From 70a269e6c9c9b38b1a37dce068c59e9a912f8578 Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Thu, 22 Jan 2009 04:56:15 +0000 Subject: netns: ipmr: allocate mroute_socket per-namespace. Preliminary work to make IPv4 multicast routing netns-aware. Make IPv4 multicast routing mroute_socket per-namespace, moves it into struct netns_ipv4. At the moment, mroute_socket is only referenced in init_net. Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- net/ipv4/ipmr.c | 28 +++++++++++++--------------- 1 file changed, 13 insertions(+), 15 deletions(-) (limited to 'net/ipv4/ipmr.c') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 14666449dc1c..ac324b702e8b 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -67,9 +67,6 @@ #define CONFIG_IP_PIMSM 1 #endif -static struct sock *mroute_socket; - - /* Big lock, protecting vif table, mrt cache and mroute socket state. Note that the changes are semaphored via rtnl_lock. */ @@ -658,7 +655,7 @@ static int ipmr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) skb->transport_header = skb->network_header; } - if (mroute_socket == NULL) { + if (init_net.ipv4.mroute_sk == NULL) { kfree_skb(skb); return -EINVAL; } @@ -666,7 +663,8 @@ static int ipmr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) /* * Deliver to mrouted */ - if ((ret = sock_queue_rcv_skb(mroute_socket, skb))<0) { + ret = sock_queue_rcv_skb(init_net.ipv4.mroute_sk, skb); + if (ret < 0) { if (net_ratelimit()) printk(KERN_WARNING "mroute: pending queue full, dropping entries.\n"); kfree_skb(skb); @@ -896,11 +894,11 @@ static void mroute_clean_tables(struct sock *sk) static void mrtsock_destruct(struct sock *sk) { rtnl_lock(); - if (sk == mroute_socket) { + if (sk == init_net.ipv4.mroute_sk) { IPV4_DEVCONF_ALL(sock_net(sk), MC_FORWARDING)--; write_lock_bh(&mrt_lock); - mroute_socket = NULL; + init_net.ipv4.mroute_sk = NULL; write_unlock_bh(&mrt_lock); mroute_clean_tables(sk); @@ -922,7 +920,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int struct mfcctl mfc; if (optname != MRT_INIT) { - if (sk != mroute_socket && !capable(CAP_NET_ADMIN)) + if (sk != init_net.ipv4.mroute_sk && !capable(CAP_NET_ADMIN)) return -EACCES; } @@ -935,7 +933,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int return -ENOPROTOOPT; rtnl_lock(); - if (mroute_socket) { + if (init_net.ipv4.mroute_sk) { rtnl_unlock(); return -EADDRINUSE; } @@ -943,7 +941,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int ret = ip_ra_control(sk, 1, mrtsock_destruct); if (ret == 0) { write_lock_bh(&mrt_lock); - mroute_socket = sk; + init_net.ipv4.mroute_sk = sk; write_unlock_bh(&mrt_lock); IPV4_DEVCONF_ALL(sock_net(sk), MC_FORWARDING)++; @@ -951,7 +949,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int rtnl_unlock(); return ret; case MRT_DONE: - if (sk != mroute_socket) + if (sk != init_net.ipv4.mroute_sk) return -EACCES; return ip_ra_control(sk, 0, NULL); case MRT_ADD_VIF: @@ -964,7 +962,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int return -ENFILE; rtnl_lock(); if (optname == MRT_ADD_VIF) { - ret = vif_add(&vif, sk==mroute_socket); + ret = vif_add(&vif, sk == init_net.ipv4.mroute_sk); } else { ret = vif_delete(vif.vifc_vifi, 0); } @@ -985,7 +983,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int if (optname == MRT_DEL_MFC) ret = ipmr_mfc_delete(&mfc); else - ret = ipmr_mfc_add(&mfc, sk==mroute_socket); + ret = ipmr_mfc_add(&mfc, sk == init_net.ipv4.mroute_sk); rtnl_unlock(); return ret; /* @@ -1425,9 +1423,9 @@ int ip_mr_input(struct sk_buff *skb) that we can forward NO IGMP messages. */ read_lock(&mrt_lock); - if (mroute_socket) { + if (init_net.ipv4.mroute_sk) { nf_reset(skb); - raw_rcv(mroute_socket, skb); + raw_rcv(init_net.ipv4.mroute_sk, skb); read_unlock(&mrt_lock); return 0; } -- cgit v1.2.2 From cf958ae377ee2545ae70cf48d38e7eb4308c27ea Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Thu, 22 Jan 2009 04:56:16 +0000 Subject: netns: ipmr: dynamically allocate vif_table Preliminary work to make IPv6 multicast routing netns-aware. Dynamically allocate interface table vif_table and move it to struct netns_ipv4, and update MIF_EXISTS() macro. At the moment, vif_table is only referenced in init_net. Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- net/ipv4/ipmr.c | 109 +++++++++++++++++++++++++++++++++++--------------------- 1 file changed, 68 insertions(+), 41 deletions(-) (limited to 'net/ipv4/ipmr.c') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index ac324b702e8b..75a5f79cc226 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -77,10 +77,7 @@ static DEFINE_RWLOCK(mrt_lock); * Multicast router control variables */ -static struct vif_device vif_table[MAXVIFS]; /* Devices */ -static int maxvif; - -#define VIF_EXISTS(idx) (vif_table[idx].dev != NULL) +#define VIF_EXISTS(_net, _idx) ((_net)->ipv4.vif_table[_idx].dev != NULL) static int mroute_do_assert; /* Set in PIM assert */ static int mroute_do_pim; @@ -286,10 +283,10 @@ static int vif_delete(int vifi, int notify) struct net_device *dev; struct in_device *in_dev; - if (vifi < 0 || vifi >= maxvif) + if (vifi < 0 || vifi >= init_net.ipv4.maxvif) return -EADDRNOTAVAIL; - v = &vif_table[vifi]; + v = &init_net.ipv4.vif_table[vifi]; write_lock_bh(&mrt_lock); dev = v->dev; @@ -305,13 +302,13 @@ static int vif_delete(int vifi, int notify) reg_vif_num = -1; #endif - if (vifi+1 == maxvif) { + if (vifi+1 == init_net.ipv4.maxvif) { int tmp; for (tmp=vifi-1; tmp>=0; tmp--) { - if (VIF_EXISTS(tmp)) + if (VIF_EXISTS(&init_net, tmp)) break; } - maxvif = tmp+1; + init_net.ipv4.maxvif = tmp+1; } write_unlock_bh(&mrt_lock); @@ -411,8 +408,9 @@ static