Changing struct ip_vs_iphdr is a potential minor kABI breaker,
because external modules can be compiled with another version of
this struct. This should not matter, as they would most-likely
be using a compiled-in version of ip_vs_fill_iphdr(). When
recompiled, they will notice ip_vs_fill_iphdr() no longer exists,
and they have to used ip_vs_fill_iph_skb() instead.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:06:41 -06:00
|
|
|
const struct ipv6hdr *iph =
|
|
|
|
(struct ipv6hdr *)skb_network_header(skb);
|
2011-11-20 20:39:03 -07:00
|
|
|
iphdr->saddr.in6 = iph->saddr;
|
|
|
|
iphdr->daddr.in6 = iph->daddr;
|
2012-09-26 06:06:59 -06:00
|
|
|
/* ipv6_find_hdr() updates len, flags, thoff_reasm */
|
|
|
|
iphdr->thoff_reasm = 0;
|
ipvs: Fix faulty IPv6 extension header handling in IPVS
IPv6 packets can contain extension headers, thus its wrong to assume
that the transport/upper-layer header, starts right after (struct
ipv6hdr) the IPv6 header. IPVS uses this false assumption, and will
write SNAT & DNAT modifications at a fixed pos which will corrupt the
message.
To fix this, proper header position must be found before modifying
packets. Introducing ip_vs_fill_iph_skb(), which uses ipv6_find_hdr()
to skip the exthdrs. It finds (1) the transport header offset, (2) the
protocol, and (3) detects if the packet is a fragment.
Note, that fragments in IPv6 is represented via an exthdr. Thus, this
is detected while skipping through the exthdrs.
This patch depends on commit 84018f55a:
"netfilter: ip6_tables: add flags parameter to ipv6_find_hdr()"
This also adds a dependency to ip6_tables.
Originally based on patch from: Hans Schillstrom
kABI notes:
Changing struct ip_vs_iphdr is a potential minor kABI breaker,
because external modules can be compiled with another version of
this struct. This should not matter, as they would most-likely
be using a compiled-in version of ip_vs_fill_iphdr(). When
recompiled, they will notice ip_vs_fill_iphdr() no longer exists,
and they have to used ip_vs_fill_iph_skb() instead.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:06:41 -06:00
|
|
|
iphdr->len = 0;
|
|
|
|
iphdr->flags = 0;
|
|
|
|
iphdr->protocol = ipv6_find_hdr(skb, &iphdr->len, -1,
|
|
|
|
&iphdr->fragoffs,
|
|
|
|
&iphdr->flags);
|
2012-09-26 06:06:59 -06:00
|
|
|
/* get proto from re-assembled packet and it's offset */
|
|
|
|
if (skb_nfct_reasm(skb))
|
|
|
|
iphdr->protocol = ipv6_find_hdr(skb_nfct_reasm(skb),
|
|
|
|
&iphdr->thoff_reasm,
|
|
|
|
-1, NULL, NULL);
|
|
|
|
|
2008-09-02 07:55:34 -06:00
|
|
|
} else
|
|
|
|
#endif
|
|
|
|
{
|
ipvs: Fix faulty IPv6 extension header handling in IPVS
IPv6 packets can contain extension headers, thus its wrong to assume
that the transport/upper-layer header, starts right after (struct
ipv6hdr) the IPv6 header. IPVS uses this false assumption, and will
write SNAT & DNAT modifications at a fixed pos which will corrupt the
message.
To fix this, proper header position must be found before modifying
packets. Introducing ip_vs_fill_iph_skb(), which uses ipv6_find_hdr()
to skip the exthdrs. It finds (1) the transport header offset, (2) the
protocol, and (3) detects if the packet is a fragment.
Note, that fragments in IPv6 is represented via an exthdr. Thus, this
is detected while skipping through the exthdrs.
This patch depends on commit 84018f55a:
"netfilter: ip6_tables: add flags parameter to ipv6_find_hdr()"
This also adds a dependency to ip6_tables.
Originally based on patch from: Hans Schillstrom
kABI notes:
Changing struct ip_vs_iphdr is a potential minor kABI breaker,
because external modules can be compiled with another version of
this struct. This should not matter, as they would most-likely
be using a compiled-in version of ip_vs_fill_iphdr(). When
recompiled, they will notice ip_vs_fill_iphdr() no longer exists,
and they have to used ip_vs_fill_iph_skb() instead.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:06:41 -06:00
|
|
|
const struct iphdr *iph =
|
|
|
|
(struct iphdr *)skb_network_header(skb);
|
|
|
|
iphdr->len = iph->ihl * 4;
|
|
|
|
iphdr->fragoffs = 0;
|
|
|
|
iphdr->protocol = iph->protocol;
|
|
|
|
iphdr->saddr.ip = iph->saddr;
|
|
|
|
iphdr->daddr.ip = iph->daddr;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* This function is a faster version of ip_vs_fill_iph_skb().
|
|
|
|
* Where we only populate {s,d}addr (and avoid calling ipv6_find_hdr()).
|
|
|
|
* This is used by the some of the ip_vs_*_schedule() functions.
|
|
|
|
* (Mostly done to avoid ABI breakage of external schedulers)
|
|
|
|
*/
|
|
|
|
static inline void
|
|
|
|
ip_vs_fill_iph_addr_only(int af, const struct sk_buff *skb,
|
|
|
|
struct ip_vs_iphdr *iphdr)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_IP_VS_IPV6
|
|
|
|
if (af == AF_INET6) {
|
|
|
|
const struct ipv6hdr *iph =
|
|
|
|
(struct ipv6hdr *)skb_network_header(skb);
|
|
|
|
iphdr->saddr.in6 = iph->saddr;
|
|
|
|
iphdr->daddr.in6 = iph->daddr;
|
2012-10-22 13:22:45 -06:00
|
|
|
} else
|
ipvs: Fix faulty IPv6 extension header handling in IPVS
IPv6 packets can contain extension headers, thus its wrong to assume
that the transport/upper-layer header, starts right after (struct
ipv6hdr) the IPv6 header. IPVS uses this false assumption, and will
write SNAT & DNAT modifications at a fixed pos which will corrupt the
message.
To fix this, proper header position must be found before modifying
packets. Introducing ip_vs_fill_iph_skb(), which uses ipv6_find_hdr()
to skip the exthdrs. It finds (1) the transport header offset, (2) the
protocol, and (3) detects if the packet is a fragment.
Note, that fragments in IPv6 is represented via an exthdr. Thus, this
is detected while skipping through the exthdrs.
This patch depends on commit 84018f55a:
"netfilter: ip6_tables: add flags parameter to ipv6_find_hdr()"
This also adds a dependency to ip6_tables.
Originally based on patch from: Hans Schillstrom
kABI notes:
Changing struct ip_vs_iphdr is a potential minor kABI breaker,
because external modules can be compiled with another version of
this struct. This should not matter, as they would most-likely
be using a compiled-in version of ip_vs_fill_iphdr(). When
recompiled, they will notice ip_vs_fill_iphdr() no longer exists,
and they have to used ip_vs_fill_iph_skb() instead.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:06:41 -06:00
|
|
|
#endif
|
2012-10-22 13:22:45 -06:00
|
|
|
{
|
ipvs: Fix faulty IPv6 extension header handling in IPVS
IPv6 packets can contain extension headers, thus its wrong to assume
that the transport/upper-layer header, starts right after (struct
ipv6hdr) the IPv6 header. IPVS uses this false assumption, and will
write SNAT & DNAT modifications at a fixed pos which will corrupt the
message.
To fix this, proper header position must be found before modifying
packets. Introducing ip_vs_fill_iph_skb(), which uses ipv6_find_hdr()
to skip the exthdrs. It finds (1) the transport header offset, (2) the
protocol, and (3) detects if the packet is a fragment.
Note, that fragments in IPv6 is represented via an exthdr. Thus, this
is detected while skipping through the exthdrs.
This patch depends on commit 84018f55a:
"netfilter: ip6_tables: add flags parameter to ipv6_find_hdr()"
This also adds a dependency to ip6_tables.
Originally based on patch from: Hans Schillstrom
kABI notes:
Changing struct ip_vs_iphdr is a potential minor kABI breaker,
because external modules can be compiled with another version of
this struct. This should not matter, as they would most-likely
be using a compiled-in version of ip_vs_fill_iphdr(). When
recompiled, they will notice ip_vs_fill_iphdr() no longer exists,
and they have to used ip_vs_fill_iph_skb() instead.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:06:41 -06:00
|
|
|
const struct iphdr *iph =
|
|
|
|
(struct iphdr *)skb_network_header(skb);
|
2008-09-02 07:55:34 -06:00
|
|
|
iphdr->saddr.ip = iph->saddr;
|
|
|
|
iphdr->daddr.ip = iph->daddr;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ip_vs_addr_copy(int af, union nf_inet_addr *dst,
|
|
|
|
const union nf_inet_addr *src)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_IP_VS_IPV6
|
|
|
|
if (af == AF_INET6)
|
2011-11-20 20:39:03 -07:00
|
|
|
dst->in6 = src->in6;
|
2008-09-02 07:55:34 -06:00
|
|
|
else
|
|
|
|
#endif
|
|
|
|
dst->ip = src->ip;
|
|
|
|
}
|
|
|
|
|
2013-03-21 03:58:12 -06:00
|
|
|
static inline void ip_vs_addr_set(int af, union nf_inet_addr *dst,
|
|
|
|
const union nf_inet_addr *src)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_IP_VS_IPV6
|
|
|
|
if (af == AF_INET6) {
|
|
|
|
dst->in6 = src->in6;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
dst->ip = src->ip;
|
|
|
|
dst->all[1] = 0;
|
|
|
|
dst->all[2] = 0;
|
|
|
|
dst->all[3] = 0;
|
|
|
|
}
|
|
|
|
|
2008-09-02 07:55:34 -06:00
|
|
|
static inline int ip_vs_addr_equal(int af, const union nf_inet_addr *a,
|
|
|
|
const union nf_inet_addr *b)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_IP_VS_IPV6
|
|
|
|
if (af == AF_INET6)
|
|
|
|
return ipv6_addr_equal(&a->in6, &b->in6);
|
|
|
|
#endif
|
|
|
|
return a->ip == b->ip;
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
#ifdef CONFIG_IP_VS_DEBUG
|
2005-12-26 21:43:12 -07:00
|
|
|
#include <linux/net.h>
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
