From 61b1ab4583e275af216c8454b9256de680499b19 Mon Sep 17 00:00:00 2001 From: Hans Schillstrom Date: Mon, 3 Jan 2011 14:44:42 +0100 Subject: [PATCH] IPVS: netns, add basic init per netns. Preparation for network name-space init, in this stage some empty functions exists. In most files there is a check if it is root ns i.e. init_net if (!net_eq(net, &init_net)) return ... this will be removed by the last patch, when enabling name-space. *v3 ip_vs_conn.c merge error corrected. net_ipvs #ifdef removed as sugested by Jan Engelhardt [ horms@verge.net.au: Removed whitespace-change-only hunks ] Signed-off-by: Hans Schillstrom Acked-by: Julian Anastasov Signed-off-by: Simon Horman --- include/net/ip_vs.h | 11 ++++++ include/net/net_namespace.h | 2 + include/net/netns/ip_vs.h | 25 +++++++++++++ net/netfilter/ipvs/ip_vs_app.c | 28 ++++++++++++-- net/netfilter/ipvs/ip_vs_conn.c | 34 ++++++++++++++--- net/netfilter/ipvs/ip_vs_core.c | 63 +++++++++++++++++++++++++++++++- net/netfilter/ipvs/ip_vs_ctl.c | 49 ++++++++++++++++++++----- net/netfilter/ipvs/ip_vs_est.c | 20 +++++++++- net/netfilter/ipvs/ip_vs_ftp.c | 34 +++++++++++++++-- net/netfilter/ipvs/ip_vs_lblc.c | 37 +++++++++++++++++-- net/netfilter/ipvs/ip_vs_lblcr.c | 38 +++++++++++++++++-- net/netfilter/ipvs/ip_vs_proto.c | 19 ++++++++++ net/netfilter/ipvs/ip_vs_sync.c | 27 ++++++++++++++ 13 files changed, 354 insertions(+), 33 deletions(-) create mode 100644 include/net/netns/ip_vs.h diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index d858264217ba..c1c2ece3ed94 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -28,6 +28,15 @@ #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) #include #endif +#include /* Netw namespace */ + +/* + * Generic access of ipvs struct + */ +static inline struct netns_ipvs *net_ipvs(struct net* net) +{ + return net->ipvs; +} /* Connections' size value needed by ip_vs_ctl.c */ extern int ip_vs_conn_tab_size; @@ -922,6 +931,8 @@ extern char ip_vs_backup_mcast_ifn[IP_VS_IFNAME_MAXLEN]; extern int start_sync_thread(int state, char *mcast_ifn, __u8 syncid); extern int stop_sync_thread(int state); extern void ip_vs_sync_conn(struct ip_vs_conn *cp); +extern int ip_vs_sync_init(void); +extern void ip_vs_sync_cleanup(void); /* diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index 1bf812b21fb7..b3b4a34cb2cc 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -20,6 +20,7 @@ #include #endif #include +#include struct proc_dir_entry; struct net_device; @@ -94,6 +95,7 @@ struct net { #ifdef CONFIG_XFRM struct netns_xfrm xfrm; #endif + struct netns_ipvs *ipvs; }; diff --git a/include/net/netns/ip_vs.h b/include/net/netns/ip_vs.h new file mode 100644 index 000000000000..12fe84087cec --- /dev/null +++ b/include/net/netns/ip_vs.h @@ -0,0 +1,25 @@ +/* + * IP Virtual Server + * Data structure for network namspace + * + */ + +#ifndef IP_VS_H_ +#define IP_VS_H_ + +#include +#include +#include +#include +#include +#include + +struct ip_vs_stats; +struct ip_vs_sync_buff; +struct ctl_table_header; + +struct netns_ipvs { + int gen; /* Generation */ +}; + +#endif /* IP_VS_H_ */ diff --git a/net/netfilter/ipvs/ip_vs_app.c b/net/netfilter/ipvs/ip_vs_app.c index a475edee0912..40b09ccc4896 100644 --- a/net/netfilter/ipvs/ip_vs_app.c +++ b/net/netfilter/ipvs/ip_vs_app.c @@ -569,15 +569,35 @@ static const struct file_operations ip_vs_app_fops = { }; #endif +static int __net_init __ip_vs_app_init(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + + proc_net_fops_create(net, "ip_vs_app", 0, &ip_vs_app_fops); + return 