void ipmr_update_thresholds(struct mfc_cache *cache, unsigned char *ttls) cache->mfc_un.res.maxvif = 0; memset(cache->mfc_un.res.ttls, 255, MAXVIFS); - for (vifi=0; vifimfc_un.res.ttls[vifi] = ttls[vifi]; if (cache->mfc_un.res.minvif > vifi) cache->mfc_un.res.minvif = vifi; @@ -425,13 +423,13 @@ static void ipmr_update_thresholds(struct mfc_cache *cache, unsigned char *ttls) static int vif_add(struct vifctl *vifc, int mrtsock) { int vifi = vifc->vifc_vifi; - struct vif_device *v = &vif_table[vifi]; + struct vif_device *v = &init_net.ipv4.vif_table[vifi]; struct net_device *dev; struct in_device *in_dev; int err; /* Is vif busy ? */ - if (VIF_EXISTS(vifi)) + if (VIF_EXISTS(&init_net, vifi)) return -EADDRINUSE; switch (vifc->vifc_flags) { @@ -509,8 +507,8 @@ static int vif_add(struct vifctl *vifc, int mrtsock) if (v->flags&VIFF_REGISTER) reg_vif_num = vifi; #endif - if (vifi+1 > maxvif) - maxvif = vifi+1; + if (vifi+1 > init_net.ipv4.maxvif) + init_net.ipv4.maxvif = vifi+1; write_unlock_bh(&mrt_lock); return 0; } @@ -849,8 +847,8 @@ static void mroute_clean_tables(struct sock *sk) /* * Shut down all active vif entries */ - for (i=0; i= maxvif) + if (vr.vifi >= init_net.ipv4.maxvif) return -EINVAL; read_lock(&mrt_lock); - vif=&vif_table[vr.vifi]; - if (VIF_EXISTS(vr.vifi)) { + vif = &init_net.ipv4.vif_table[vr.vifi]; + if (VIF_EXISTS(&init_net, vr.vifi)) { vr.icount = vif->pkt_in; vr.ocount = vif->pkt_out; vr.ibytes = vif->bytes_in; @@ -1140,8 +1138,8 @@ static int ipmr_device_event(struct notifier_block *this, unsigned long event, v if (event != NETDEV_UNREGISTER) return NOTIFY_DONE; - v=&vif_table[0]; - for (ct=0; ctdev == dev) vif_delete(ct, 1); } @@ -1204,7 +1202,7 @@ static inline int ipmr_forward_finish(struct sk_buff *skb) static void ipmr_queue_xmit(struct sk_buff *skb, struct mfc_cache *c, int vifi) { const struct iphdr *iph = ip_hdr(skb); - struct vif_device *vif = &vif_table[vifi]; + struct vif_device *vif = &init_net.ipv4.vif_table[vifi]; struct net_device *dev; struct rtable *rt; int encap = 0; @@ -1305,8 +1303,8 @@ out_free: static int ipmr_find_vif(struct net_device *dev) { int ct; - for (ct=maxvif-1; ct>=0; ct--) { - if (vif_table[ct].dev == dev) + for (ct = init_net.ipv4.maxvif-1; ct >= 0; ct--) { + if (init_net.ipv4.vif_table[ct].dev == dev) break; } return ct; @@ -1326,7 +1324,7 @@ static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local /* * Wrong interface: drop packet and (maybe) send PIM assert. */ - if (vif_table[vif].dev != skb->dev) { + if (init_net.ipv4.vif_table[vif].dev != skb->dev) { int true_vifi; if (skb->rtable->fl.iif == 0) { @@ -1362,8 +1360,8 @@ static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local goto dont_forward; } - vif_table[vif].pkt_in++; - vif_table[vif].bytes_in += skb->len; + init_net.ipv4.vif_table[vif].pkt_in++; + init_net.ipv4.vif_table[vif].bytes_in += skb->len; /* * Forward the frame @@ -1500,7 +1498,7 @@ static int __pim_rcv(struct sk_buff *skb, unsigned int pimlen) read_lock(&mrt_lock); if (reg_vif_num >= 0) - reg_dev = vif_table[reg_vif_num].dev; + reg_dev = init_net.ipv4.vif_table[reg_vif_num].dev; if (reg_dev) dev_hold(reg_dev); read_unlock(&mrt_lock); @@ -1581,7 +1579,7 @@ ipmr_fill_mroute(struct sk_buff *skb, struct mfc_cache *c, struct rtmsg *rtm) { int ct; struct rtnexthop *nhp; - struct net_device *dev = vif_table[c->mfc_parent].dev; + struct net_device *dev = init_net.ipv4.vif_table[c->mfc_parent].dev; u8 *b = skb_tail_pointer(skb); struct rtattr *mp_head; @@ -1597,7 +1595,7 @@ ipmr_fill_mroute(struct sk_buff *skb, struct mfc_cache *c, struct rtmsg *rtm) nhp = (struct rtnexthop *)skb_put(skb, RTA_ALIGN(sizeof(*nhp))); nhp->rtnh_flags = 0; nhp->rtnh_hops = c->mfc_un.res.ttls[ct]; - nhp->rtnh_ifindex = vif_table[ct].dev->ifindex; + nhp->rtnh_ifindex = init_net.ipv4.vif_table[ct].dev->ifindex; nhp->rtnh_len = sizeof(*nhp); } } @@ -1672,11 +1670,11 @@ struct ipmr_vif_iter { static struct vif_device *ipmr_vif_seq_idx(struct ipmr_vif_iter *iter, loff_t pos) { - for (iter->ct = 0; iter->ct < maxvif; ++iter->ct) { - if (!VIF_EXISTS(iter->ct)) + for (iter->ct = 0; iter->ct < init_net.ipv4.maxvif; ++iter->ct) { + if (!VIF_EXISTS(&init_net, iter->ct)) continue; if (pos-- == 0) - return &vif_table[iter->ct]; + return &init_net.ipv4.vif_table[iter->ct]; } return NULL; } @@ -1697,10 +1695,10 @@ static void *ipmr_vif_seq_next(struct seq_file *seq, void *v, loff_t *pos) if (v == SEQ_START_TOKEN) return ipmr_vif_seq_idx(iter, 0); - while (++iter->ct < maxvif) { - if (!VIF_EXISTS(iter->ct)) + while (++iter->ct < init_net.ipv4.maxvif) { + if (!VIF_EXISTS(&init_net, iter->ct)) continue; - return &vif_table[iter->ct]; + return &init_net.ipv4.vif_table[iter->ct]; } return NULL; } @@ -1722,7 +1720,7 @@ static int ipmr_vif_seq_show(struct seq_file *seq, void *v) seq_printf(seq, "%2Zd %-10s %8ld %7ld %8ld %7ld %05X %08X %08X\n", - vif - vif_table, + vif - init_net.ipv4.vif_table, name, vif->bytes_in, vif->pkt_in, vif->bytes_out, vif->pkt_out, vif->flags, vif->local, vif->remote); @@ -1864,9 +1862,9 @@ static int ipmr_mfc_seq_show(struct seq_file *seq, void *v) mfc->mfc_un.res.wrong_if); for (n = mfc->mfc_un.res.minvif; n < mfc->mfc_un.res.maxvif; n++ ) { - if (VIF_EXISTS(n) - && mfc->mfc_un.res.ttls[n] < 255) - seq_printf(seq, + if (VIF_EXISTS(&init_net, n) && + mfc->mfc_un.res.ttls[n] < 255) + seq_printf(seq, " %2d:%-3d", n, mfc->mfc_un.res.ttls[n]); } @@ -1913,6 +1911,29 @@ static struct net_protocol pim_protocol = { /* * Setup for IP multicast routing */ +static int __net_init ipmr_net_init(struct net *net) +{ + int err = 0; + + net->ipv4.vif_table = kcalloc(MAXVIFS, sizeof(struct vif_device), + GFP_KERNEL); + if (!net->ipv4.vif_table) { + err = -ENOMEM; + goto fail; + } +fail: + return err; +} + +static void __net_exit ipmr_net_exit(struct net *net) +{ + kfree(net->ipv4.vif_table); +} + +static struct pernet_operations ipmr_net_ops = { + .init = ipmr_net_init, + .exit = ipmr_net_exit, +}; int __init ip_mr_init(void) { @@ -1925,6 +1946,10 @@ int __init ip_mr_init(void) if (!mrt_cachep) return -ENOMEM; + err = register_pernet_subsys(&ipmr_net_ops); + if (err) + goto reg_pernet_fail; + setup_timer(&ipmr_expire_timer, ipmr_expire_process, 0); err = register_netdevice_notifier(&ip_mr_notifier); if (err) @@ -1945,6 +1970,8 @@ proc_vif_fail: #endif reg_notif_fail: del_timer(&ipmr_expire_timer); + unregister_pernet_subsys(&ipmr_net_ops); +reg_pernet_fail: kmem_cache_destroy(mrt_cachep); return err; } -- cgit v1.2.2 From 5c0a66f5f3c9c59e2c341400048e2cff768e67a9 Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Thu, 22 Jan 2009 04:56:17 +0000 Subject: netns: ipmr: store netns in struct mfc_cache This patch stores into struct mfc_cache the network namespace each mfc_cache belongs to. The new member is mfc_net. mfc_net is assigned at cache allocation and doesn't change during the rest of the cache entry life. A new net parameter is added to ipmr_cache_alloc/ipmr_cache_alloc_unres. This will help to retrieve the current netns around the IPv4 multicast routing code. At the moment, all mfc_cache are allocated in init_net. Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- net/ipv4/ipmr.c | 26 +++++++++++++++++--------- 1 file changed, 17 insertions(+), 9 deletions(-) (limited to 'net/ipv4/ipmr.c') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 75a5f79cc226..8428a0fb5c10 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -327,6 +327,12 @@ static int vif_delete(int vifi, int notify) return 0; } +static inline void ipmr_cache_free(struct mfc_cache *c) +{ + release_net(mfc_net(c)); + kmem_cache_free(mrt_cachep, c); +} + /* Destroy an unresolved cache entry, killing queued skbs and reporting error to netlink readers. */ @@ -353,7 +359,7 @@ static void ipmr_destroy_unres(struct mfc_cache *c) kfree_skb(skb); } - kmem_cache_free(mrt_cachep, c); + ipmr_cache_free(c); } @@ -528,22 +534,24 @@ static struct mfc_cache *ipmr_cache_find(__be32 origin, __be32 mcastgrp) /* * Allocate a multicast cache entry */ -static struct mfc_cache *ipmr_cache_alloc(void) +static struct mfc_cache *ipmr_cache_alloc(struct net *net) { struct mfc_cache *c = kmem_cache_zalloc(mrt_cachep, GFP_KERNEL); if (c == NULL) return NULL; c->mfc_un.res.minvif = MAXVIFS; + mfc_net_set(c, net); return c; } -static struct mfc_cache *ipmr_cache_alloc_unres(void) +static struct mfc_cache *ipmr_cache_alloc_unres(struct net *net) { struct mfc_cache *c = kmem_cache_zalloc(mrt_cachep, GFP_ATOMIC); if (c == NULL) return NULL; skb_queue_head_init(&c->mfc_un.unres.unresolved); c->mfc_un.unres.expires = jiffies + 10*HZ; + mfc_net_set(c, net); return c; } @@ -695,7 +703,7 @@ ipmr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) */ if (atomic_read(&cache_resolve_queue_len) >= 10 || - (c=ipmr_cache_alloc_unres())==NULL) { + (c = ipmr_cache_alloc_unres(&init_net)) == NULL) { spin_unlock_bh(&mfc_unres_lock); kfree_skb(skb); @@ -718,7 +726,7 @@ ipmr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) */ spin_unlock_bh(&mfc_unres_lock); - kmem_cache_free(mrt_cachep, c); + ipmr_cache_free(c); kfree_skb(skb); return err; } @@ -763,7 +771,7 @@ static int ipmr_mfc_delete(struct mfcctl *mfc) *cp = c->next; write_unlock_bh(&mrt_lock); - kmem_cache_free(mrt_cachep, c); + ipmr_cache_free(c); return 0; } } @@ -796,7 +804,7 @@ static int ipmr_mfc_add(struct mfcctl *mfc, int mrtsock) if (!ipv4_is_multicast(mfc->mfcc_mcastgrp.s_addr)) return -EINVAL; - c = ipmr_cache_alloc(); + c = ipmr_cache_alloc(&init_net); if (c == NULL) return -ENOMEM; @@ -831,7 +839,7 @@ static int ipmr_mfc_add(struct mfcctl *mfc, int mrtsock) if (uc) { ipmr_cache_resolve(uc, c); - kmem_cache_free(mrt_cachep, uc); + ipmr_cache_free(uc); } return 0; } @@ -868,7 +876,7 @@ static void mroute_clean_tables(struct sock *sk) *cp = c->next; write_unlock_bh(&mrt_lock); - kmem_cache_free(mrt_cachep, c); + ipmr_cache_free(c); } } -- cgit v1.2.2 From 2bb8b26c3ea8bde1943dc5cd4dda2dc9f48fb281 Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Thu, 22 Jan 2009 04:56:18 +0000 Subject: netns: ipmr: dynamically allocate mfc_cache_array Preliminary work to make IPv4 multicast routing netns-aware. Dynamically allocate IPv4 multicast forwarding cache, mfc_cache_array, and move it to struct netns_ipv4. At the moment, mfc_cache_array is only referenced in init_net. Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- net/ipv4/ipmr.c | 41 ++++++++++++++++++++++++++++------------- 1 file changed, 28 insertions(+), 13 deletions(-) (limited to 'net/ipv4/ipmr.c') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 8428a0fb5c10..35b868dd3bfd 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -82,8 +82,6 @@ static DEFINE_RWLOCK(mrt_lock); static int mroute_do_assert; /* Set in PIM assert */ static int mroute_do_pim; -static struct mfc_cache *mfc_cache_array[MFC_LINES]; /* Forwarding cache */ - static struct mfc_cache *mfc_unres_queue; /* Queue of unresolved entries */ static atomic_t cache_resolve_queue_len; /* Size of unresolved */ @@ -524,7 +522,7 @@ static struct mfc_cache *ipmr_cache_find(__be32 origin, __be32 mcastgrp) int line = MFC_HASH(mcastgrp, origin); struct mfc_cache *c; - for (c=mfc_cache_array[line]; c; c = c->next) { + for (c = init_net.ipv4.mfc_cache_array[line]; c; c = c->next) { if (c->mfc_origin==origin && c->mfc_mcastgrp==mcastgrp) break; } @@ -764,7 +762,8 @@ static int ipmr_mfc_delete(struct mfcctl *mfc) line = MFC_HASH(mfc->mfcc_mcastgrp.s_addr, mfc->mfcc_origin.s_addr); - for (cp=&mfc_cache_array[line]; (c=*cp) != NULL; cp = &c->next) { + for (cp = &init_net.ipv4.mfc_cache_array[line]; + (c = *cp) != NULL; cp = &c->next) { if (c->mfc_origin == mfc->mfcc_origin.s_addr && c->mfc_mcastgrp == mfc->mfcc_mcastgrp.s_addr) { write_lock_bh(&mrt_lock); @@ -785,7 +784,8 @@ static int ipmr_mfc_add(struct mfcctl *mfc, int mrtsock) line = MFC_HASH(mfc->mfcc_mcastgrp.s_addr, mfc->mfcc_origin.s_addr); - for (cp=&mfc_cache_array[line]; (c=*cp) != NULL; cp = &c->next) { + for (cp = &init_net.ipv4.mfc_cache_array[line]; + (c = *cp) != NULL; cp = &c->next) { if (c->mfc_origin == mfc->mfcc_origin.s_addr && c->mfc_mcastgrp == mfc->mfcc_mcastgrp.s_addr) break; @@ -816,8 +816,8 @@ static int ipmr_mfc_add(struct mfcctl *mfc, int mrtsock) c->mfc_flags |= MFC_STATIC; write_lock_bh(&mrt_lock); - c->next = mfc_cache_array[line]; - mfc_cache_array[line] = c; + c->next = init_net.ipv4.mfc_cache_array[line]; + init_net.ipv4.mfc_cache_array[line] = c; write_unlock_bh(&mrt_lock); /* @@ -866,7 +866,7 @@ static void mroute_clean_tables(struct sock *sk) for (i=0; imfc_flags&MFC_STATIC) { cp = &c->next; @@ -1767,10 +1767,11 @@ static struct mfc_cache *ipmr_mfc_seq_idx(struct ipmr_mfc_iter *it, loff_t pos) { struct mfc_cache *mfc; - it->cache = mfc_cache_array; + it->cache = init_net.ipv4.mfc_cache_array; read_lock(&mrt_lock); for (it->ct = 0; it->ct < MFC_LINES; it->ct++) - for (mfc = mfc_cache_array[it->ct]; mfc; mfc = mfc->next) + for (mfc = init_net.ipv4.mfc_cache_array[it->ct]; + mfc; mfc = mfc->next) if (pos-- == 0) return mfc; read_unlock(&mrt_lock); @@ -1812,10 +1813,10 @@ static void *ipmr_mfc_seq_next(struct seq_file *seq, void *v, loff_t *pos) if (it->cache == &mfc_unres_queue) goto end_of_list; - BUG_ON(it->cache != mfc_cache_array); + BUG_ON(it->cache != init_net.ipv4.mfc_cache_array); while (++it->ct < MFC_LINES) { - mfc = mfc_cache_array[it->ct]; + mfc = init_net.ipv4.mfc_cache_array[it->ct]; if (mfc) return mfc; } @@ -1843,7 +1844,7 @@ static void ipmr_mfc_seq_stop(struct seq_file *seq, void *v) if (it->cache == &mfc_unres_queue) spin_unlock_bh(&mfc_unres_lock); - else if (it->cache == mfc_cache_array) + else if (it->cache == init_net.ipv4.mfc_cache_array) read_unlock(&mrt_lock); } @@ -1929,12 +1930,26 @@ static int __net_init ipmr_net_init(struct net *net) err = -ENOMEM; goto fail; } + + /* Forwarding cache */ + net->ipv4.mfc_cache_array = kcalloc(MFC_LINES, + sizeof(struct mfc_cache *), + GFP_KERNEL); + if (!net->ipv4.mfc_cache_array) { + err = -ENOMEM; + goto fail_mfc_cache; + } + return 0; + +fail_mfc_cache: + kfree(net->ipv4.vif_table); fail: return err; } static void __net_exit ipmr_net_exit(struct net *net) { + kfree(net->ipv4.mfc_cache_array); kfree(net->ipv4.vif_table); } -- cgit v1.2.2 From 1e8fb3b6a4ac6c5e486298d88289038456957545 Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Thu, 22 Jan 2009 04:56:19 +0000 Subject: netns: ipmr: declare counter cache_resolve_queue_len per-namespace Preliminary work to make IPv4 multicast routing netns-aware. Declare variable cache_resolve_queue_len per-namespace: move it into struct netns_ipv4. This variable counts the number of unresolved cache entries queued in the list mfc_unres_queue. This list is kept global to all netns as the number of entries per namespace is limited to 10 (hardcoded in routine ipmr_cache_unresolved). Entries belonging to different namespaces in mfc_unres_queue will be identified by matching the mfc_net member introduced previously in struct mfc_cache. Keeping this list global to all netns, also allows us to keep a single timer (ipmr_expire_timer) to handle their expiration. In some places cache_resolve_queue_len value was tested for arming or deleting the timer. These tests were equivalent to testing mfc_unres_queue value instead and are replaced in this patch. At the moment, cache_resolve_queue_len is only referenced in init_net. Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- net/ipv4/ipmr.c | 39 +++++++++++++++++++++------------------ 1 file changed, 21 insertions(+), 18 deletions(-) (limited to 'net/ipv4/ipmr.c') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 35b868dd3bfd..feafd14eb7b9 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -83,7 +83,6 @@ static int mroute_do_assert; /* Set in PIM assert */ static int mroute_do_pim; static struct mfc_cache *mfc_unres_queue; /* Queue of unresolved entries */ -static atomic_t cache_resolve_queue_len; /* Size of unresolved */ /* Special spinlock for queue of unresolved entries */ static DEFINE_SPINLOCK(mfc_unres_lock); @@ -340,7 +339,7 @@ static void ipmr_destroy_unres(struct mfc_cache *c) struct sk_buff *skb; struct nlmsgerr *e; - atomic_dec(&cache_resolve_queue_len); + atomic_dec(&init_net.ipv4.cache_resolve_queue_len); while ((skb = skb_dequeue(&c->mfc_un.unres.unresolved))) { if (ip_hdr(skb)->version == 0) { @@ -374,7 +373,7 @@ static void ipmr_expire_process(unsigned long dummy) return; } - if (atomic_read(&cache_resolve_queue_len) == 0) + if (mfc_unres_queue == NULL) goto out; now = jiffies; @@ -395,7 +394,7 @@ static void ipmr_expire_process(unsigned long dummy) ipmr_destroy_unres(c); } - if (atomic_read(&cache_resolve_queue_len)) + if (mfc_unres_queue != NULL) mod_timer(&ipmr_expire_timer, jiffies + expires); out: @@ -690,7 +689,8 @@ ipmr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) spin_lock_bh(&mfc_unres_lock); for (c=mfc_unres_queue; c; c=c->next) { - if (c->mfc_mcastgrp == iph->daddr && + if (net_eq(mfc_net(c), &init_net) && + c->mfc_mcastgrp == iph->daddr && c->mfc_origin == iph->saddr) break; } @@ -700,7 +700,7 @@ ipmr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) * Create a new entry if allowable */ - if (atomic_read(&cache_resolve_queue_len) >= 10 || + if (atomic_read(&init_net.ipv4.cache_resolve_queue_len) >= 10 || (c = ipmr_cache_alloc_unres(&init_net)) == NULL) { spin_unlock_bh(&mfc_unres_lock); @@ -729,7 +729,7 @@ ipmr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) return err; } - atomic_inc(&cache_resolve_queue_len); + atomic_inc(&init_net.ipv4.cache_resolve_queue_len); c->next = mfc_unres_queue; mfc_unres_queue = c; @@ -827,14 +827,16 @@ static int ipmr_mfc_add(struct mfcctl *mfc, int mrtsock) spin_lock_bh(&mfc_unres_lock); for (cp = &mfc_unres_queue; (uc=*cp) != NULL; cp = &uc->next) { - if (uc->mfc_origin == c->mfc_origin && + if (net_eq(mfc_net(uc), &init_net) && + uc->mfc_origin == c->mfc_origin && uc->mfc_mcastgrp == c->mfc_mcastgrp) { *cp = uc->next; - if (atomic_dec_and_test(&cache_resolve_queue_len)) - del_timer(&ipmr_expire_timer); + atomic_dec(&init_net.ipv4.cache_resolve_queue_len); break; } } + if (mfc_unres_queue == NULL) + del_timer(&ipmr_expire_timer); spin_unlock_bh(&mfc_unres_lock); if (uc) { @@ -880,18 +882,19 @@ static void mroute_clean_tables(struct sock *sk) } } - if (atomic_read(&cache_resolve_queue_len) != 0) { - struct mfc_cache *c; + if (atomic_read(&init_net.ipv4.cache_resolve_queue_len) != 0) { + struct mfc_cache *c, **cp; spin_lock_bh(&mfc_unres_lock); - while (mfc_unres_queue != NULL) { - c = mfc_unres_queue; - mfc_unres_queue = c->next; - spin_unlock_bh(&mfc_unres_lock); + cp = &mfc_unres_queue; + while ((c = *cp) != NULL) { + if (!net_eq(mfc_net(c), &init_net)) { + cp = &c->next; + continue; + } + *cp = c->next; ipmr_destroy_unres(c); - - spin_lock_bh(&mfc_unres_lock); } spin_unlock_bh(&mfc_unres_lock); } -- cgit v1.2.2 From 6f9374a9342e896c68df7cf7c0b039ab5cca994c Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Thu, 22 Jan 2009 04:56:20 +0000 Subject: netns: ipmr: declare mroute_do_assert and mroute_do_pim per-namespace Preliminary work to make IPv4 multicast routing netns-aware. Declare IPv multicast routing variables 'mroute_do_assert' and 'mroute_do_pim' per-namespace in struct netns_ipv4. At the moment, these variables are only referenced in init_net. Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- net/ipv4/ipmr.c | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-) (limited to 'net/ipv4/ipmr.c') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index feafd14eb7b9..d6a28acc0683 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -79,9 +79,6 @@ static DEFINE_RWLOCK(mrt_lock); #define VIF_EXISTS(_net, _idx) ((_net)->ipv4.vif_table[_idx].dev != NULL) -static int mroute_do_assert; /* Set in PIM assert */ -static int mroute_do_pim; - static struct mfc_cache *mfc_unres_queue; /* Queue of unresolved entries */ /* Special spinlock for queue of unresolved entries */ @@ -1003,7 +1000,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int int v; if (get_user(v,(int __user *)optval)) return -EFAULT; - mroute_do_assert=(v)?1:0; + init_net.ipv4.mroute_do_assert = (v) ? 1 : 0; return 0; } #ifdef CONFIG_IP_PIMSM @@ -1017,11 +1014,11 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int rtnl_lock(); ret = 0; - if (v != mroute_do_pim) { - mroute_do_pim = v; - mroute_do_assert = v; + if (v != init_net.ipv4.mroute_do_pim) { + init_net.ipv4.mroute_do_pim = v; + init_net.ipv4.mroute_do_assert = v; #ifdef CONFIG_IP_PIMSM_V2 - if (mroute_do_pim) + if (init_net.ipv4.mroute_do_pim) ret = inet_add_protocol(&pim_protocol, IPPROTO_PIM); else @@ -1073,10 +1070,10 @@ int ip_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, int val = 0x0305; #ifdef CONFIG_IP_PIMSM else if (optname == MRT_PIM) - val = mroute_do_pim; + val = init_net.ipv4.mroute_do_pim; #endif else - val = mroute_do_assert; + val = init_net.ipv4.mroute_do_assert; if (copy_to_user(optval, &val, olr)) return -EFAULT; return 0; @@ -1356,13 +1353,14 @@ static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local cache->mfc_un.res.wrong_if++; true_vifi = ipmr_find_vif(skb->dev); - if (true_vifi >= 0 && mroute_do_assert && + if (true_vifi >= 0 && init_net.ipv4.mroute_do_assert && /* pimsm uses asserts, when switching from RPT to SPT, so that we cannot check that packet arrived on an oif. It is bad, but otherwise we would need to move pretty large chunk of pimd to kernel. Ough... --ANK */ - (mroute_do_pim || cache->mfc_un.res.ttls[true_vifi] < 255) && + (init_net.ipv4.mroute_do_pim || + cache->mfc_un.res.ttls[true_vifi] < 255) && time_after(jiffies, cache->mfc_un.res.last_assert + MFC_ASSERT_THRESH)) { cache->mfc_un.res.last_assert = jiffies; @@ -1550,7 +1548,7 @@ int pim_rcv_v1(struct sk_buff * skb) pim = igmp_hdr(skb); - if (!mroute_do_pim || + if (!init_net.ipv4.mroute_do_pim || pim->group != PIM_V1_VERSION || pim->code != PIM_V1_REGISTER) goto drop; -- cgit v1.2.2 From 6c5143dbcfe50ac722965dc7d096abbeeec8bb33 Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Thu, 22 Jan 2009 04:56:21 +0000 Subject: netns: ipmr: declare reg_vif_num per-namespace Preliminary work to make IPv4 multicast routing netns-aware. Declare variable 'reg_vif_num' per-namespace, move into struct netns_ipv4. At the moment, this variable is only referenced in init_net. Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- net/ipv4/ipmr.