extern int ip_vs_get_debug_level(void);
|
2008-09-02 07:55:35 -06:00
|
|
|
|
|
|
|
static inline const char *ip_vs_dbg_addr(int af, char *buf, size_t buf_len,
|
|
|
|
const union nf_inet_addr *addr,
|
|
|
|
int *idx)
|
|
|
|
{
|
|
|
|
int len;
|
|
|
|
#ifdef CONFIG_IP_VS_IPV6
|
|
|
|
if (af == AF_INET6)
|
2012-09-26 06:05:53 -06:00
|
|
|
len = snprintf(&buf[*idx], buf_len - *idx, "[%pI6c]",
|
2008-10-28 17:09:23 -06:00
|
|
|
&addr->in6) + 1;
|
2008-09-02 07:55:35 -06:00
|
|
|
else
|
|
|
|
#endif
|
2008-10-31 01:56:49 -06:00
|
|
|
len = snprintf(&buf[*idx], buf_len - *idx, "%pI4",
|
|
|
|
&addr->ip) + 1;
|
2008-09-02 07:55:35 -06:00
|
|
|
|
|
|
|
*idx += len;
|
|
|
|
BUG_ON(*idx > buf_len + 1);
|
|
|
|
return &buf[*idx - len];
|
|
|
|
}
|
|
|
|
|
2009-07-30 15:29:44 -06:00
|
|
|
#define IP_VS_DBG_BUF(level, msg, ...) \
|
|
|
|
do { \
|
|
|
|
char ip_vs_dbg_buf[160]; \
|
|
|
|
int ip_vs_dbg_idx = 0; \
|
|
|
|
if (level <= ip_vs_get_debug_level()) \
|
|
|
|
printk(KERN_DEBUG pr_fmt(msg), ##__VA_ARGS__); \
|
|
|
|
} while (0)
|
|
|
|
#define IP_VS_ERR_BUF(msg...) \
|
|
|
|
do { \
|
|
|
|
char ip_vs_dbg_buf[160]; \
|
|
|
|
int ip_vs_dbg_idx = 0; \
|
|
|
|
pr_err(msg); \
|
|
|
|
} while (0)
|
2008-09-02 07:55:35 -06:00
|
|
|
|
|
|
|
/* Only use from within IP_VS_DBG_BUF() or IP_VS_ERR_BUF macros */
|
2009-07-30 15:29:44 -06:00
|
|
|
#define IP_VS_DBG_ADDR(af, addr) \
|
|
|
|
ip_vs_dbg_addr(af, ip_vs_dbg_buf, \
|
|
|
|
sizeof(ip_vs_dbg_buf), addr, \
|
|
|
|
&ip_vs_dbg_idx)
|
|
|
|
|
|
|
|
#define IP_VS_DBG(level, msg, ...) \
|
|
|
|
do { \
|
|
|
|
if (level <= ip_vs_get_debug_level()) \
|
|
|
|
printk(KERN_DEBUG pr_fmt(msg), ##__VA_ARGS__); \
|
|
|
|
} while (0)
|
|
|
|
#define IP_VS_DBG_RL(msg, ...) \
|
|
|
|
do { \
|
|
|
|
if (net_ratelimit()) \
|
|
|
|
printk(KERN_DEBUG pr_fmt(msg), ##__VA_ARGS__); \
|
|
|
|
} while (0)
|
2010-10-17 07:46:17 -06:00
|
|
|
#define IP_VS_DBG_PKT(level, af, pp, skb, ofs, msg) \
|
2009-07-30 15:29:44 -06:00
|
|
|
do { \
|
|
|
|
if (level <= ip_vs_get_debug_level()) \
|
2010-10-17 07:46:17 -06:00
|
|
|
pp->debug_packet(af, pp, skb, ofs, msg); \
|
2009-07-30 15:29:44 -06:00
|
|
|
} while (0)
|
2010-10-17 07:46:17 -06:00
|
|
|
#define IP_VS_DBG_RL_PKT(level, af, pp, skb, ofs, msg) \
|
2009-07-30 15:29:44 -06:00
|
|
|
do { \
|
|
|
|
if (level <= ip_vs_get_debug_level() && \
|
|
|
|
net_ratelimit()) \
|
2010-10-17 07:46:17 -06:00
|
|
|
pp->debug_packet(af, pp, skb, ofs, msg); \
|
2009-07-30 15:29:44 -06:00
|
|
|
} while (0)
|
2005-04-16 16:20:36 -06:00
|
|
|
#else /* NO DEBUGGING at ALL */
|
2008-09-02 07:55:35 -06:00
|
|
|
#define IP_VS_DBG_BUF(level, msg...) do {} while (0)
|
|
|
|
#define IP_VS_ERR_BUF(msg...) do {} while (0)
|
2005-04-16 16:20:36 -06:00
|
|
|
#define IP_VS_DBG(level, msg...) do {} while (0)
|
|
|
|
#define IP_VS_DBG_RL(msg...) do {} while (0)
|
2010-10-17 07:46:17 -06:00
|
|
|
#define IP_VS_DBG_PKT(level, af, pp, skb, ofs, msg) do {} while (0)
|
|
|
|
#define IP_VS_DBG_RL_PKT(level, af, pp, skb, ofs, msg) do {} while (0)
|
2005-04-16 16:20:36 -06:00
|
|
|
#endif
|
|
|
|
|
|
|
|
#define IP_VS_BUG() BUG()
|
2009-08-02 05:05:41 -06:00
|
|
|
#define IP_VS_ERR_RL(msg, ...) \
|
2009-07-30 15:29:44 -06:00
|
|
|
do { \
|
|
|
|
if (net_ratelimit()) \
|
2009-08-02 05:05:41 -06:00
|
|
|
pr_err(msg, ##__VA_ARGS__); \
|
2009-07-30 15:29:44 -06:00
|
|
|
} while (0)
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
#ifdef CONFIG_IP_VS_DEBUG
|
|
|
|
#define EnterFunction(level) \
|
2009-07-30 15:29:44 -06:00
|
|
|
do { \
|
|
|
|
if (level <= ip_vs_get_debug_level()) \
|
|
|
|
printk(KERN_DEBUG \
|
|
|
|
pr_fmt("Enter: %s, %s line %i\n"), \
|
|
|
|
__func__, __FILE__, __LINE__); \
|
|
|
|
} while (0)
|
|
|
|
#define LeaveFunction(level) \
|
|
|
|
do { \
|
|
|
|
if (level <= ip_vs_get_debug_level()) \
|
|
|
|
printk(KERN_DEBUG \
|
|
|
|
pr_fmt("Leave: %s, %s line %i\n"), \
|
|
|
|
__func__, __FILE__, __LINE__); \
|
|
|
|
} while (0)
|
2005-04-16 16:20:36 -06:00
|
|
|
#else
|
|
|
|
#define EnterFunction(level) do {} while (0)
|
|
|
|
#define LeaveFunction(level) do {} while (0)
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The port number of FTP service (in network order).
|
|
|
|
*/
|
2009-02-14 23:58:35 -07:00
|
|
|
#define FTPPORT cpu_to_be16(21)
|
|
|
|
#define FTPDATA cpu_to_be16(20)
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* TCP State Values
|
|
|
|
*/
|
|
|
|
enum {
|
|
|
|
IP_VS_TCP_S_NONE = 0,
|
|
|
|
IP_VS_TCP_S_ESTABLISHED,
|
|
|
|
IP_VS_TCP_S_SYN_SENT,
|
|
|
|
IP_VS_TCP_S_SYN_RECV,
|
|
|
|
IP_VS_TCP_S_FIN_WAIT,
|
|
|
|
IP_VS_TCP_S_TIME_WAIT,
|
|
|
|
IP_VS_TCP_S_CLOSE,
|
|
|
|
IP_VS_TCP_S_CLOSE_WAIT,
|
|
|
|
IP_VS_TCP_S_LAST_ACK,
|
|
|
|
IP_VS_TCP_S_LISTEN,
|
|
|
|
IP_VS_TCP_S_SYNACK,
|
|
|
|
IP_VS_TCP_S_LAST
|
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* UDP State Values
|
|
|
|
*/
|
|
|
|
enum {
|
|
|
|
IP_VS_UDP_S_NORMAL,
|
|
|
|
IP_VS_UDP_S_LAST,
|
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* ICMP State Values
|
|
|
|
*/
|
|
|
|
enum {
|
|
|
|
IP_VS_ICMP_S_NORMAL,
|
|
|
|
IP_VS_ICMP_S_LAST,
|
|
|
|
};
|
|
|
|
|
2010-02-18 04:31:05 -07:00
|
|
|
/*
|
|
|
|
* SCTP State Values
|
|
|
|
*/
|
|
|
|
enum ip_vs_sctp_states {
|
|
|
|
IP_VS_SCTP_S_NONE,
|
|
|
|
IP_VS_SCTP_S_INIT_CLI,
|
|
|
|
IP_VS_SCTP_S_INIT_SER,
|
|
|
|
IP_VS_SCTP_S_INIT_ACK_CLI,
|
|
|
|
IP_VS_SCTP_S_INIT_ACK_SER,
|
|
|
|
IP_VS_SCTP_S_ECHO_CLI,
|
|
|
|
IP_VS_SCTP_S_ECHO_SER,
|
|
|
|
IP_VS_SCTP_S_ESTABLISHED,
|
|
|
|
IP_VS_SCTP_S_SHUT_CLI,
|
|
|
|
IP_VS_SCTP_S_SHUT_SER,
|
|
|
|
IP_VS_SCTP_S_SHUT_ACK_CLI,
|
|
|
|
IP_VS_SCTP_S_SHUT_ACK_SER,
|
|
|
|
IP_VS_SCTP_S_CLOSED,
|
|
|
|
IP_VS_SCTP_S_LAST
|
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* Delta sequence info structure
|
|
|
|
* Each ip_vs_conn has 2 (output AND input seq. changes).
|
|
|
|
* Only used in the VS/NAT.
|
|
|
|
*/
|
|
|
|
struct ip_vs_seq {
|
|
|
|
__u32 init_seq; /* Add delta from this seq */
|
|
|
|
__u32 delta; /* Delta in sequence numbers */
|
|
|
|
__u32 previous_delta; /* Delta in sequence numbers
|
|
|
|
before last resized pkt */
|
|
|
|
};
|
|
|
|
|
2011-01-03 06:44:56 -07:00
|
|
|
/*
|
|
|
|
* counters per cpu
|
|
|
|
*/
|
|
|
|
struct ip_vs_counters {
|
|
|
|
__u32 conns; /* connections scheduled */
|
|
|
|
__u32 inpkts; /* incoming packets */
|
|
|
|
__u32 outpkts; /* outgoing packets */
|
|
|
|
__u64 inbytes; /* incoming bytes */
|
|
|
|
__u64 outbytes; /* outgoing bytes */
|
|
|
|
};
|
|
|
|
/*
|
|
|
|
* Stats per cpu
|
|
|
|
*/
|
|
|
|
struct ip_vs_cpu_stats {
|
|
|
|
struct ip_vs_counters ustats;
|
|
|
|
struct u64_stats_sync syncp;
|
|
|
|
};
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
2008-08-10 12:24:41 -06:00
|
|
|
* IPVS statistics objects
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
2008-08-10 12:24:41 -06:00
|
|
|
struct ip_vs_estimator {
|
|
|
|
struct list_head list;
|
|
|
|
|
|
|
|
u64 last_inbytes;
|
|
|
|
u64 last_outbytes;
|
|
|
|
u32 last_conns;
|
|
|
|
u32 last_inpkts;
|
|
|
|
u32 last_outpkts;
|
|
|
|
|
|
|
|
u32 cps;
|
|
|
|
u32 inpps;
|
|
|
|
u32 outpps;
|
|
|
|
u32 inbps;
|
|
|
|
u32 outbps;
|
|
|
|
};
|
|
|
|
|
2009-11-02 20:26:03 -07:00
|
|
|
struct ip_vs_stats {
|
2011-01-03 06:44:56 -07:00
|
|
|
struct ip_vs_stats_user ustats; /* statistics */
|
2008-09-08 05:39:04 -06:00
|
|
|
struct ip_vs_estimator est; /* estimator */
|
2013-03-09 14:25:08 -07:00
|
|
|
struct ip_vs_cpu_stats __percpu *cpustats; /* per cpu counters */
|
2011-01-03 06:44:56 -07:00
|
|
|
spinlock_t lock; /* spin lock */
|
2011-03-13 17:37:49 -06:00
|
|
|
struct ip_vs_stats_user ustats0; /* reset values */
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2005-12-26 21:43:12 -07:00
|
|
|
struct dst_entry;
|
|
|
|
struct iphdr;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct ip_vs_conn;
|
|
|
|
struct ip_vs_app;
|
2005-12-26 21:43:12 -07:00
|
|
|
struct sk_buff;
|
2011-01-03 06:44:46 -07:00
|
|
|
struct ip_vs_proto_data;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
struct ip_vs_protocol {
|
|
|
|
struct ip_vs_protocol *next;
|
|
|
|
char *name;
|
2008-04-29 04:21:23 -06:00
|
|
|
u16 protocol;
|
|
|
|
u16 num_states;
|
2005-04-16 16:20:36 -06:00
|
|
|
int dont_defrag;
|
|
|
|
|
|
|
|
void (*init)(struct ip_vs_protocol *pp);
|
|
|
|
|
|
|
|
void (*exit)(struct ip_vs_protocol *pp);
|
|
|
|
|
2012-04-26 01:45:35 -06:00
|
|
|
int (*init_netns)(struct net *net, struct ip_vs_proto_data *pd);
|
2011-01-03 06:44:46 -07:00
|
|
|
|
|
|
|
void (*exit_netns)(struct net *net, struct ip_vs_proto_data *pd);
|
|
|
|
|
2008-09-02 07:55:40 -06:00
|
|
|
int (*conn_schedule)(int af, struct sk_buff *skb,
|
2011-01-03 06:44:51 -07:00
|
|
|
struct ip_vs_proto_data *pd,
|
ipvs: API change to avoid rescan of IPv6 exthdr
Reduce the number of times we scan/skip the IPv6 exthdrs.
This patch contains a lot of API changes. This is done, to avoid
repeating the scan of finding the IPv6 headers, via ipv6_find_hdr(),
which is called by ip_vs_fill_iph_skb().
Finding the IPv6 headers is done as early as possible, and passed on
as a pointer "struct ip_vs_iphdr *" to the affected functions.
This patch reduce/removes 19 calls to ip_vs_fill_iph_skb().
Notice, I have choosen, not to change the API of function
pointer "(*schedule)" (in struct ip_vs_scheduler) as it can be
used by external schedulers, via {un,}register_ip_vs_scheduler.