0; +} + +static void __net_exit __ip_vs_app_cleanup(struct net *net) +{ + proc_net_remove(net, "ip_vs_app"); +} + +static struct pernet_operations ip_vs_app_ops = { + .init = __ip_vs_app_init, + .exit = __ip_vs_app_cleanup, +}; + int __init ip_vs_app_init(void) { - /* we will replace it with proc_net_ipvs_create() soon */ - proc_net_fops_create(&init_net, "ip_vs_app", 0, &ip_vs_app_fops); - return 0; + int rv; + + rv = register_pernet_subsys(&ip_vs_app_ops); + return rv; } void ip_vs_app_cleanup(void) { - proc_net_remove(&init_net, "ip_vs_app"); + unregister_pernet_subsys(&ip_vs_app_ops); } diff --git a/net/netfilter/ipvs/ip_vs_conn.c b/net/netfilter/ipvs/ip_vs_conn.c index 66e4662925d5..7c1b502f8d8d 100644 --- a/net/netfilter/ipvs/ip_vs_conn.c +++ b/net/netfilter/ipvs/ip_vs_conn.c @@ -1201,11 +1201,36 @@ static void ip_vs_conn_flush(void) goto flush_again; } } +/* + * per netns init and exit + */ +int __net_init __ip_vs_conn_init(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + proc_net_fops_create(net, "ip_vs_conn", 0, &ip_vs_conn_fops); + proc_net_fops_create(net, "ip_vs_conn_sync", 0, &ip_vs_conn_sync_fops); + return 0; +} + +static void __net_exit __ip_vs_conn_cleanup(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return; + + proc_net_remove(net, "ip_vs_conn"); + proc_net_remove(net, "ip_vs_conn_sync"); +} +static struct pernet_operations ipvs_conn_ops = { + .init = __ip_vs_conn_init, + .exit = __ip_vs_conn_cleanup, +}; int __init ip_vs_conn_init(void) { int idx; + int retc; /* Compute size and mask */ ip_vs_conn_tab_size = 1 << ip_vs_conn_tab_bits; @@ -1243,24 +1268,21 @@ int __init ip_vs_conn_init(void) rwlock_init(&__ip_vs_conntbl_lock_array[idx].l); } - proc_net_fops_create(&init_net, "ip_vs_conn", 0, &ip_vs_conn_fops); - proc_net_fops_create(&init_net, "ip_vs_conn_sync", 0, &ip_vs_conn_sync_fops); + retc = register_pernet_subsys(&ipvs_conn_ops); /* calculate the random value for connection hash */ get_random_bytes(&ip_vs_conn_rnd, sizeof(ip_vs_conn_rnd)); - return 0; + return retc; } - void ip_vs_conn_cleanup(void) { + unregister_pernet_subsys(&ipvs_conn_ops); /* flush all the connection entries first */ ip_vs_conn_flush(); /* Release the empty cache */ kmem_cache_destroy(ip_vs_conn_cachep); - proc_net_remove(&init_net, "ip_vs_conn"); - proc_net_remove(&init_net, "ip_vs_conn_sync"); vfree(ip_vs_conn_tab); } diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 5287771d0647..206f40c548d7 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -41,6 +41,7 @@ #include /* for icmp_send */ #include #include +#include /* net_generic() */ #include #include @@ -68,6 +69,12 @@ EXPORT_SYMBOL(ip_vs_conn_put); EXPORT_SYMBOL(ip_vs_get_debug_level); #endif +int ip_vs_net_id __read_mostly; +#ifdef IP_VS_GENERIC_NETNS +EXPORT_SYMBOL(ip_vs_net_id); +#endif +/* netns cnt used for uniqueness */ +static atomic_t ipvs_netns_cnt = ATOMIC_INIT(0); /* ID used in ICMP lookups */ #define icmp_id(icmph) (((icmph)->un).echo.id) @@ -1813,6 +1820,44 @@ static struct nf_hook_ops ip_vs_ops[] __read_mostly = { #endif }; +/* + * Initialize IP Virtual Server netns mem. + */ +static int __net_init __ip_vs_init(struct net *net) +{ + struct netns_ipvs *ipvs; + + if (!net_eq(net, &init_net)) { + pr_err("The final patch for enabling netns is missing\n"); + return -EPERM; + } + ipvs = net_generic(net, ip_vs_net_id); + if (ipvs == NULL) { + pr_err("%s(): no memory.