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) (limited to 'net/ipv4/ipmr.c') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index d6a28acc0683..346e67b50d6c 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -197,14 +197,12 @@ failure: #ifdef CONFIG_IP_PIMSM -static int reg_vif_num = -1; - static int reg_vif_xmit(struct sk_buff *skb, struct net_device *dev) { read_lock(&mrt_lock); dev->stats.tx_bytes += skb->len; dev->stats.tx_packets++; - ipmr_cache_report(skb, reg_vif_num, IGMPMSG_WHOLEPKT); + ipmr_cache_report(skb, init_net.ipv4.mroute_reg_vif_num, IGMPMSG_WHOLEPKT); read_unlock(&mrt_lock); kfree_skb(skb); return 0; @@ -292,8 +290,8 @@ static int vif_delete(int vifi, int notify) } #ifdef CONFIG_IP_PIMSM - if (vifi == reg_vif_num) - reg_vif_num = -1; + if (vifi == init_net.ipv4.mroute_reg_vif_num) + init_net.ipv4.mroute_reg_vif_num = -1; #endif if (vifi+1 == init_net.ipv4.maxvif) { @@ -439,7 +437,7 @@ static int vif_add(struct vifctl *vifc, int mrtsock) * Special Purpose VIF in PIM * All the packets will be sent to the daemon */ - if (reg_vif_num >= 0) + if (init_net.ipv4.mroute_reg_vif_num >= 0) return -EADDRINUSE; dev = ipmr_reg_vif(); if (!dev) @@ -505,7 +503,7 @@ static int vif_add(struct vifctl *vifc, int mrtsock) v->dev = dev; #ifdef CONFIG_IP_PIMSM if (v->flags&VIFF_REGISTER) - reg_vif_num = vifi; + init_net.ipv4.mroute_reg_vif_num = vifi; #endif if (vifi+1 > init_net.ipv4.maxvif) init_net.ipv4.maxvif = vifi+1; @@ -623,7 +621,7 @@ static int ipmr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) memcpy(msg, skb_network_header(pkt), sizeof(struct iphdr)); msg->im_msgtype = IGMPMSG_WHOLEPKT; msg->im_mbz = 0; - msg->im_vif = reg_vif_num; + msg->im_vif = init_net.ipv4.mroute_reg_vif_num; ip_hdr(skb)->ihl = sizeof(struct iphdr) >> 2; ip_hdr(skb)->tot_len = htons(ntohs(ip_hdr(pkt)->tot_len) + sizeof(struct iphdr)); @@ -1506,8 +1504,8 @@ static int __pim_rcv(struct sk_buff *skb, unsigned int pimlen) return 1; read_lock(&mrt_lock); - if (reg_vif_num >= 0) - reg_dev = init_net.ipv4.vif_table[reg_vif_num].dev; + if (init_net.ipv4.mroute_reg_vif_num >= 0) + reg_dev = init_net.ipv4.vif_table[init_net.ipv4.mroute_reg_vif_num].dev; if (reg_dev) dev_hold(reg_dev); read_unlock(&mrt_lock); @@ -1940,6 +1938,10 @@ static int __net_init ipmr_net_init(struct net *net) err = -ENOMEM; goto fail_mfc_cache; } + +#ifdef CONFIG_IP_PIMSM + net->ipv4.mroute_reg_vif_num = -1; +#endif return 0; fail_mfc_cache: -- cgit v1.2.2 From f6bb451476be53d456e73bcfd82356afd680bbb0 Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Thu, 22 Jan 2009 04:56:22 +0000 Subject: netns: ipmr: declare ipmr /proc/net entries per-namespace Declare IPv4 multicast forwarding /proc/net entries per-namespace: /proc/net/ip_mr_vif /proc/net/ip_mr_cache Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- net/ipv4/ipmr.c | 101 ++++++++++++++++++++++++++++++++++---------------------- 1 file changed, 62 insertions(+), 39 deletions(-) (limited to 'net/ipv4/ipmr.c') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 346e67b50d6c..a4fd97f1920c 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1671,17 +1671,19 @@ int ipmr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait) * The /proc interfaces to multicast routing /proc/ip_mr_cache /proc/ip_mr_vif */ struct ipmr_vif_iter { + struct seq_net_private p; int ct; }; -static struct vif_device *ipmr_vif_seq_idx(struct ipmr_vif_iter *iter, +static struct vif_device *ipmr_vif_seq_idx(struct net *net, + struct ipmr_vif_iter *iter, loff_t pos) { - for (iter->ct = 0; iter->ct < init_net.ipv4.maxvif; ++iter->ct) { - if (!VIF_EXISTS(&init_net, iter->ct)) + for (iter->ct = 0; iter->ct < net->ipv4.maxvif; ++iter->ct) { + if (!VIF_EXISTS(net, iter->ct)) continue; if (pos-- == 0) - return &init_net.ipv4.vif_table[iter->ct]; + return &net->ipv4.vif_table[iter->ct]; } return NULL; } @@ -1689,23 +1691,26 @@ static struct vif_device *ipmr_vif_seq_idx(struct ipmr_vif_iter *iter, static void *ipmr_vif_seq_start(struct seq_file *seq, loff_t *pos) __acquires(mrt_lock) { + struct net *net = seq_file_net(seq); + read_lock(&mrt_lock); - return *pos ? ipmr_vif_seq_idx(seq->private, *pos - 1) + return *pos ? ipmr_vif_seq_idx(net, seq->private, *pos - 1) : SEQ_START_TOKEN; } static void *ipmr_vif_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct ipmr_vif_iter *iter = seq->private; + struct net *net = seq_file_net(seq); ++*pos; if (v == SEQ_START_TOKEN) - return ipmr_vif_seq_idx(iter, 0); + return ipmr_vif_seq_idx(net, iter, 0); - while (++iter->ct < init_net.ipv4.maxvif) { - if (!VIF_EXISTS(&init_net, iter->ct)) + while (++iter->ct < net->ipv4.maxvif) { + if (!VIF_EXISTS(net, iter->ct)) continue; - return &init_net.ipv4.vif_table[iter->ct]; + return &net->ipv4.vif_table[iter->ct]; } return NULL; } @@ -1718,6 +1723,8 @@ static void ipmr_vif_seq_stop(struct seq_file *seq, void *v) static int ipmr_vif_seq_show(struct seq_file *seq, void *v) { + struct net *net = seq_file_net(seq); + if (v == SEQ_START_TOKEN) { seq_puts(seq, "Interface BytesIn PktsIn BytesOut PktsOut Flags Local Remote\n"); @@ -1727,7 +1734,7 @@ static int ipmr_vif_seq_show(struct seq_file *seq, void *v) seq_printf(seq, "%2Zd %-10s %8ld %7ld %8ld %7ld %05X %08X %08X\n", - vif - init_net.ipv4.vif_table, + vif - net->ipv4.vif_table, name, vif->bytes_in, vif->pkt_in, vif->bytes_out, vif->pkt_out, vif->flags, vif->local, vif->remote); @@ -1744,8 +1751,8 @@ static const struct seq_operations ipmr_vif_seq_ops = { static int ipmr_vif_open(struct inode *inode, struct file *file) { - return seq_open_private(file, &ipmr_vif_seq_ops, - sizeof(struct ipmr_vif_iter)); + return seq_open_net(inode, file, &ipmr_vif_seq_ops, + sizeof(struct