Only 4 out of 10 schedulers use info from ip_vs_iphdr*, and when
they do, they are only interested in iph->{s,d}addr.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:07:17 -06:00
|
|
|
int *verdict, struct ip_vs_conn **cpp,
|
|
|
|
struct ip_vs_iphdr *iph);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
struct ip_vs_conn *
|
2008-09-02 07:55:40 -06:00
|
|
|
(*conn_in_get)(int af,
|
|
|
|
const struct sk_buff *skb,
|
|
|
|
const struct ip_vs_iphdr *iph,
|
2005-04-16 16:20:36 -06:00
|
|
|
int inverse);
|
|
|
|
|
|
|
|
struct ip_vs_conn *
|
2008-09-02 07:55:40 -06:00
|
|
|
(*conn_out_get)(int af,
|
|
|
|
const struct sk_buff *skb,
|
|
|
|
const struct ip_vs_iphdr *iph,
|
2005-04-16 16:20:36 -06:00
|
|
|
int inverse);
|
|
|
|
|
ipvs: API change to avoid rescan of IPv6 exthdr
Reduce the number of times we scan/skip the IPv6 exthdrs.
This patch contains a lot of API changes. This is done, to avoid
repeating the scan of finding the IPv6 headers, via ipv6_find_hdr(),
which is called by ip_vs_fill_iph_skb().
Finding the IPv6 headers is done as early as possible, and passed on
as a pointer "struct ip_vs_iphdr *" to the affected functions.
This patch reduce/removes 19 calls to ip_vs_fill_iph_skb().
Notice, I have choosen, not to change the API of function
pointer "(*schedule)" (in struct ip_vs_scheduler) as it can be
used by external schedulers, via {un,}register_ip_vs_scheduler.
Only 4 out of 10 schedulers use info from ip_vs_iphdr*, and when
they do, they are only interested in iph->{s,d}addr.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:07:17 -06:00
|
|
|
int (*snat_handler)(struct sk_buff *skb, struct ip_vs_protocol *pp,
|
|
|
|
struct ip_vs_conn *cp, struct ip_vs_iphdr *iph);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
ipvs: API change to avoid rescan of IPv6 exthdr
Reduce the number of times we scan/skip the IPv6 exthdrs.
This patch contains a lot of API changes. This is done, to avoid
repeating the scan of finding the IPv6 headers, via ipv6_find_hdr(),
which is called by ip_vs_fill_iph_skb().
Finding the IPv6 headers is done as early as possible, and passed on
as a pointer "struct ip_vs_iphdr *" to the affected functions.
This patch reduce/removes 19 calls to ip_vs_fill_iph_skb().
Notice, I have choosen, not to change the API of function
pointer "(*schedule)" (in struct ip_vs_scheduler) as it can be
used by external schedulers, via {un,}register_ip_vs_scheduler.
Only 4 out of 10 schedulers use info from ip_vs_iphdr*, and when
they do, they are only interested in iph->{s,d}addr.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:07:17 -06:00
|
|
|
int (*dnat_handler)(struct sk_buff *skb, struct ip_vs_protocol *pp,
|
|
|
|
struct ip_vs_conn *cp, struct ip_vs_iphdr *iph);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-09-02 07:55:40 -06:00
|
|
|
int (*csum_check)(int af, struct sk_buff *skb,
|
|
|
|
struct ip_vs_protocol *pp);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
const char *(*state_name)(int state);
|
|
|
|
|
2011-09-15 23:11:49 -06:00
|
|
|
void (*state_transition)(struct ip_vs_conn *cp, int direction,
|
|
|
|
const struct sk_buff *skb,
|
|
|
|
struct ip_vs_proto_data *pd);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-01-03 06:44:53 -07:00
|
|
|
int (*register_app)(struct net *net, struct ip_vs_app *inc);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-01-03 06:44:53 -07:00
|
|
|
void (*unregister_app)(struct net *net, struct ip_vs_app *inc);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
int (*app_conn_bind)(struct ip_vs_conn *cp);
|
|
|
|
|
2010-10-17 07:46:17 -06:00
|
|
|
void (*debug_packet)(int af, struct ip_vs_protocol *pp,
|
2005-04-16 16:20:36 -06:00
|
|
|
const struct sk_buff *skb,
|
|
|
|
int offset,
|
|
|
|
const char *msg);
|
|
|
|
|
2011-01-03 06:44:51 -07:00
|
|
|
void (*timeout_change)(struct ip_vs_proto_data *pd, int flags);
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2011-01-03 06:44:46 -07:00
|
|
|
/*
|
|
|
|
* protocol data per netns
|
|
|
|
*/
|
|
|
|
struct ip_vs_proto_data {
|
|
|
|
struct ip_vs_proto_data *next;
|
|
|
|
struct ip_vs_protocol *pp;
|
|
|
|
int *timeout_table; /* protocol timeout table */
|
|
|
|
atomic_t appcnt; /* counter of proto app incs. */
|
|
|
|
struct tcp_states_t *tcp_state_table;
|
|
|
|
};
|
|
|
|
|
|
|
|
extern struct ip_vs_protocol *ip_vs_proto_get(unsigned short proto);
|
|
|
|
extern struct ip_vs_proto_data *ip_vs_proto_data_get(struct net *net,
|
|
|
|
unsigned short proto);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-08-22 06:37:52 -06:00
|
|
|
struct ip_vs_conn_param {
|
2011-01-03 06:44:57 -07:00
|
|
|
struct net *net;
|
2010-08-22 06:37:52 -06:00
|
|
|
const union nf_inet_addr *caddr;
|
|
|
|
const union nf_inet_addr *vaddr;
|
|
|
|
__be16 cport;
|
|
|
|
__be16 vport;
|
|
|
|
__u16 protocol;
|
|
|
|
u16 af;
|
2010-08-22 06:37:53 -06:00
|
|
|
|
|
|
|
const struct ip_vs_pe *pe;
|
|
|
|
char *pe_data;
|
|
|
|
__u8 pe_data_len;
|
2010-08-22 06:37:52 -06:00
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* IP_VS structure allocated for each dynamically scheduled connection
|
|
|
|
*/
|
|
|
|
struct ip_vs_conn {
|
2011-02-19 03:05:08 -07:00
|
|
|
struct hlist_node c_list; /* hashed list heads */
|
2005-04-16 16:20:36 -06:00
|
|
|
/* Protocol, addresses and port numbers */
|
2011-01-03 06:44:57 -07:00
|
|
|
__be16 cport;
|
|
|
|
__be16 dport;
|
2013-03-21 03:58:11 -06:00
|
|
|
__be16 vport;
|
|
|
|
u16 af; /* address family */
|
2011-01-03 06:44:57 -07:00
|
|
|
union nf_inet_addr caddr; /* client address */
|
|
|
|
union nf_inet_addr vaddr; /* virtual address */
|
|
|
|
union nf_inet_addr daddr; /* destination address */
|
|
|
|
volatile __u32 flags; /* status flags */
|
2005-04-16 16:20:36 -06:00
|
|
|
__u16 protocol; /* Which protocol (TCP/UDP) */
|
2013-03-21 03:58:11 -06:00
|
|
|
#ifdef CONFIG_NET_NS
|
|
|
|
struct net *net; /* Name space */
|
|
|
|
#endif
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* counter and timer */
|
|
|
|
atomic_t refcnt; /* reference count */
|
|
|
|
struct timer_list timer; /* Expiration timer */
|
|
|
|
volatile unsigned long timeout; /* timeout */
|
|
|
|
|
|
|
|
/* Flags and state transition */
|
|
|
|
spinlock_t lock; /* lock for state transition */
|
|
|
|
volatile __u16 state; /* state info */
|
2007-11-07 03:36:55 -07:00
|
|
|
volatile __u16 old_state; /* old state, to be used for
|
|
|
|
* state transition triggerd
|
|
|
|
* synchronization
|
|
|
|
*/
|
2013-03-21 03:58:11 -06:00
|
|
|
__u32 fwmark; /* Fire wall mark from skb */
|
2012-04-24 14:46:40 -06:00
|
|
|
unsigned long sync_endtime; /* jiffies + sent_retries */
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* Control members */
|
|
|
|
struct ip_vs_conn *control; /* Master control connection */
|
|
|
|
atomic_t n_control; /* Number of controlled ones */
|
|
|
|
struct ip_vs_dest *dest; /* real server */
|
|
|
|
atomic_t in_pkts; /* incoming packet counter */
|
|
|
|
|
|
|
|
/* packet transmitter for different forwarding methods. If it
|
|
|
|
mangles the packet, it must return NF_DROP or better NF_STOLEN,
|
|
|
|
otherwise this must be changed to a sk_buff **.
|
2010-10-17 07:38:15 -06:00
|
|
|
NF_ACCEPT can be returned when destination is local.
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
int (*packet_xmit)(struct sk_buff *skb, struct ip_vs_conn *cp,
|
ipvs: API change to avoid rescan of IPv6 exthdr
Reduce the number of times we scan/skip the IPv6 exthdrs.
This patch contains a lot of API changes. This is done, to avoid
repeating the scan of finding the IPv6 headers, via ipv6_find_hdr(),
which is called by ip_vs_fill_iph_skb().
Finding the IPv6 headers is done as early as possible, and passed on
as a pointer "struct ip_vs_iphdr *" to the affected functions.
This patch reduce/removes 19 calls to ip_vs_fill_iph_skb().
Notice, I have choosen, not to change the API of function
pointer "(*schedule)" (in struct ip_vs_scheduler) as it can be
used by external schedulers, via {un,}register_ip_vs_scheduler.
Only 4 out of 10 schedulers use info from ip_vs_iphdr*, and when
they do, they are only interested in iph->{s,d}addr.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:07:17 -06:00
|
|
|
struct ip_vs_protocol *pp, struct ip_vs_iphdr *iph);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* Note: we can group the following members into a structure,
|
|
|
|
in order to save more space, and the following members are
|
|
|
|
only used in VS/NAT anyway */
|
|
|
|
struct ip_vs_app *app; /* bound ip_vs_app object */
|
|
|
|
void *app_data; /* Application private data */
|
|
|
|
struct ip_vs_seq in_seq; /* incoming seq. struct */
|
|
|
|
struct ip_vs_seq out_seq; /* outgoing seq. struct */
|
2010-08-22 06:37:53 -06:00
|
|
|
|
2010-11-08 04:05:57 -07:00
|
|
|
const struct ip_vs_pe *pe;
|
2010-08-22 06:37:53 -06:00
|
|
|
char *pe_data;
|
|
|
|
__u8 pe_data_len;
|
2013-03-21 03:58:10 -06:00
|
|
|
|
|
|
|
struct rcu_head rcu_head;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2011-01-03 06:44:57 -07:00
|
|
|
/*
|
|
|
|
* To save some memory in conn table when name space is disabled.
|
|
|
|
*/
|
|
|
|
static inline struct net *ip_vs_conn_net(const struct ip_vs_conn *cp)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_NET_NS
|
|
|
|
return cp->net;
|
|
|
|
#else
|
|
|
|
return &init_net;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
static inline void ip_vs_conn_net_set(struct ip_vs_conn *cp, struct net *net)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_NET_NS
|
|
|
|
cp->net = net;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int ip_vs_conn_net_eq(const struct ip_vs_conn *cp,
|
|
|
|
struct net *net)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_NET_NS
|
|
|
|
return cp->net == net;
|
|
|
|
#else
|
|
|
|
return 1;
|
|
|
|
#endif
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-09-02 07:55:36 -06:00
|
|
|
/*
|
|
|
|
* Extended internal versions of struct ip_vs_service_user and
|
|
|
|
* ip_vs_dest_user for IPv6 support.
|
|
|
|
*
|
|
|
|
* We need these to conveniently pass around service and destination
|
|
|
|
* options, but unfortunately, we also need to keep the old definitions to
|
|
|
|
* maintain userspace backwards compatibility for the setsockopt interface.