\n", __func__); + return -ENOMEM; + } + /* Counters used for creating unique names */ + ipvs->gen = atomic_read(&ipvs_netns_cnt); + atomic_inc(&ipvs_netns_cnt); + net->ipvs = ipvs; + printk(KERN_INFO "IPVS: Creating netns size=%lu id=%d\n", + sizeof(struct netns_ipvs), ipvs->gen); + return 0; +} + +static void __net_exit __ip_vs_cleanup(struct net *net) +{ + struct netns_ipvs *ipvs = net_ipvs(net); + + IP_VS_DBG(10, "ipvs netns %d released\n", ipvs->gen); +} + +static struct pernet_operations ipvs_core_ops = { + .init = __ip_vs_init, + .exit = __ip_vs_cleanup, + .id = &ip_vs_net_id, + .size = sizeof(struct netns_ipvs), +}; /* * Initialize IP Virtual Server @@ -1821,8 +1866,11 @@ static int __init ip_vs_init(void) { int ret; - ip_vs_estimator_init(); + ret = register_pernet_subsys(&ipvs_core_ops); /* Alloc ip_vs struct */ + if (ret < 0) + return ret; + ip_vs_estimator_init(); ret = ip_vs_control_init(); if (ret < 0) { pr_err("can't setup control.\n"); @@ -1843,15 +1891,23 @@ static int __init ip_vs_init(void) goto cleanup_app; } + ret = ip_vs_sync_init(); + if (ret < 0) { + pr_err("can't setup sync data.\n"); + goto cleanup_conn; + } + ret = nf_register_hooks(ip_vs_ops, ARRAY_SIZE(ip_vs_ops)); if (ret < 0) { pr_err("can't register hooks.\n"); - goto cleanup_conn; + goto cleanup_sync; } pr_info("ipvs loaded.\n"); return ret; +cleanup_sync: + ip_vs_sync_cleanup(); cleanup_conn: ip_vs_conn_cleanup(); cleanup_app: @@ -1861,17 +1917,20 @@ static int __init ip_vs_init(void) ip_vs_control_cleanup(); cleanup_estimator: ip_vs_estimator_cleanup(); + unregister_pernet_subsys(&ipvs_core_ops); /* free ip_vs struct */ return ret; } static void __exit ip_vs_cleanup(void) { nf_unregister_hooks(ip_vs_ops, ARRAY_SIZE(ip_vs_ops)); + ip_vs_sync_cleanup(); ip_vs_conn_cleanup(); ip_vs_app_cleanup(); ip_vs_protocol_cleanup(); ip_vs_control_cleanup(); ip_vs_estimator_cleanup(); + unregister_pernet_subsys(&ipvs_core_ops); /* free ip_vs struct */ pr_info("ipvs unloaded.\n"); } diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index ca49e928f302..ceeef4352d34 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -3406,6 +3406,42 @@ static void ip_vs_genl_unregister(void) /* End of Generic Netlink interface definitions */ +/* + * per netns intit/exit func. + */ +int __net_init __ip_vs_control_init(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + + proc_net_fops_create(net, "ip_vs", 0, &ip_vs_info_fops); + proc_net_fops_create(net, "ip_vs_stats", 0, &ip_vs_stats_fops); + sysctl_header = register_net_sysctl_table(net, net_vs_ctl_path, + vs_vars); + if (sysctl_header == NULL) + goto err_reg; + ip_vs_new_estimator(&ip_vs_stats); + return 0; + +err_reg: + return -ENOMEM; +} + +static void __net_exit __ip_vs_control_cleanup(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return; + + ip_vs_kill_estimator(&ip_vs_stats); + unregister_net_sysctl_table(sysctl_header); + proc_net_remove(net, "ip_vs_stats"); + proc_net_remove(net, "ip_vs"); +} + +static struct pernet_operations ipvs_control_ops = { + .init = __ip_vs_control_init, + .exit = __ip_vs_control_cleanup, +}; int __init ip_vs_control_init(void) { @@ -3437,12 +3473,9 @@ int __init ip_vs_control_init(void) return ret; } - proc_net_fops_create(&init_net, "ip_vs", 0, &ip_vs_info_fops); - proc_net_fops_create(&init_net, "ip_vs_stats",0, &ip_vs_stats_fops); - - sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars); - - ip_vs_new_estimator(&ip_vs_stats); + ret = register_pernet_subsys(&ipvs_control_ops); + if (ret) + return ret; /* Hook the defense timer */ schedule_delayed_work(&defense_work, DEFENSE_TIMER_PERIOD); @@ -3459,9 +3492,7 @@ void