ipmr_vif_iter)); } static const struct file_operations ipmr_vif_fops = { @@ -1753,23 +1760,25 @@ static const struct file_operations ipmr_vif_fops = { .open = ipmr_vif_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release_private, + .release = seq_release_net, }; struct ipmr_mfc_iter { + struct seq_net_private p; struct mfc_cache **cache; int ct; }; -static struct mfc_cache *ipmr_mfc_seq_idx(struct ipmr_mfc_iter *it, loff_t pos) +static struct mfc_cache *ipmr_mfc_seq_idx(struct net *net, + struct ipmr_mfc_iter *it, loff_t pos) { struct mfc_cache *mfc; - it->cache = init_net.ipv4.mfc_cache_array; + it->cache = net->ipv4.mfc_cache_array; read_lock(&mrt_lock); for (it->ct = 0; it->ct < MFC_LINES; it->ct++) - for (mfc = init_net.ipv4.mfc_cache_array[it->ct]; + for (mfc = net->ipv4.mfc_cache_array[it->ct]; mfc; mfc = mfc->next) if (pos-- == 0) return mfc; @@ -1778,7 +1787,8 @@ static struct mfc_cache *ipmr_mfc_seq_idx(struct ipmr_mfc_iter *it, loff_t pos) it->cache = &mfc_unres_queue; spin_lock_bh(&mfc_unres_lock); for (mfc = mfc_unres_queue; mfc; mfc = mfc->next) - if (pos-- == 0) + if (net_eq(mfc_net(mfc), net) && + pos-- == 0) return mfc; spin_unlock_bh(&mfc_unres_lock); @@ -1790,9 +1800,11 @@ static struct mfc_cache *ipmr_mfc_seq_idx(struct ipmr_mfc_iter *it, loff_t pos) static void *ipmr_mfc_seq_start(struct seq_file *seq, loff_t *pos) { struct ipmr_mfc_iter *it = seq->private; + struct net *net = seq_file_net(seq); + it->cache = NULL; it->ct = 0; - return *pos ? ipmr_mfc_seq_idx(seq->private, *pos - 1) + return *pos ? ipmr_mfc_seq_idx(net, seq->private, *pos - 1) : SEQ_START_TOKEN; } @@ -1800,11 +1812,12 @@ static void *ipmr_mfc_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct mfc_cache *mfc = v; struct ipmr_mfc_iter *it = seq->private; + struct net *net = seq_file_net(seq); ++*pos; if (v == SEQ_START_TOKEN) - return ipmr_mfc_seq_idx(seq->private, 0); + return ipmr_mfc_seq_idx(net, seq->private, 0); if (mfc->next) return mfc->next; @@ -1812,10 +1825,10 @@ static void *ipmr_mfc_seq_next(struct seq_file *seq, void *v, loff_t *pos) if (it->cache == &mfc_unres_queue) goto end_of_list; - BUG_ON(it->cache != init_net.ipv4.mfc_cache_array); + BUG_ON(it->cache != net->ipv4.mfc_cache_array); while (++it->ct < MFC_LINES) { - mfc = init_net.ipv4.mfc_cache_array[it->ct]; + mfc = net->ipv4.mfc_cache_array[it->ct]; if (mfc) return mfc; } @@ -1827,6 +1840,8 @@ static void *ipmr_mfc_seq_next(struct seq_file *seq, void *v, loff_t *pos) spin_lock_bh(&mfc_unres_lock); mfc = mfc_unres_queue; + while (mfc && !net_eq(mfc_net(mfc), net)) + mfc = mfc->next; if (mfc) return mfc; @@ -1840,16 +1855,18 @@ static void *ipmr_mfc_seq_next(struct seq_file *seq, void *v, loff_t *pos) static void ipmr_mfc_seq_stop(struct seq_file *seq, void *v) { struct ipmr_mfc_iter *it = seq->private; + struct net *net = seq_file_net(seq); if (it->cache == &mfc_unres_queue) spin_unlock_bh(&mfc_unres_lock); - else if (it->cache == init_net.ipv4.mfc_cache_array) + else if (it->cache == net->ipv4.mfc_cache_array) read_unlock(&mrt_lock); } static int ipmr_mfc_seq_show(struct seq_file *seq, void *v) { int n; + struct net *net = seq_file_net(seq); if (v == SEQ_START_TOKEN) { seq_puts(seq, @@ -1870,7 +1887,7 @@ static int ipmr_mfc_seq_show(struct seq_file *seq, void *v) mfc->mfc_un.res.wrong_if); for (n = mfc->mfc_un.res.minvif; n < mfc->mfc_un.res.maxvif; n++ ) { - if (VIF_EXISTS(&init_net, n) && + if (VIF_EXISTS(net, n) && mfc->mfc_un.res.ttls[n] < 255) seq_printf(seq, " %2d:%-3d", @@ -1896,8 +1913,8 @@ static const struct seq_operations ipmr_mfc_seq_ops = { static int ipmr_mfc_open(struct inode *inode, struct file *file) { - return seq_open_private(file, &ipmr_mfc_seq_ops, - sizeof(struct ipmr_mfc_iter)); + return seq_open_net(inode, file, &ipmr_mfc_seq_ops, + sizeof(struct ipmr_mfc_iter)); } static const struct file_operations ipmr_mfc_fops = { @@ -1905,7 +1922,7 @@ static const struct file_operations ipmr_mfc_fops = { .open = ipmr_mfc_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release_private, + .release = seq_release_net, }; #endif @@ -1942,8 +1959,22 @@ static int __net_init ipmr_net_init(struct net *net) #ifdef CONFIG_IP_PIMSM net->ipv4.mroute_reg_vif_num = -1; #endif + +#ifdef CONFIG_PROC_FS + err = -ENOMEM; + if (!proc_net_fops_create(net, "ip_mr_vif", 0, &ipmr_vif_fops)) + goto proc_vif_fail; + if (!proc_net_fops_create(net, "ip_mr_cache", 0, &ipmr_mfc_fops)) + goto proc_cache_fail; +#endif return 0; +#ifdef CONFIG_PROC_FS +proc_cache_fail: + proc_net_remove(net, "ip_mr_vif"); +proc_vif_fail: + kfree(net->ipv4.mfc_cache_array); +#endif fail_mfc_cache: kfree(net->ipv4.vif_table); fail: @@ -1952,6 +1983,10 @@ fail: static void __net_exit ipmr_net_exit(struct net *net) { +#ifdef CONFIG_PROC_FS + proc_net_remove(net, "ip_mr_cache"); + proc_net_remove(net, "ip_mr_vif"); +#endif kfree(net->ipv4.mfc_cache_array); kfree(net->ipv4.vif_table); } @@ -1980,20 +2015,8 @@ int __init ip_mr_init(void) err = register_netdevice_notifier(&ip_mr_notifier); if (err) goto reg_notif_fail; -#ifdef CONFIG_PROC_FS - err = -ENOMEM; - if (!proc_net_fops_create(&init_net, "ip_mr_vif", 0, &ipmr_vif_fops)) - goto proc_vif_fail; - if (!proc_net_fops_create(&init_net, "ip_mr_cache", 0, &ipmr_mfc_fops)) - goto proc_cache_fail; -#endif return 0; -#ifdef CONFIG_PROC_FS -proc_cache_fail: - proc_net_remove(&init_net, "ip_mr_vif"); -proc_vif_fail: - unregister_netdevice_notifier(&ip_mr_notifier); -#endif + reg_notif_fail: del_timer(&ipmr_expire_timer); unregister_pernet_subsys(&ipmr_net_ops); -- cgit v1.2.2 From 