|
|
|
|
*/
|
|
|
|
struct ip_vs_service_user_kern {
|
|
|
|
/* virtual service addresses */
|
|
|
|
u16 af;
|
|
|
|
u16 protocol;
|
|
|
|
union nf_inet_addr addr; /* virtual ip address */
|
2013-04-17 14:50:49 -06:00
|
|
|
__be16 port;
|
2008-09-02 07:55:36 -06:00
|
|
|
u32 fwmark; /* firwall mark of service */
|
|
|
|
|
|
|
|
/* virtual service options */
|
|
|
|
char *sched_name;
|
2010-08-22 06:37:54 -06:00
|
|
|
char *pe_name;
|
2012-04-14 23:58:06 -06:00
|
|
|
unsigned int flags; /* virtual service flags */
|
|
|
|
unsigned int timeout; /* persistent timeout in sec */
|
2013-04-17 14:50:49 -06:00
|
|
|
__be32 netmask; /* persistent netmask or plen */
|
2008-09-02 07:55:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
struct ip_vs_dest_user_kern {
|
|
|
|
/* destination server address */
|
|
|
|
union nf_inet_addr addr;
|
2013-04-17 14:50:49 -06:00
|
|
|
__be16 port;
|
2008-09-02 07:55:36 -06:00
|
|
|
|
|
|
|
/* real server options */
|
2012-04-14 23:58:06 -06:00
|
|
|
unsigned int conn_flags; /* connection flags */
|
2008-09-02 07:55:36 -06:00
|
|
|
int weight; /* destination weight */
|
|
|
|
|
|
|
|
/* thresholds for active connections */
|
|
|
|
u32 u_threshold; /* upper threshold */
|
|
|
|
u32 l_threshold; /* lower threshold */
|
|
|
|
};
|
|
|
|
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* The information about the virtual service offered to the net
|
|
|
|
* and the forwarding entries
|
|
|
|
*/
|
|
|
|
struct ip_vs_service {
|
2013-03-22 03:46:53 -06:00
|
|
|
struct hlist_node s_list; /* for normal service table */
|
|
|
|
struct hlist_node f_list; /* for fwmark-based service table */
|
2005-04-16 16:20:36 -06:00
|
|
|
atomic_t refcnt; /* reference counter */
|
|
|
|
|
2008-09-02 07:55:33 -06:00
|
|
|
u16 af; /* address family */
|
2005-04-16 16:20:36 -06:00
|
|
|
__u16 protocol; /* which protocol (TCP/UDP) */
|
2008-09-02 07:55:33 -06:00
|
|
|
union nf_inet_addr addr; /* IP address for virtual service */
|
2006-09-28 15:29:52 -06:00
|
|
|
__be16 port; /* port number for the service */
|
2005-04-16 16:20:36 -06:00
|
|
|
__u32 fwmark; /* firewall mark of the service */
|
2012-04-14 23:58:06 -06:00
|
|
|
unsigned int flags; /* service status flags */
|
|
|
|
unsigned int timeout; /* persistent timeout in ticks */
|
2013-04-17 14:50:49 -06:00
|
|
|
__be32 netmask; /* grouping granularity, mask/plen */
|
2011-01-03 06:44:43 -07:00
|
|
|
struct net *net;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
struct list_head destinations; /* real server d-linked list */
|
|
|
|
__u32 num_dests; /* number of servers */
|
|
|
|
struct ip_vs_stats stats; /* statistics for the service */
|
|
|
|
|
|
|
|
/* for scheduling */
|
2013-03-22 03:46:53 -06:00
|
|
|
struct ip_vs_scheduler __rcu *scheduler; /* bound scheduler object */
|
2013-03-22 03:46:51 -06:00
|
|
|
spinlock_t sched_lock; /* lock sched_data */
|
2005-04-16 16:20:36 -06:00
|
|
|
void *sched_data; /* scheduler application data */
|
2010-08-22 06:37:53 -06:00
|
|
|
|
|
|
|
/* alternate persistence engine */
|
2013-03-22 03:46:53 -06:00
|
|
|
struct ip_vs_pe __rcu *pe;
|
|
|
|
|
|
|
|
struct rcu_head rcu_head;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
2013-03-21 03:58:06 -06:00
|
|
|
/* Information for cached dst */
|
|
|
|
struct ip_vs_dest_dst {
|
|
|
|
struct dst_entry *dst_cache; /* destination cache entry */
|
|
|
|
u32 dst_cookie;
|
|
|
|
union nf_inet_addr dst_saddr;
|
|
|
|
struct rcu_head rcu_head;
|
|
|
|
};
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2013-03-22 03:46:49 -06:00
|
|
|
/* In grace period after removing */
|
|
|
|
#define IP_VS_DEST_STATE_REMOVING 0x01
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* The real server destination forwarding entry
|
|
|
|
* with ip address, port number, and so on.
|
|
|
|
*/
|
|
|
|
struct ip_vs_dest {
|
|
|
|
struct list_head n_list; /* for the dests in the service */
|
2013-03-21 03:58:08 -06:00
|
|
|
struct hlist_node d_list; /* for table with all the dests */
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-09-02 07:55:33 -06:00
|
|
|
u16 af; /* address family */
|
2006-09-28 15:29:52 -06:00
|
|
|
__be16 port; /* port number of the server */
|
2011-01-03 06:45:00 -07:00
|
|
|
union nf_inet_addr addr; /* IP address of the server */
|
2012-04-14 23:58:06 -06:00
|
|
|
volatile unsigned int flags; /* dest status flags */
|
2005-04-16 16:20:36 -06:00
|
|
|
atomic_t conn_flags; /* flags to copy to conn */
|
|
|
|
atomic_t weight; /* server weight */
|
|
|
|
|
|
|
|
atomic_t refcnt; /* reference counter */
|
|
|
|
struct ip_vs_stats stats; /* statistics */
|
2013-03-22 03:46:49 -06:00
|
|
|
unsigned long state; /* state flags */
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* connection counters and thresholds */
|
|
|
|
atomic_t activeconns; /* active connections */
|
|
|
|
atomic_t inactconns; /* inactive connections */
|
|
|
|
atomic_t persistconns; /* persistent connections */
|
|
|
|
__u32 u_threshold; /* upper threshold */
|
|
|
|
__u32 l_threshold; /* lower threshold */
|
|
|
|
|
|
|
|
/* for destination cache */
|
|
|
|
spinlock_t dst_lock; /* lock of dst_cache */
|
2013-03-21 03:58:06 -06:00
|
|
|
struct ip_vs_dest_dst __rcu *dest_dst; /* cached dst info */
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* for virtual service */
|
|
|
|
struct ip_vs_service *svc; /* service it belongs to */
|
|
|
|
__u16 protocol; /* which protocol (TCP/UDP) */
|
2006-09-28 15:29:52 -06:00
|
|
|
__be16 vport; /* virtual port number */
|
2011-01-03 06:45:00 -07:00
|
|
|
union nf_inet_addr vaddr; /* virtual IP address */
|
2005-04-16 16:20:36 -06:00
|
|
|
__u32 vfwmark; /* firewall mark of service */
|
2013-03-21 03:58:08 -06:00
|
|
|
|
2013-03-22 03:46:49 -06:00
|
|
|
struct list_head t_list; /* in dest_trash */
|
2013-03-21 03:58:08 -06:00
|
|
|
struct rcu_head rcu_head;
|
|
|
|
unsigned int in_rs_table:1; /* we are in rs_table */
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The scheduler object
|
|
|
|
*/
|
|
|
|
struct ip_vs_scheduler {
|
|
|
|
struct list_head n_list; /* d-linked list head */
|
|
|
|
char *name; /* scheduler name */
|
|
|
|
atomic_t refcnt; /* reference counter */
|
|
|
|
struct module *module; /* THIS_MODULE/NULL */
|
|
|
|
|
|
|
|
/* scheduler initializing service */
|
|
|
|
int (*init_service)(struct ip_vs_service *svc);
|
|
|
|
/* scheduling service finish */
|
2013-03-22 03:46:50 -06:00
|
|
|
void (*done_service)(struct ip_vs_service *svc);
|
2013-03-22 03:46:37 -06:00
|
|
|
/* dest is linked */
|
|
|
|
int (*add_dest)(struct ip_vs_service *svc, struct ip_vs_dest *dest);
|
|
|
|
/* dest is unlinked */
|
|
|
|
int (*del_dest)(struct ip_vs_service *svc, struct ip_vs_dest *dest);
|
|
|
|
/* dest is updated */
|
|
|
|
int (*upd_dest)(struct ip_vs_service *svc, struct ip_vs_dest *dest);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* selecting a server from the given service */
|
|
|
|
struct ip_vs_dest* (*schedule)(struct ip_vs_service *svc,
|
|
|
|
const struct sk_buff *skb);
|
|
|
|
};
|
|
|
|
|
2010-08-22 06:37:53 -06:00
|
|
|
/* The persistence engine object */
|
|
|
|
struct ip_vs_pe {
|
|
|
|
struct list_head n_list; /* d-linked list head */
|
|
|
|
char *name; /* scheduler name */
|
|
|
|
atomic_t refcnt; /* reference counter */
|
|
|
|
struct module *module; /* THIS_MODULE/NULL */
|
|
|
|
|
|
|
|
/* get the connection template, if any */
|
|
|
|
int (*fill_param)(struct ip_vs_conn_param *p, struct sk_buff *skb);
|
|
|
|
bool (*ct_match)(const struct ip_vs_conn_param *p,
|
|
|
|
struct ip_vs_conn *ct);
|
|
|
|
u32 (*hashkey_raw)(const struct ip_vs_conn_param *p, u32 initval,
|
|
|
|
bool inverse);
|
2010-08-22 06:37:53 -06:00
|
|
|
int (*show_pe_data)(const struct ip_vs_conn *cp, char *buf);
|
2010-08-22 06:37:53 -06:00
|
|
|
};
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* The application module object (a.k.a. app incarnation)
|
|
|
|
*/
|
2009-11-02 20:26:03 -07:00
|
|
|
struct ip_vs_app {
|
2005-04-16 16:20:36 -06:00
|
|
|
struct list_head a_list; /* member in app list */
|
|
|
|
int type; /* IP_VS_APP_TYPE_xxx */
|
|
|
|
char *name; /* application module name */
|
|
|
|
__u16 protocol;
|
|
|
|
struct module *module; /* THIS_MODULE/NULL */
|
|
|
|
struct list_head incs_list; /* list of incarnations */
|
|
|
|
|
|
|
|
/* members for application incarnations */
|
|
|
|
struct list_head p_list; /* member in proto app list */
|
|
|
|
struct ip_vs_app *app; /* its real application */
|
2006-09-28 15:29:52 -06:00
|
|
|
__be16 port; /* port number in net order */
|
2005-04-16 16:20:36 -06:00
|
|
|
atomic_t usecnt; /* usage counter */
|
2013-03-21 03:58:07 -06:00
|
|
|
struct rcu_head rcu_head;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-10-17 07:17:20 -06:00
|
|
|
/*
|
|
|
|
* output hook: Process packet in inout direction, diff set for TCP.
|
|
|
|
* Return: 0=Error, 1=Payload Not Mangled/Mangled but checksum is ok,
|
|
|
|
* 2=Mangled but checksum was not updated
|
|
|
|
*/
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*pkt_out)(struct ip_vs_app *, struct ip_vs_conn *,
|
2007-10-15 01:53:15 -06:00
|
|
|
struct sk_buff *, int *diff);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-10-17 07:17:20 -06:00
|
|
|
/*
|
|
|
|
* input hook: Process packet in outin direction, diff set for TCP.
|
|
|
|
* Return: 0=Error, 1=Payload Not Mangled/Mangled but checksum is ok,
|
|
|
|
* 2=Mangled but checksum was not updated
|
|
|
|
*/
|
2005-04-16 16:20:36 -06:00
|
|
|
int (*pkt_in)(struct ip_vs_app *, struct ip_vs_conn *,
|
2007-10-15 01:53:15 -06:00
|
|
|
struct sk_buff *, int *diff);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* ip_vs_app initializer */
|
|
|
|
int (*init_conn)(struct ip_vs_app *, struct ip_vs_conn *);
|
|
|
|
|
|
|
|
/* ip_vs_app finish */
|
|
|
|
int (*done_conn)(struct ip_vs_app *, struct ip_vs_conn *);
|
|
|
|
|
|
|
|
|
|
|
|
/* not used now */
|
|
|
|
int (*bind_conn)(struct ip_vs_app *, struct ip_vs_conn *,
|
|
|
|
struct ip_vs_protocol *);
|
|
|
|
|
|
|
|
void (*unbind_conn)(struct ip_vs_app *, struct ip_vs_conn *);
|
|
|
|
|
|
|
|
int * timeout_table;
|
|
|
|
int * timeouts;
|
|
|
|
int timeouts_size;
|
|
|
|
|
|
|
|
int (*conn_schedule)(struct sk_buff *skb, struct ip_vs_app *app,
|
|
|
|
int *verdict, struct ip_vs_conn **cpp);
|
|
|
|
|
|
|
|
struct ip_vs_conn *
|
|
|
|
(*conn_in_get)(const struct sk_buff *skb, struct ip_vs_app *app,
|
ipvs: API change to avoid rescan of IPv6 exthdr
Reduce the number of times we scan/skip the IPv6 exthdrs.