ip_vs_control_cleanup(void) cancel_delayed_work_sync(&defense_work); cancel_work_sync(&defense_work.work); ip_vs_kill_estimator(&ip_vs_stats); - unregister_sysctl_table(sysctl_header); - proc_net_remove(&init_net, "ip_vs_stats"); - proc_net_remove(&init_net, "ip_vs"); + unregister_pernet_subsys(&ipvs_control_ops); ip_vs_genl_unregister(); nf_unregister_sockopt(&ip_vs_sockopts); LeaveFunction(2); diff --git a/net/netfilter/ipvs/ip_vs_est.c b/net/netfilter/ipvs/ip_vs_est.c index ff28801962e0..7417a0c1408b 100644 --- a/net/netfilter/ipvs/ip_vs_est.c +++ b/net/netfilter/ipvs/ip_vs_est.c @@ -157,13 +157,31 @@ void ip_vs_zero_estimator(struct ip_vs_stats *stats) est->outbps = 0; } +static int __net_init __ip_vs_estimator_init(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + + return 0; +} + +static struct pernet_operations ip_vs_app_ops = { + .init = __ip_vs_estimator_init, +}; + int __init ip_vs_estimator_init(void) { + int rv; + + rv = register_pernet_subsys(&ip_vs_app_ops); + if (rv < 0) + return rv; mod_timer(&est_timer, jiffies + 2 * HZ); - return 0; + return rv; } void ip_vs_estimator_cleanup(void) { del_timer_sync(&est_timer); + unregister_pernet_subsys(&ip_vs_app_ops); } diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index 84aef65b37d1..0e762f322aa3 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -399,15 +399,17 @@ static struct ip_vs_app ip_vs_ftp = { .pkt_in = ip_vs_ftp_in, }; - /* - * ip_vs_ftp initialization + * per netns ip_vs_ftp initialization */ -static int __init ip_vs_ftp_init(void) +static int __net_init __ip_vs_ftp_init(struct net *net) { int i, ret; struct ip_vs_app *app = &ip_vs_ftp; + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + ret = register_ip_vs_app(app); if (ret) return ret; @@ -427,14 +429,38 @@ static int __init ip_vs_ftp_init(void) return ret; } +/* + * netns exit + */ +static void __ip_vs_ftp_exit(struct net *net) +{ + struct ip_vs_app *app = &ip_vs_ftp; + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return; + + unregister_ip_vs_app(app); +} + +static struct pernet_operations ip_vs_ftp_ops = { + .init = __ip_vs_ftp_init, + .exit = __ip_vs_ftp_exit, +}; + +int __init ip_vs_ftp_init(void) +{ + int rv; + + rv = register_pernet_subsys(&ip_vs_ftp_ops); + return rv; +} /* * ip_vs_ftp finish. */ static void __exit ip_vs_ftp_exit(void) { - unregister_ip_vs_app(&ip_vs_ftp); + unregister_pernet_subsys(&ip_vs_ftp_ops); } diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c index 9323f8944199..84278fb4e055 100644 --- a/net/netfilter/ipvs/ip_vs_lblc.c +++ b/net/netfilter/ipvs/ip_vs_lblc.c @@ -543,23 +543,54 @@ static struct ip_vs_scheduler ip_vs_lblc_scheduler = .schedule = ip_vs_lblc_schedule, }; +/* + * per netns init. + */ +static int __net_init __ip_vs_lblc_init(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + + sysctl_header = register_net_sysctl_table(net, net_vs_ctl_path, + vs_vars_table); + if (!sysctl_header) + return -ENOMEM; + + return 0; +} + +static void __net_exit __ip_vs_lblc_exit(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return; + + unregister_net_sysctl_table(sysctl_header); +} + +static struct pernet_operations ip_vs_lblc_ops = { + .init = __ip_vs_lblc_init, + .exit = __ip_vs_lblc_exit, +}; static int __init ip_vs_lblc_init(void) { int ret; - sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars_table); + ret = register_pernet_subsys(&ip_vs_lblc_ops); + if (ret) + return ret; + ret = register_ip_vs_scheduler(&ip_vs_lblc_scheduler); if (ret) - unregister_sysctl_table(sysctl_header); + unregister_pernet_subsys(&ip_vs_lblc_ops); return