This patch contains a lot of API changes. This is done, to avoid
repeating the scan of finding the IPv6 headers, via ipv6_find_hdr(),
which is called by ip_vs_fill_iph_skb().
Finding the IPv6 headers is done as early as possible, and passed on
as a pointer "struct ip_vs_iphdr *" to the affected functions.
This patch reduce/removes 19 calls to ip_vs_fill_iph_skb().
Notice, I have choosen, not to change the API of function
pointer "(*schedule)" (in struct ip_vs_scheduler) as it can be
used by external schedulers, via {un,}register_ip_vs_scheduler.
Only 4 out of 10 schedulers use info from ip_vs_iphdr*, and when
they do, they are only interested in iph->{s,d}addr.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:07:17 -06:00
|
|
|
const struct iphdr *iph, int inverse);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
struct ip_vs_conn *
|
|
|
|
(*conn_out_get)(const struct sk_buff *skb, struct ip_vs_app *app,
|
ipvs: API change to avoid rescan of IPv6 exthdr
Reduce the number of times we scan/skip the IPv6 exthdrs.
This patch contains a lot of API changes. This is done, to avoid
repeating the scan of finding the IPv6 headers, via ipv6_find_hdr(),
which is called by ip_vs_fill_iph_skb().
Finding the IPv6 headers is done as early as possible, and passed on
as a pointer "struct ip_vs_iphdr *" to the affected functions.
This patch reduce/removes 19 calls to ip_vs_fill_iph_skb().
Notice, I have choosen, not to change the API of function
pointer "(*schedule)" (in struct ip_vs_scheduler) as it can be
used by external schedulers, via {un,}register_ip_vs_scheduler.
Only 4 out of 10 schedulers use info from ip_vs_iphdr*, and when
they do, they are only interested in iph->{s,d}addr.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:07:17 -06:00
|
|
|
const struct iphdr *iph, int inverse);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
int (*state_transition)(struct ip_vs_conn *cp, int direction,
|
|
|
|
const struct sk_buff *skb,
|
|
|
|
struct ip_vs_app *app);
|
|
|
|
|
|
|
|
void (*timeout_change)(struct ip_vs_app *app, int flags);
|
|
|
|
};
|
|
|
|
|
2012-05-08 11:40:30 -06:00
|
|
|
struct ipvs_master_sync_state {
|
|
|
|
struct list_head sync_queue;
|
|
|
|
struct ip_vs_sync_buff *sync_buff;
|
|
|
|
int sync_queue_len;
|
|
|
|
unsigned int sync_queue_delay;
|
|
|
|
struct task_struct *master_thread;
|
|
|
|
struct delayed_work master_wakeup_work;
|
|
|
|
struct netns_ipvs *ipvs;
|
|
|
|
};
|
|
|
|
|
2013-03-22 03:46:49 -06:00
|
|
|
/* How much time to keep dests in trash */
|
|
|
|
#define IP_VS_DEST_TRASH_PERIOD (120 * HZ)
|
|
|
|
|
2011-03-04 03:18:07 -07:00
|
|
|
/* IPVS in network namespace */
|
|
|
|
struct netns_ipvs {
|
|
|
|
int gen; /* Generation */
|
2011-05-03 14:09:31 -06:00
|
|
|
int enable; /* enable like nf_hooks do */
|
2011-03-04 03:18:07 -07:00
|
|
|
/*
|
|
|
|
* Hash table: for real service lookups
|
|
|
|
*/
|
|
|
|
#define IP_VS_RTAB_BITS 4
|
|
|
|
#define IP_VS_RTAB_SIZE (1 << IP_VS_RTAB_BITS)
|
|
|
|
#define IP_VS_RTAB_MASK (IP_VS_RTAB_SIZE - 1)
|
|
|
|
|
2013-03-21 03:58:08 -06:00
|
|
|
struct hlist_head rs_table[IP_VS_RTAB_SIZE];
|
2011-03-04 03:18:07 -07:00
|
|
|
/* ip_vs_app */
|
|
|
|
struct list_head app_list;
|
|
|
|
/* ip_vs_proto */
|
|
|
|
#define IP_VS_PROTO_TAB_SIZE 32 /* must be power of 2 */
|
|
|
|
struct ip_vs_proto_data *proto_data_table[IP_VS_PROTO_TAB_SIZE];
|
|
|
|
/* ip_vs_proto_tcp */
|
|
|
|
#ifdef CONFIG_IP_VS_PROTO_TCP
|
|
|
|
#define TCP_APP_TAB_BITS 4
|
|
|
|
#define TCP_APP_TAB_SIZE (1 << TCP_APP_TAB_BITS)
|
|
|
|
#define TCP_APP_TAB_MASK (TCP_APP_TAB_SIZE - 1)
|
|
|
|
struct list_head tcp_apps[TCP_APP_TAB_SIZE];
|
|
|
|
#endif
|
|
|
|
/* ip_vs_proto_udp */
|
|
|
|
#ifdef CONFIG_IP_VS_PROTO_UDP
|
|
|
|
#define UDP_APP_TAB_BITS 4
|
|
|
|
#define UDP_APP_TAB_SIZE (1 << UDP_APP_TAB_BITS)
|
|
|
|
#define UDP_APP_TAB_MASK (UDP_APP_TAB_SIZE - 1)
|
|
|
|
struct list_head udp_apps[UDP_APP_TAB_SIZE];
|
|
|
|
#endif
|
|
|
|
/* ip_vs_proto_sctp */
|
|
|
|
#ifdef CONFIG_IP_VS_PROTO_SCTP
|
|
|
|
#define SCTP_APP_TAB_BITS 4
|
|
|
|
#define SCTP_APP_TAB_SIZE (1 << SCTP_APP_TAB_BITS)
|
|
|
|
#define SCTP_APP_TAB_MASK (SCTP_APP_TAB_SIZE - 1)
|
|
|
|
/* Hash table for SCTP application incarnations */
|
|
|
|
struct list_head sctp_apps[SCTP_APP_TAB_SIZE];
|
|
|
|
#endif
|
|
|
|
/* ip_vs_conn */
|
|
|
|
atomic_t conn_count; /* connection counter */
|
|
|
|
|
|
|
|
/* ip_vs_ctl */
|
2011-03-04 03:20:35 -07:00
|
|
|
struct ip_vs_stats tot_stats; /* Statistics & est. */
|
2011-03-04 03:18:07 -07:00
|
|
|
|
|
|
|
int num_services; /* no of virtual services */
|
2011-02-04 02:33:02 -07:00
|
|
|
|
|
|
|
/* Trash for destinations */
|
|
|
|
struct list_head dest_trash;
|
2013-03-22 03:46:49 -06:00
|
|
|
spinlock_t dest_trash_lock;
|
|
|
|
struct timer_list dest_trash_timer; /* expiration timer */
|
2011-02-04 02:33:02 -07:00
|
|
|
/* Service counters */
|
|
|
|
atomic_t ftpsvc_counter;
|
|
|
|
atomic_t nullsvc_counter;
|
|
|
|
|
|
|
|
#ifdef CONFIG_SYSCTL
|
2011-03-04 03:18:07 -07:00
|
|
|
/* 1/rate drop and drop-entry variables */
|
|
|
|
struct delayed_work defense_work; /* Work handler */
|
|
|
|
int drop_rate;
|
|
|
|
int drop_counter;
|
|
|
|
atomic_t dropentry;
|
|
|
|
/* locks in ctl.c */
|
|
|
|
spinlock_t dropentry_lock; /* drop entry handling */
|
|
|
|
spinlock_t droppacket_lock; /* drop packet handling */
|
|
|
|
spinlock_t securetcp_lock; /* state and timeout tables */
|
|
|
|
|
|
|
|
/* sys-ctl struct */
|
|
|
|
struct ctl_table_header *sysctl_hdr;
|
|
|
|
struct ctl_table *sysctl_tbl;
|
2011-02-04 02:33:02 -07:00
|
|
|
#endif
|
|
|
|
|
2011-03-04 03:18:07 -07:00
|
|
|
/* sysctl variables */
|
|
|
|
int sysctl_amemthresh;
|
|
|
|
int sysctl_am_droprate;
|
|
|
|
int sysctl_drop_entry;
|
|
|
|
int sysctl_drop_packet;
|
|
|
|
int sysctl_secure_tcp;
|
|
|
|
#ifdef CONFIG_IP_VS_NFCT
|
|
|
|
int sysctl_conntrack;
|
|
|
|
#endif
|
|
|
|
int sysctl_snat_reroute;
|
|
|
|
int sysctl_sync_ver;
|
2012-05-08 11:40:30 -06:00
|
|
|
int sysctl_sync_ports;
|
2012-05-08 11:39:49 -06:00
|
|
|
int sysctl_sync_qlen_max;
|
|
|
|
int sysctl_sync_sock_size;
|
2011-03-04 03:18:07 -07:00
|
|
|
int sysctl_cache_bypass;
|
|
|
|
int sysctl_expire_nodest_conn;
|
|
|
|
int sysctl_expire_quiescent_template;
|
|
|
|
int sysctl_sync_threshold[2];
|
2012-04-24 14:46:40 -06:00
|
|
|
unsigned int sysctl_sync_refresh_period;
|
|
|
|
int sysctl_sync_retries;
|
2011-03-04 03:18:07 -07:00
|
|
|
int sysctl_nat_icmp_send;
|
2012-07-20 02:59:53 -06:00
|
|
|
int sysctl_pmtu_disc;
|
2013-03-09 14:25:04 -07:00
|
|
|
int sysctl_backup_only;
|
2011-03-04 03:18:07 -07:00
|
|
|
|
|
|
|
/* ip_vs_lblc */
|
|
|
|
int sysctl_lblc_expiration;
|
|
|
|
struct ctl_table_header *lblc_ctl_header;
|
|
|
|
struct ctl_table *lblc_ctl_table;
|
|
|
|
/* ip_vs_lblcr */
|
|
|
|
int sysctl_lblcr_expiration;
|
|
|
|
struct ctl_table_header *lblcr_ctl_header;
|
|
|
|
struct ctl_table *lblcr_ctl_table;
|
|
|
|
/* ip_vs_est */
|
|
|
|
struct list_head est_list; /* estimator list */
|
|
|
|
spinlock_t est_lock;
|
|
|
|
struct timer_list est_timer; /* Estimation timer */
|
|
|
|
/* ip_vs_sync */
|
|
|
|
spinlock_t sync_lock;
|
2012-05-08 11:40:30 -06:00
|
|
|
struct ipvs_master_sync_state *ms;
|
2011-03-04 03:18:07 -07:00
|
|
|
spinlock_t sync_buff_lock;
|
2012-05-08 11:40:30 -06:00
|
|
|
struct task_struct **backup_threads;
|
|
|
|
int threads_mask;
|
2011-03-04 03:18:07 -07:00
|
|
|
int send_mesg_maxlen;
|
|
|
|
int recv_mesg_maxlen;
|
|
|
|
volatile int sync_state;
|
|
|
|
volatile int master_syncid;
|
|
|
|
volatile int backup_syncid;
|
2011-10-10 19:54:35 -06:00
|
|
|
struct mutex sync_mutex;
|
2011-03-04 03:18:07 -07:00
|
|
|
/* multicast interface name */
|
|
|
|
char master_mcast_ifn[IP_VS_IFNAME_MAXLEN];
|
|
|
|
char backup_mcast_ifn[IP_VS_IFNAME_MAXLEN];
|
|
|
|
/* net name space ptr */
|
|
|
|
struct net *net; /* Needed by timer routines */
|
|
|
|
};
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-02-04 02:33:01 -07:00
|
|
|
#define DEFAULT_SYNC_THRESHOLD 3
|
|
|
|
#define DEFAULT_SYNC_PERIOD 50
|
2011-02-04 02:33:01 -07:00
|
|
|
#define DEFAULT_SYNC_VER 1
|
2012-04-24 14:46:40 -06:00
|
|
|
#define DEFAULT_SYNC_REFRESH_PERIOD (0U * HZ)
|
|
|
|
#define DEFAULT_SYNC_RETRIES 0
|
2012-05-08 11:39:49 -06:00
|
|
|
#define IPVS_SYNC_WAKEUP_RATE 8
|
|
|
|
#define IPVS_SYNC_QLEN_MAX (IPVS_SYNC_WAKEUP_RATE * 4)
|
|
|
|
#define IPVS_SYNC_SEND_DELAY (HZ / 50)
|
|
|
|
#define IPVS_SYNC_CHECK_PERIOD HZ
|
2012-04-24 14:46:40 -06:00
|
|
|
#define IPVS_SYNC_FLUSH_TIME (HZ * 2)
|
2012-05-08 11:40:30 -06:00
|
|
|
#define IPVS_SYNC_PORTS_MAX (1 << 6)
|
2011-02-04 02:33:01 -07:00
|
|
|
|
|
|
|
#ifdef CONFIG_SYSCTL
|
|
|
|
|
|
|
|
static inline int sysctl_sync_threshold(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return ipvs->sysctl_sync_threshold[0];
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int sysctl_sync_period(struct netns_ipvs *ipvs)
|
|
|
|
{
|
2012-04-24 14:46:40 -06:00
|
|
|
return ACCESS_ONCE(ipvs->sysctl_sync_threshold[1]);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline unsigned int sysctl_sync_refresh_period(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return ACCESS_ONCE(ipvs->sysctl_sync_refresh_period);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int sysctl_sync_retries(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return ipvs->sysctl_sync_retries;
|
2011-02-04 02:33:01 -07:00
|
|
|
}
|
|
|
|
|
2011-02-04 02:33:01 -07:00
|
|
|