ret; } static void __exit ip_vs_lblc_cleanup(void) { - unregister_sysctl_table(sysctl_header); unregister_ip_vs_scheduler(&ip_vs_lblc_scheduler); + unregister_pernet_subsys(&ip_vs_lblc_ops); } diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c index dbeed8ea421a..7c7396a6acbf 100644 --- a/net/netfilter/ipvs/ip_vs_lblcr.c +++ b/net/netfilter/ipvs/ip_vs_lblcr.c @@ -744,23 +744,53 @@ static struct ip_vs_scheduler ip_vs_lblcr_scheduler = .schedule = ip_vs_lblcr_schedule, }; +/* + * per netns init. + */ +static int __net_init __ip_vs_lblcr_init(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return -EPERM; + + sysctl_header = register_net_sysctl_table(net, net_vs_ctl_path, + vs_vars_table); + if (!sysctl_header) + return -ENOMEM; + + return 0; +} + +static void __net_exit __ip_vs_lblcr_exit(struct net *net) +{ + if (!net_eq(net, &init_net)) /* netns not enabled yet */ + return; + + unregister_net_sysctl_table(sysctl_header); +} + +static struct pernet_operations ip_vs_lblcr_ops = { + .init = __ip_vs_lblcr_init, + .exit = __ip_vs_lblcr_exit, +}; static int __init ip_vs_lblcr_init(void) { int ret; - sysctl_header = register_sysctl_paths(net_vs_ctl_path, vs_vars_table); + ret = register_pernet_subsys(&ip_vs_lblcr_ops); + if (ret) + return ret; + ret = register_ip_vs_scheduler(&ip_vs_lblcr_scheduler); if (ret) - unregister_sysctl_table(sysctl_header); + unregister_pernet_subsys(&ip_vs_lblcr_ops); return ret; } - static void __exit ip_vs_lblcr_cleanup(void) { - unregister_sysctl_table(sysctl_header); unregister_ip_vs_scheduler(&ip_vs_lblcr_scheduler); + unregister_pernet_subsys(&ip_vs_lblcr_ops); } diff --git a/net/netfilter/ipvs/ip_vs_proto.c b/net/netfilter/ipvs/ip_vs_proto.c index c53998390877..45392942d0e7 100644 --- a/net/netfilter/ipvs/ip_vs_proto.c +++ b/net/netfilter/ipvs/ip_vs_proto.c @@ -236,6 +236,23 @@ ip_vs_tcpudp_debug_packet(int af, struct ip_vs_protocol *pp, ip_vs_tcpudp_debug_packet_v4(pp, skb, offset, msg); } +/* + * per network name-space init + */ +static int __net_init __ip_vs_protocol_init(struct net *net) +{ + return 0; +} + +static void __net_exit __ip_vs_protocol_cleanup(struct net *net) +{ + /* empty */ +} + +static struct pernet_operations ipvs_proto_ops = { + .init = __ip_vs_protocol_init, + .exit = __ip_vs_protocol_cleanup, +}; int __init ip_vs_protocol_init(void) { @@ -265,6 +282,7 @@ int __init ip_vs_protocol_init(void) REGISTER_PROTOCOL(&ip_vs_protocol_esp); #endif pr_info("Registered protocols (%s)\n", &protocols[2]); + return register_pernet_subsys(&ipvs_proto_ops); return 0; } @@ -275,6 +293,7 @@ void ip_vs_protocol_cleanup(void) struct ip_vs_protocol *pp; int i; + unregister_pernet_subsys(&ipvs_proto_ops); /* unregister all the ipvs protocols */ for (i = 0; i < IP_VS_PROTO_TAB_SIZE; i++) { while ((pp = ip_vs_proto_table[i]) != NULL) diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c index c1c167ab73ee..3668739a6d06 100644 --- a/net/netfilter/ipvs/ip_vs_sync.c +++ b/net/netfilter/ipvs/ip_vs_sync.c @@ -1639,3 +1639,30 @@ int stop_sync_thread(int state) return 0; } + +/* + * Initialize data struct for each netns + */ +static int __net_init __ip_vs_sync_init(struct net *net) +{ + return 0; +} + +static void __ip_vs_sync_cleanup(struct net *net) +{ +} +static struct pernet_operations ipvs_sync_ops = { + .init = __ip_vs_sync_init, + .exit = __ip_vs_sync_cleanup, +}; + + +int __init ip_vs_sync_init(void) +{ + return register_pernet_subsys(&ipvs_sync_ops); +} + +void __exit ip_vs_sync_cleanup(void) +{ + unregister_pernet_subsys(&ipvs_sync_ops); +}