static inline int sysctl_sync_ver(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return ipvs->sysctl_sync_ver;
|
|
|
|
}
|
|
|
|
|
2012-05-08 11:40:30 -06:00
|
|
|
static inline int sysctl_sync_ports(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return ACCESS_ONCE(ipvs->sysctl_sync_ports);
|
|
|
|
}
|
|
|
|
|
2012-05-08 11:39:49 -06:00
|
|
|
static inline int sysctl_sync_qlen_max(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return ipvs->sysctl_sync_qlen_max;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int sysctl_sync_sock_size(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return ipvs->sysctl_sync_sock_size;
|
|
|
|
}
|
|
|
|
|
2012-07-20 02:59:53 -06:00
|
|
|
static inline int sysctl_pmtu_disc(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return ipvs->sysctl_pmtu_disc;
|
|
|
|
}
|
|
|
|
|
2013-03-09 14:25:04 -07:00
|
|
|
static inline int sysctl_backup_only(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return ipvs->sync_state & IP_VS_STATE_BACKUP &&
|
|
|
|
ipvs->sysctl_backup_only;
|
|
|
|
}
|
|
|
|
|
2011-02-04 02:33:01 -07:00
|
|
|
#else
|
|
|
|
|
|
|
|
static inline int sysctl_sync_threshold(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return DEFAULT_SYNC_THRESHOLD;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int sysctl_sync_period(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return DEFAULT_SYNC_PERIOD;
|
|
|
|
}
|
|
|
|
|
2012-04-24 14:46:40 -06:00
|
|
|
static inline unsigned int sysctl_sync_refresh_period(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return DEFAULT_SYNC_REFRESH_PERIOD;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int sysctl_sync_retries(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return DEFAULT_SYNC_RETRIES & 3;
|
|
|
|
}
|
|
|
|
|
2011-02-04 02:33:01 -07:00
|
|
|
static inline int sysctl_sync_ver(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return DEFAULT_SYNC_VER;
|
|
|
|
}
|
|
|
|
|
2012-05-08 11:40:30 -06:00
|
|
|
static inline int sysctl_sync_ports(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2012-05-08 11:39:49 -06:00
|
|
|
static inline int sysctl_sync_qlen_max(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return IPVS_SYNC_QLEN_MAX;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int sysctl_sync_sock_size(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2012-07-20 02:59:53 -06:00
|
|
|
static inline int sysctl_pmtu_disc(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2013-03-09 14:25:04 -07:00
|
|
|
static inline int sysctl_backup_only(struct netns_ipvs *ipvs)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-02-04 02:33:01 -07:00
|
|
|
#endif
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* IPVS core functions
|
|
|
|
* (from ip_vs_core.c)
|
|
|
|
*/
|
2012-04-14 23:58:06 -06:00
|
|
|
extern const char *ip_vs_proto_name(unsigned int proto);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern void ip_vs_init_hash_table(struct list_head *table, int rows);
|
2008-08-10 03:18:01 -06:00
|
|
|
#define IP_VS_INIT_HASH_TABLE(t) ip_vs_init_hash_table((t), ARRAY_SIZE((t)))
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
#define IP_VS_APP_TYPE_FTP 1
|
|
|
|
|
|
|
|
/*
|
|
|
|
* ip_vs_conn handling functions
|
|
|
|
* (from ip_vs_conn.c)
|
|
|
|
*/
|
|
|
|
|
|
|
|
enum {
|
|
|
|
IP_VS_DIR_INPUT = 0,
|
|
|
|
IP_VS_DIR_OUTPUT,
|
|
|
|
IP_VS_DIR_INPUT_ONLY,
|
|
|
|
IP_VS_DIR_LAST,
|
|
|
|
};
|
|
|
|
|
2011-01-03 06:44:57 -07:00
|
|
|
static inline void ip_vs_conn_fill_param(struct net *net, int af, int protocol,
|
2010-08-22 06:37:52 -06:00
|
|
|
const union nf_inet_addr *caddr,
|
|
|
|
__be16 cport,
|
|
|
|
const union nf_inet_addr *vaddr,
|
|
|
|
__be16 vport,
|
|
|
|
struct ip_vs_conn_param *p)
|
|
|
|
{
|
2011-01-03 06:44:57 -07:00
|
|
|
p->net = net;
|
2010-08-22 06:37:52 -06:00
|
|
|
p->af = af;
|
|
|
|
p->protocol = protocol;
|
|
|
|
p->caddr = caddr;
|
|
|
|
p->cport = cport;
|
|
|
|
p->vaddr = vaddr;
|
|
|
|
p->vport = vport;
|
2010-08-22 06:37:53 -06:00
|
|
|
p->pe = NULL;
|
|
|
|
p->pe_data = NULL;
|
2010-08-22 06:37:52 -06:00
|
|
|
}
|
2008-09-02 07:55:43 -06:00
|
|
|
|
2010-08-22 06:37:52 -06:00
|
|
|
struct ip_vs_conn *ip_vs_conn_in_get(const struct ip_vs_conn_param *p);
|
|
|
|
struct ip_vs_conn *ip_vs_ct_in_get(const struct ip_vs_conn_param *p);
|
2008-09-02 07:55:43 -06:00
|
|
|
|
2010-08-02 09:12:44 -06:00
|
|
|
struct ip_vs_conn * ip_vs_conn_in_get_proto(int af, const struct sk_buff *skb,
|
|
|
|
const struct ip_vs_iphdr *iph,
|
|
|
|
int inverse);
|
|
|
|
|
2010-08-22 06:37:52 -06:00
|
|
|
struct ip_vs_conn *ip_vs_conn_out_get(const struct ip_vs_conn_param *p);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-08-02 09:12:44 -06:00
|
|
|
struct ip_vs_conn * ip_vs_conn_out_get_proto(int af, const struct sk_buff *skb,
|
|
|
|
const struct ip_vs_iphdr *iph,
|
|
|
|
int inverse);
|
|
|
|
|
2013-03-21 03:58:10 -06:00
|
|
|
/* Get reference to gain full access to conn.
|
|
|
|
* By default, RCU read-side critical sections have access only to
|
|
|
|
* conn fields and its PE data, see ip_vs_conn_rcu_free() for reference.
|
|
|
|
*/
|
|
|
|
static inline bool __ip_vs_conn_get(struct ip_vs_conn *cp)
|
|
|
|
{
|
|
|
|
return atomic_inc_not_zero(&cp->refcnt);
|
|
|
|
}
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/* put back the conn without restarting its timer */
|
|
|
|
static inline void __ip_vs_conn_put(struct ip_vs_conn *cp)
|
|
|
|
{
|
2013-03-21 03:58:10 -06:00
|
|
|
smp_mb__before_atomic_dec();
|
2005-04-16 16:20:36 -06:00
|
|
|
atomic_dec(&cp->refcnt);
|
|
|
|
}
|
|
|
|
extern void ip_vs_conn_put(struct ip_vs_conn *cp);
|
2006-09-28 15:29:52 -06:00
|
|
|
extern void ip_vs_conn_fill_cport(struct ip_vs_conn *cp, __be16 cport);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-08-22 06:37:52 -06:00
|
|
|
struct ip_vs_conn *ip_vs_conn_new(const struct ip_vs_conn_param *p,
|
|
|
|
const union nf_inet_addr *daddr,
|
2012-04-14 23:58:06 -06:00
|
|
|
__be16 dport, unsigned int flags,
|
2010-11-19 06:25:07 -07:00
|
|
|
struct ip_vs_dest *dest, __u32 fwmark);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern void ip_vs_conn_expire_now(struct ip_vs_conn *cp);
|
|
|
|
|
|
|
|
extern const char * ip_vs_state_name(__u16 proto, int state);
|
|
|
|
|
2011-01-03 06:44:47 -07:00
|
|
|
extern void ip_vs_tcp_conn_listen(struct net *net, struct ip_vs_conn *cp);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern int ip_vs_check_template(struct ip_vs_conn *ct);
|
2011-01-03 06:44:59 -07:00
|
|
|
extern void ip_vs_random_dropentry(struct net *net);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern int ip_vs_conn_init(void);
|
|
|
|
extern void ip_vs_conn_cleanup(void);
|
|
|
|
|
|
|
|
static inline void ip_vs_control_del(struct ip_vs_conn *cp)
|
|
|
|
{
|
|
|
|
struct ip_vs_conn *ctl_cp = cp->control;
|
|
|
|
if (!ctl_cp) {
|
2008-09-02 07:55:53 -06:00
|
|
|
IP_VS_ERR_BUF("request control DEL for uncontrolled: "
|
|
|
|
"%s:%d to %s:%d\n",
|
|
|
|
IP_VS_DBG_ADDR(cp->af, &cp->caddr),
|
|
|
|
ntohs(cp->cport),
|
|
|
|
IP_VS_DBG_ADDR(cp->af, &cp->vaddr),
|
|
|
|
ntohs(cp->vport));
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2008-09-02 07:55:53 -06:00
|
|
|
IP_VS_DBG_BUF(7, "DELeting control for: "
|
|
|
|
"cp.dst=%s:%d ctl_cp.dst=%s:%d\n",
|
|
|
|
IP_VS_DBG_ADDR(cp->af, &cp->caddr),
|
|
|
|
ntohs(cp->cport),
|
|
|
|
IP_VS_DBG_ADDR(cp->af, &ctl_cp->caddr),
|
|
|
|
ntohs(ctl_cp->cport));
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
cp->control = NULL;
|
|
|
|
if (atomic_read(&ctl_cp->n_control) == 0) {
|
2008-09-02 07:55:53 -06:00
|
|
|
IP_VS_ERR_BUF("BUG control DEL with n=0 : "
|
|
|
|
"%s:%d to %s:%d\n",
|
|
|
|
IP_VS_DBG_ADDR(cp->af, &cp->caddr),
|
|
|
|
ntohs(cp->cport),
|
|
|
|
IP_VS_DBG_ADDR(cp->af, &cp->vaddr),
|
|
|
|
ntohs(cp->vport));
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
atomic_dec(&ctl_cp->n_control);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void
|
|
|
|
ip_vs_control_add(struct ip_vs_conn *cp, struct ip_vs_conn *ctl_cp)
|
|
|
|
{
|
|
|
|
if (cp->control) {
|
2008-09-02 07:55:53 -06:00
|
|
|
IP_VS_ERR_BUF("request control ADD for already controlled: "
|
|
|
|
"%s:%d to %s:%d\n",
|
|
|
|
IP_VS_DBG_ADDR(cp->af, &cp->caddr),
|
|
|
|
ntohs(cp->cport),
|
|
|
|
IP_VS_DBG_ADDR(cp->af, &cp->vaddr),
|
|
|
|
ntohs(cp->vport));
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
ip_vs_control_del(cp);
|
|
|
|
}
|
|
|
|
|
2008-09-02 07:55:53 -06:00
|
|
|
IP_VS_DBG_BUF(7, "ADDing control for: "
|
|
|
|
"cp.dst=%s:%d ctl_cp.dst=%s:%d\n",
|
|
|
|
IP_VS_DBG_ADDR(cp->af, &cp->caddr),
|
|
|
|
ntohs(cp->cport),
|
|
|
|
IP_VS_DBG_ADDR(cp->af, &ctl_cp->caddr),
|
|
|
|
ntohs(ctl_cp->cport));
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
cp->control = ctl_cp;
|
|
|
|
atomic_inc(&ctl_cp->n_control);
|
|
|
|
}
|
|
|
|
|
2011-05-03 14:09:31 -06:00
|
|
|
/*
|
|
|
|
* IPVS netns init & cleanup functions
|
|
|
|
*/
|
2011-05-01 10:50:16 -06:00
|
|
|
extern int ip_vs_estimator_net_init(struct net *net);
|
|
|
|
extern int ip_vs_control_net_init(struct net *net);
|
|
|
|
extern int ip_vs_protocol_net_init(struct net *net);
|
|
|
|
extern int ip_vs_app_net_init(struct net *net);
|
|
|
|
extern int ip_vs_conn_net_init(struct net *net);
|
|
|
|
extern int ip_vs_sync_net_init(struct net *net);
|
|
|
|
extern void ip_vs_conn_net_cleanup(struct net *net);
|
|
|
|
extern void ip_vs_app_net_cleanup(struct net *net);
|
|
|
|
extern void ip_vs_protocol_net_cleanup(struct net *net);
|
|
|
|
extern void ip_vs_control_net_cleanup(struct net *net);
|
|
|
|
extern void ip_vs_estimator_net_cleanup(struct net *net);
|
|
|
|
extern void ip_vs_sync_net_cleanup(struct net *net);
|
|
|
|
extern void ip_vs_service_net_cleanup(struct net *net);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* IPVS application functions
|
|
|
|
* (from ip_vs_app.c)
|
|
|
|
*/
|
|
|
|
#define IP_VS_APP_MAX_PORTS 8
|
2012-07-12 14:06:20 -06:00
|
|
|
extern struct ip_vs_app *register_ip_vs_app(struct net *net,
|
|
|
|
struct ip_vs_app *app);
|
2011-01-03 06:44:53 -07:00
|
|
|
extern void unregister_ip_vs_app(struct net *net, struct ip_vs_app *app);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern int ip_vs_bind_app(struct ip_vs_conn *cp, struct ip_vs_protocol *pp);
|
|
|
|
extern void ip_vs_unbind_app(struct ip_vs_conn *cp);
|
2011-01-03 06:44:53 -07:00
|
|
|
extern int register_ip_vs_app_inc(struct net *net, struct ip_vs_app *app,
|
|
|
|
__u16 proto, __u16 port);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern int ip_vs_app_inc_get(struct ip_vs_app *inc);
|
|
|
|
extern void ip_vs_app_inc_put(struct ip_vs_app *inc);
|
|
|
|
|
2007-10-15 01:53:15 -06:00
|
|
|
extern int ip_vs_app_pkt_out(struct ip_vs_conn *, struct sk_buff *skb);
|
|
|
|
extern int ip_vs_app_pkt_in(struct ip_vs_conn *, struct sk_buff *skb);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2010-08-22 06:37:54 -06:00
|
|
|
int register_ip_vs_pe(struct ip_vs_pe *pe);
|
|
|
|
int unregister_ip_vs_pe(struct ip_vs_pe *pe);
|
2010-11-08 04:05:57 -07:00
|
|
|
struct ip_vs_pe *ip_vs_pe_getbyname(const char *name);
|
2010-11-19 06:25:12 -07:00
|
|
|
struct ip_vs_pe *__ip_vs_pe_getbyname(const char *pe_name);
|
2010-11-08 04:05:57 -07:00
|
|
|
|
2011-08-29 10:40:15 -06:00
|
|
|
/*
|
|
|
|
* Use a #define to avoid all of module.h just for these trivial ops
|
|
|
|
*/
|
|
|
|
#define ip_vs_pe_get(pe) \
|
|
|
|
if (pe && pe->module) \
|
2010-11-08 04:05:57 -07:00
|
|
|
__module_get(pe->module);
|
|
|
|
|
2011-08-29 10:40:15 -06:00
|
|
|
#define ip_vs_pe_put(pe) \
|
|
|
|
if (pe && pe->module) \
|
2010-11-08 04:05:57 -07:00
|
|
|
module_put(pe->module);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* IPVS protocol functions (from ip_vs_proto.c)
|
|
|
|
*/
|
|
|
|
extern int ip_vs_protocol_init(void);
|
|
|
|
extern void ip_vs_protocol_cleanup(void);
|
2011-01-03 06:44:51 -07:00
|
|
|
extern void ip_vs_protocol_timeout_change(struct netns_ipvs *ipvs, int flags);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern int *ip_vs_create_timeout_table(int *table, int size);
|
|
|
|
extern int
|
2009-08-05 11:42:58 -06:00
|
|
|
ip_vs_set_state_timeout(int *table, int num, const char *const *names,
|
|
|
|
const char *name, int to);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern void
|
2010-10-17 07:46:17 -06:00
|
|
|
ip_vs_tcpudp_debug_packet(int af, struct ip_vs_protocol *pp,
|
|
|
|
const struct sk_buff *skb,
|
2005-04-16 16:20:36 -06:00
|
|
|
int offset, const char *msg);
|
|
|
|
|
|
|
|
extern struct ip_vs_protocol ip_vs_protocol_tcp;
|
|
|
|
extern struct ip_vs_protocol ip_vs_protocol_udp;
|
|
|
|
extern struct ip_vs_protocol ip_vs_protocol_icmp;
|
|
|
|
extern struct ip_vs_protocol ip_vs_protocol_esp;
|
|
|
|
extern struct ip_vs_protocol ip_vs_protocol_ah;
|
2010-02-18 04:31:05 -07:00
|
|
|
extern struct ip_vs_protocol ip_vs_protocol_sctp;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Registering/unregistering scheduler functions
|
|
|
|
* (from ip_vs_sched.c)
|
|
|
|
*/
|
|
|
|
extern int register_ip_vs_scheduler(struct ip_vs_scheduler *scheduler);
|
|
|
|
extern int unregister_ip_vs_scheduler(struct ip_vs_scheduler *scheduler);
|
|
|
|
extern int ip_vs_bind_scheduler(struct ip_vs_service *svc,
|
|
|
|
struct ip_vs_scheduler *scheduler);
|
2013-03-22 03:46:53 -06:00
|
|
|
extern void ip_vs_unbind_scheduler(struct ip_vs_service *svc,
|
|
|
|
struct ip_vs_scheduler *sched);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern struct ip_vs_scheduler *ip_vs_scheduler_get(const char *sched_name);
|
|
|
|
extern void ip_vs_scheduler_put(struct ip_vs_scheduler *scheduler);
|
|
|
|
extern struct ip_vs_conn *
|
2010-10-17 07:24:37 -06:00
|
|
|
ip_vs_schedule(struct ip_vs_service *svc, struct sk_buff *skb,
|
ipvs: API change to avoid rescan of IPv6 exthdr
Reduce the number of times we scan/skip the IPv6 exthdrs.
This patch contains a lot of API changes. This is done, to avoid
repeating the scan of finding the IPv6 headers, via ipv6_find_hdr(),
which is called by ip_vs_fill_iph_skb().
Finding the IPv6 headers is done as early as possible, and passed on
as a pointer "struct ip_vs_iphdr *" to the affected functions.
This patch reduce/removes 19 calls to ip_vs_fill_iph_skb().
Notice, I have choosen, not to change the API of function
pointer "(*schedule)" (in struct ip_vs_scheduler) as it can be
used by external schedulers, via {un,}register_ip_vs_scheduler.
Only 4 out of 10 schedulers use info from ip_vs_iphdr*, and when
they do, they are only interested in iph->{s,d}addr.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:07:17 -06:00
|
|
|
struct ip_vs_proto_data *pd, int *ignored,
|
|
|
|
struct ip_vs_iphdr *iph);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern int ip_vs_leave(struct ip_vs_service *svc, struct sk_buff *skb,
|
ipvs: API change to avoid rescan of IPv6 exthdr
Reduce the number of times we scan/skip the IPv6 exthdrs.
This patch contains a lot of API changes. This is done, to avoid
repeating the scan of finding the IPv6 headers, via ipv6_find_hdr(),
which is called by ip_vs_fill_iph_skb().
Finding the IPv6 headers is done as early as possible, and passed on
as a pointer "struct ip_vs_iphdr *" to the affected functions.
This patch reduce/removes 19 calls to ip_vs_fill_iph_skb().
Notice, I have choosen, not to change the API of function
pointer "(*schedule)" (in struct ip_vs_scheduler) as it can be
used by external schedulers, via {un,}register_ip_vs_scheduler.
Only 4 out of 10 schedulers use info from ip_vs_iphdr*, and when
they do, they are only interested in iph->{s,d}addr.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:07:17 -06:00
|
|
|
struct ip_vs_proto_data *pd, struct ip_vs_iphdr *iph);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-02-11 06:01:12 -07:00
|
|
|
extern void ip_vs_scheduler_err(struct ip_vs_service *svc, const char *msg);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* IPVS control data and functions (from ip_vs_ctl.c)
|
|
|
|
*/
|
|
|
|
extern struct ip_vs_stats ip_vs_stats;
|
2010-11-19 06:25:14 -07:00
|
|
|
extern int sysctl_ip_vs_sync_ver;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
extern struct ip_vs_service *
|
2013-03-22 03:46:53 -06:00
|
|
|
ip_vs_service_find(struct net *net, int af, __u32 fwmark, __u16 protocol,
|
2008-09-02 07:55:38 -06:00
|
|
|
const union nf_inet_addr *vaddr, __be16 vport);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2013-03-21 03:58:08 -06:00
|
|
|
extern bool
|
|
|
|
ip_vs_has_real_service(struct net *net, int af, __u16 protocol,
|
|
|
|
const union nf_inet_addr *daddr, __be16 dport);
|
2008-09-02 07:55:48 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
extern int ip_vs_use_count_inc(void);
|
|
|
|
extern void ip_vs_use_count_dec(void);
|
2012-04-25 23:47:44 -06:00
|
|
|
extern int ip_vs_register_nl_ioctl(void);
|
|
|
|
extern void ip_vs_unregister_nl_ioctl(void);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern int ip_vs_control_init(void);
|
|
|
|
extern void ip_vs_control_cleanup(void);
|
2007-11-07 03:35:54 -07:00
|
|
|
extern struct ip_vs_dest *
|
2011-01-03 06:44:43 -07:00
|
|
|
ip_vs_find_dest(struct net *net, int af, const union nf_inet_addr *daddr,
|
|
|
|
__be16 dport, const union nf_inet_addr *vaddr, __be16 vport,
|
2011-12-29 22:19:02 -07:00
|
|
|
__u16 protocol, __u32 fwmark, __u32 flags);
|
2013-03-22 03:46:52 -06:00
|
|
|
extern void ip_vs_try_bind_dest(struct ip_vs_conn *cp);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2013-03-22 03:46:38 -06:00
|
|
|
static inline void ip_vs_dest_hold(struct ip_vs_dest *dest)
|
|
|
|
{
|
|
|
|
atomic_inc(&dest->refcnt);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ip_vs_dest_put(struct ip_vs_dest *dest)
|
|
|
|
{
|
|
|
|
smp_mb__before_atomic_dec();
|
|
|
|
atomic_dec(&dest->refcnt);
|
|
|
|
}
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* IPVS sync daemon data and function prototypes
|
|
|
|
* (from ip_vs_sync.c)
|
|
|
|
*/
|
2011-01-03 06:44:55 -07:00
|
|
|
extern int start_sync_thread(struct net *net, int state, char *mcast_ifn,
|
|
|
|
__u8 syncid);
|
|
|
|
extern int stop_sync_thread(struct net *net, int state);
|
2012-04-24 14:46:40 -06:00
|
|
|
extern void ip_vs_sync_conn(struct net *net, struct ip_vs_conn *cp, int pkts);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
* IPVS rate estimator prototypes (from ip_vs_est.c)
|
|
|
|
*/
|
2011-03-13 17:44:28 -06:00
|
|
|
extern void ip_vs_start_estimator(struct net *net, struct ip_vs_stats *stats);
|
|
|
|
extern void ip_vs_stop_estimator(struct net *net, struct ip_vs_stats *stats);
|
2005-04-16 16:20:36 -06:00
|
|
|
extern void ip_vs_zero_estimator(struct ip_vs_stats *stats);
|
2011-03-13 17:41:54 -06:00
|
|
|
extern void ip_vs_read_estimator(struct ip_vs_stats_user *dst,
|
|
|
|
struct ip_vs_stats *stats);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Various IPVS packet transmitters (from ip_vs_xmit.c)
|
|
|
|
*/
|
ipvs: API change to avoid rescan of IPv6 exthdr
Reduce the number of times we scan/skip the IPv6 exthdrs.
This patch contains a lot of API changes. This is done, to avoid
repeating the scan of finding the IPv6 headers, via ipv6_find_hdr(),
which is called by ip_vs_fill_iph_skb().
Finding the IPv6 headers is done as early as possible, and passed on
as a pointer "struct ip_vs_iphdr *" to the affected functions.
This patch reduce/removes 19 calls to ip_vs_fill_iph_skb().
Notice, I have choosen, not to change the API of function
pointer "(*schedule)" (in struct ip_vs_scheduler) as it can be
used by external schedulers, via {un,}register_ip_vs_scheduler.
Only 4 out of 10 schedulers use info from ip_vs_iphdr*, and when
they do, they are only interested in iph->{s,d}addr.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:07:17 -06:00
|
|
|
extern int ip_vs_null_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
struct ip_vs_protocol *pp, struct ip_vs_iphdr *iph);
|
|
|
|
extern int ip_vs_bypass_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
struct ip_vs_protocol *pp,
|
|
|
|
struct ip_vs_iphdr *iph);
|
|
|
|
extern int ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
struct ip_vs_protocol *pp, struct ip_vs_iphdr *iph);
|
|
|
|
extern int ip_vs_tunnel_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
struct ip_vs_protocol *pp,
|
|
|
|
struct ip_vs_iphdr *iph);
|
|
|
|
extern int ip_vs_dr_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
struct ip_vs_protocol *pp, struct ip_vs_iphdr *iph);
|
|
|
|
extern int ip_vs_icmp_xmit(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
struct ip_vs_protocol *pp, int offset,
|
|
|
|
unsigned int hooknum, struct ip_vs_iphdr *iph);
|
2013-03-21 03:58:06 -06:00
|
|
|
extern void ip_vs_dest_dst_rcu_free(struct rcu_head *head);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-09-02 07:55:45 -06:00
|
|
|
#ifdef CONFIG_IP_VS_IPV6
|
ipvs: API change to avoid rescan of IPv6 exthdr
Reduce the number of times we scan/skip the IPv6 exthdrs.
This patch contains a lot of API changes. This is done, to avoid
repeating the scan of finding the IPv6 headers, via ipv6_find_hdr(),
which is called by ip_vs_fill_iph_skb().
Finding the IPv6 headers is done as early as possible, and passed on
as a pointer "struct ip_vs_iphdr *" to the affected functions.
This patch reduce/removes 19 calls to ip_vs_fill_iph_skb().
Notice, I have choosen, not to change the API of function
pointer "(*schedule)" (in struct ip_vs_scheduler) as it can be
used by external schedulers, via {un,}register_ip_vs_scheduler.
Only 4 out of 10 schedulers use info from ip_vs_iphdr*, and when
they do, they are only interested in iph->{s,d}addr.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
2012-09-26 06:07:17 -06:00
|
|
|
extern int ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
struct ip_vs_protocol *pp,
|
|
|
|
struct ip_vs_iphdr *iph);
|
|
|
|
extern int ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
struct ip_vs_protocol *pp,
|
|
|
|
struct ip_vs_iphdr *iph);
|
|
|
|
extern int ip_vs_tunnel_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
struct ip_vs_protocol *pp,
|
|
|
|
struct ip_vs_iphdr *iph);
|
|
|
|
extern int ip_vs_dr_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
struct ip_vs_protocol *pp, struct ip_vs_iphdr *iph);
|
|
|
|
extern int ip_vs_icmp_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
struct ip_vs_protocol *pp, int offset,
|
|
|
|
unsigned int hooknum, struct ip_vs_iphdr *iph);
|
2008-09-02 07:55:45 -06:00
|
|
|
#endif
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2011-02-04 02:33:02 -07:00
|
|
|
#ifdef CONFIG_SYSCTL
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* This is a simple mechanism to ignore packets when
|
|
|
|
* we are loaded. Just set ip_vs_drop_rate to 'n' and
|
|
|
|
* we start to drop 1/rate of the packets
|
|
|
|
*/
|
|
|
|
|
2011-01-03 06:44:58 -07:00
|
|
|
static inline int ip_vs_todrop(struct netns_ipvs *ipvs)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2011-01-03 06:44:58 -07:00
|
|
|
if (!ipvs->drop_rate)
|
|
|
|
return 0;
|
|
|
|
if (--ipvs->drop_counter > 0)
|
|
|
|
return 0;
|
|
|
|
ipvs->drop_counter = ipvs->drop_rate;
|
2005-04-16 16:20:36 -06:00
|
|
|
return 1;
|
|
|
|
}
|
2011-02-04 02:33:02 -07:00
|
|
|
#else
|
|
|
|
static inline int ip_vs_todrop(struct netns_ipvs *ipvs) { return 0; }
|
|
|
|
#endif
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* ip_vs_fwd_tag returns the forwarding tag of the connection
|
|
|
|
*/
|
|
|
|
#define IP_VS_FWD_METHOD(cp) (cp->flags & IP_VS_CONN_F_FWD_MASK)
|
|
|
|
|
2005-09-01 18:40:26 -06:00
|
|
|
static inline char ip_vs_fwd_tag(struct ip_vs_conn *cp)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
|
|
|
char fwd;
|
|
|
|
|
|
|
|
switch (IP_VS_FWD_METHOD(cp)) {
|
|
|
|
case IP_VS_CONN_F_MASQ:
|
|
|
|
fwd = 'M'; break;
|
|
|
|
case IP_VS_CONN_F_LOCALNODE:
|
|
|
|
fwd = 'L'; break;
|
|
|
|
case IP_VS_CONN_F_TUNNEL:
|
|
|
|
fwd = 'T'; break;
|
|
|
|
case IP_VS_CONN_F_DROUTE:
|
|
|
|
fwd = 'R'; break;
|
|
|
|
case IP_VS_CONN_F_BYPASS:
|
|
|
|
fwd = 'B'; break;
|
|
|
|
default:
|
|
|
|
fwd = '?'; break;
|
|
|
|
}
|
|
|
|
return fwd;
|
|
|
|
}
|
|
|
|
|
|
|
|
extern void ip_vs_nat_icmp(struct sk_buff *skb, struct ip_vs_protocol *pp,
|
2008-09-02 07:55:45 -06:00
|
|
|
struct ip_vs_conn *cp, int dir);
|
|
|
|
|
|
|
|
#ifdef CONFIG_IP_VS_IPV6
|
|
|
|
extern void ip_vs_nat_icmp_v6(struct sk_buff *skb, struct ip_vs_protocol *pp,
|
|
|
|
struct ip_vs_conn *cp, int dir);
|
|
|
|
#endif
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-11-14 22:37:50 -07:00
|
|
|
extern __sum16 ip_vs_checksum_complete(struct sk_buff *skb, int offset);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-11-16 03:41:18 -07:00
|
|
|
static inline __wsum ip_vs_check_diff4(__be32 old, __be32 new, __wsum oldsum)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-11-16 03:41:18 -07:00
|
|
|
__be32 diff[2] = { ~old, new };
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2008-11-19 16:44:53 -07:00
|
|
|
return csum_partial(diff, sizeof(diff), oldsum);
|
2006-11-16 03:41:18 -07:00
|
|
|
}
|
|
|
|
|
2008-09-02 07:55:42 -06:00
|
|
|
#ifdef CONFIG_IP_VS_IPV6
|
|
|
|
static inline __wsum ip_vs_check_diff16(const __be32 *old, const __be32 *new,
|
|
|
|
__wsum oldsum)
|
|
|
|
{
|
|
|
|
__be32 diff[8] = { ~old[3], ~old[2], ~old[1], ~old[0],
|
|
|
|
new[3], new[2], new[1], new[0] };
|
|
|
|
|
2008-11-19 16:44:53 -07:00
|
|
|
return csum_partial(diff, sizeof(diff), oldsum);
|
2008-09-02 07:55:42 -06:00
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2006-11-16 03:41:18 -07:00
|
|
|
static inline __wsum ip_vs_check_diff2(__be16 old, __be16 new, __wsum oldsum)
|
|
|
|
{
|
|
|
|
__be16 diff[2] = { ~old, new };
|
|
|
|
|
2008-11-19 16:44:53 -07:00
|
|
|
return csum_partial(diff, sizeof(diff), oldsum);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2010-10-17 07:21:07 -06:00
|
|
|
/*
|
|
|
|
* Forget current conntrack (unconfirmed) and attach notrack entry
|
|
|
|
*/
|
|
|
|
static inline void ip_vs_notrack(struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
|
|
|
|
enum ip_conntrack_info ctinfo;
|
2011-03-09 14:55:05 -07:00
|
|
|
struct nf_conn *ct = nf_ct_get(skb, &ctinfo);
|
2010-10-17 07:21:07 -06:00
|
|
|
|
|
|
|
if (!ct || !nf_ct_is_untracked(ct)) {
|
2012-07-07 04:26:10 -06:00
|
|
|
nf_conntrack_put(skb->nfct);
|
2010-10-17 07:21:07 -06:00
|
|
|
skb->nfct = &nf_ct_untracked_get()->ct_general;
|
|
|
|
skb->nfctinfo = IP_CT_NEW;
|
|
|
|
nf_conntrack_get(skb->nfct);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
2010-09-21 09:35:41 -06:00
|
|
|
#ifdef CONFIG_IP_VS_NFCT
|
|
|
|
/*
|
|
|
|
* Netfilter connection tracking
|
|
|
|
* (from ip_vs_nfct.c)
|
|
|
|
*/
|
2011-01-03 06:44:58 -07:00
|
|
|
static inline int ip_vs_conntrack_enabled(struct netns_ipvs *ipvs)
|
2010-09-21 09:35:41 -06:00
|
|
|
{
|
2011-02-04 02:33:02 -07:00
|
|
|
#ifdef CONFIG_SYSCTL
|
2011-01-03 06:44:58 -07:00
|
|
|
return ipvs->sysctl_conntrack;
|
2011-02-04 02:33:02 -07:00
|
|
|
#else
|
|
|
|
return 0;
|
|
|
|
#endif
|
2010-09-21 09:35:41 -06:00
|
|
|
}
|
|
|
|
|
2010-09-05 12:02:29 -06:00
|
|
|
extern void ip_vs_update_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp,
|
|
|
|
int outin);
|
2011-09-15 23:02:19 -06:00
|
|
|
extern int ip_vs_confirm_conntrack(struct sk_buff *skb);
|
2010-09-21 09:35:41 -06:00
|
|
|
extern void ip_vs_nfct_expect_related(struct sk_buff *skb, struct nf_conn *ct,
|
|
|
|
struct ip_vs_conn *cp, u_int8_t proto,
|
|
|
|
const __be16 port, int from_rs);
|
|
|
|
extern void ip_vs_conn_drop_conntrack(struct ip_vs_conn *cp);
|
|
|
|
|
|
|
|
#else
|
|
|
|
|
2011-01-03 06:44:58 -07:00
|
|
|
static inline int ip_vs_conntrack_enabled(struct netns_ipvs *ipvs)
|
2010-09-21 09:35:41 -06:00
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ip_vs_update_conntrack(struct sk_buff *skb,
|
|
|
|
struct ip_vs_conn *cp, int outin)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2011-10-20 06:18:04 -06:00
|
|
|
static inline int ip_vs_confirm_conntrack(struct sk_buff *skb)
|
2010-09-21 09:35:41 -06:00
|
|
|
{
|
|
|
|
return NF_ACCEPT;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ip_vs_conn_drop_conntrack(struct ip_vs_conn *cp)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
/* CONFIG_IP_VS_NFCT */
|
|
|
|
#endif
|
2010-09-05 12:02:29 -06:00
|
|
|
|
2011-02-19 02:32:28 -07:00
|
|
|
static inline unsigned int
|
|
|
|
ip_vs_dest_conn_overhead(struct ip_vs_dest *dest)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* We think the overhead of processing active connections is 256
|
|
|
|
* times higher than that of inactive connections in average. (This
|
|
|
|
* 256 times might not be accurate, we will change it later) We
|
|
|
|
* use the following formula to estimate the overhead now:
|
|
|
|
* dest->activeconns*256 + dest->inactconns
|
|
|
|
*/
|
|
|
|
return (atomic_read(&dest->activeconns) << 8) +
|
|
|
|
atomic_read(&dest->inactconns);
|
|
|
|
}
|
|
|
|
|
2008-07-31 21:45:24 -06:00
|
|
|
#endif /* _NET_IP_VS_H */
|