4feb88e5c694bfe414cbc3ce0e383f7f7038f90b Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Thu, 22 Jan 2009 04:56:23 +0000 Subject: netns: ipmr: enable namespace support in ipv4 multicast routing code This last patch makes the appropriate changes to use and propagate the network namespace where needed in IPv4 multicast routing code. This consists mainly in replacing all the remaining init_net occurences with current netns pointer retrieved from sockets, net devices or mfc_caches depending on the routines' contexts. Some routines receive a new 'struct net' parameter to propagate the current netns: * vif_add/vif_delete * ipmr_new_tunnel * mroute_clean_tables * ipmr_cache_find * ipmr_cache_report * ipmr_cache_unresolved * ipmr_mfc_add/ipmr_mfc_delete * ipmr_get_route * rt_fill_info (in route.c) Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- net/ipv4/ipmr.c | 243 +++++++++++++++++++++++++++++++------------------------- 1 file changed, 135 insertions(+), 108 deletions(-) (limited to 'net/ipv4/ipmr.c') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index a4fd97f1920c..21a6dc710f20 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -95,7 +95,8 @@ static DEFINE_SPINLOCK(mfc_unres_lock); static struct kmem_cache *mrt_cachep __read_mostly; static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local); -static int ipmr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert); +static int ipmr_cache_report(struct net *net, + struct sk_buff *pkt, vifi_t vifi, int assert); static int ipmr_fill_mroute(struct sk_buff *skb, struct mfc_cache *c, struct rtmsg *rtm); #ifdef CONFIG_IP_PIMSM_V2 @@ -108,9 +109,11 @@ static struct timer_list ipmr_expire_timer; static void ipmr_del_tunnel(struct net_device *dev, struct vifctl *v) { + struct net *net = dev_net(dev); + dev_close(dev); - dev = __dev_get_by_name(&init_net, "tunl0"); + dev = __dev_get_by_name(net, "tunl0"); if (dev) { const struct net_device_ops *ops = dev->netdev_ops; struct ifreq ifr; @@ -136,11 +139,11 @@ static void ipmr_del_tunnel(struct net_device *dev, struct vifctl *v) } static -struct net_device *ipmr_new_tunnel(struct vifctl *v) +struct net_device *ipmr_new_tunnel(struct net *net, struct vifctl *v) { struct net_device *dev; - dev = __dev_get_by_name(&init_net, "tunl0"); + dev = __dev_get_by_name(net, "tunl0"); if (dev) { const struct net_device_ops *ops = dev->netdev_ops; @@ -169,7 +172,8 @@ struct net_device *ipmr_new_tunnel(struct vifctl *v) dev = NULL; - if (err == 0 && (dev = __dev_get_by_name(&init_net, p.name)) != NULL) { + if (err == 0 && + (dev = __dev_get_by_name(net, p.name)) != NULL) { dev->flags |= IFF_MULTICAST; in_dev = __in_dev_get_rtnl(dev); @@ -199,10 +203,13 @@ failure: static int reg_vif_xmit(struct sk_buff *skb, struct net_device *dev) { + struct net *net = dev_net(dev); + read_lock(&mrt_lock); dev->stats.tx_bytes += skb->len; dev->stats.tx_packets++; - ipmr_cache_report(skb, init_net.ipv4.mroute_reg_vif_num, IGMPMSG_WHOLEPKT); + ipmr_cache_report(net, skb, net->ipv4.mroute_reg_vif_num, + IGMPMSG_WHOLEPKT); read_unlock(&mrt_lock); kfree_skb(skb); return 0; @@ -269,16 +276,16 @@ failure: * @notify: Set to 1, if the caller is a notifier_call */ -static int vif_delete(int vifi, int notify) +static int vif_delete(struct net *net, int vifi, int notify) { struct vif_device *v; struct net_device *dev; struct in_device *in_dev; - if (vifi < 0 || vifi >= init_net.ipv4.maxvif) + if (vifi < 0 || vifi >= net->ipv4.maxvif) return -EADDRNOTAVAIL; - v = &init_net.ipv4.vif_table[vifi]; + v = &net->ipv4.vif_table[vifi]; write_lock_bh(&mrt_lock); dev = v->dev; @@ -290,17 +297,17 @@ static int vif_delete(int vifi, int notify) } #ifdef CONFIG_IP_PIMSM - if (vifi == init_net.ipv4.mroute_reg_vif_num) - init_net.ipv4.mroute_reg_vif_num = -1; + if (vifi == net->ipv4.mroute_reg_vif_num) + net->ipv4.mroute_reg_vif_num = -1; #endif - if (vifi+1 == init_net.ipv4.maxvif) { + if (vifi+1 == net->ipv4.maxvif) { int tmp; for (tmp=vifi-1; tmp>=0; tmp--) { - if (VIF_EXISTS(&init_net, tmp)) + if (VIF_EXISTS(net, tmp)) break; } - init_net.ipv4.maxvif = tmp+1; + net->ipv4.maxvif = tmp+1; } write_unlock_bh(&mrt_lock); @@ -333,8 +340,9 @@ static void ipmr_destroy_unres(struct mfc_cache *c) { struct sk_buff *skb; struct nlmsgerr *e; + struct net *net = mfc_net(c); - atomic_dec(&init_net.ipv4.cache_resolve_queue_len); + atomic_dec(&net->ipv4.cache_resolve_queue_len); while ((skb = skb_dequeue(&c->mfc_un.unres.unresolved))) { if (ip_hdr(skb)->version == 0) { @@ -346,7 +354,7 @@ static void ipmr_destroy_unres(struct mfc_cache *c) e->error = -ETIMEDOUT; memset(&e->msg, 0, sizeof(e->msg)); - rtnl_unicast(skb, &init_net, NETLINK_CB(skb).pid); + rtnl_unicast(skb, net, NETLINK_CB(skb).pid); } else kfree_skb(skb); } @@ -401,13 +409,14 @@ out: static void ipmr_update_thresholds(struct mfc_cache *cache, unsigned char *ttls) { int vifi; + struct net *net = mfc_net(cache); cache->mfc_un.res.minvif = MAXVIFS; cache->mfc_un.res.maxvif = 0; memset(cache->mfc_un.res.ttls, 255, MAXVIFS); - for (vifi = 0; vifi < init_net.ipv4.maxvif; vifi++) { - if (VIF_EXISTS(&init_net, vifi) && + for (vifi = 0; vifi < net->ipv4.maxvif; vifi++) { + if (VIF_EXISTS(net, vifi) && ttls[vifi] && ttls[vifi] < 255) { cache->mfc_un.res.ttls[vifi] = ttls[vifi]; if (cache->mfc_un.res.minvif > vifi) @@ -418,16 +427,16 @@ static void ipmr_update_thresholds(struct mfc_cache *cache, unsigned char *ttls) } } -static int vif_add(struct vifctl *vifc, int mrtsock) +static int vif_add(struct net *net, struct vifctl *vifc, int mrtsock) { int vifi = vifc->vifc_vifi; - struct vif_device *v = &init_net.ipv4.vif_table[vifi]; + struct vif_device *v = &net->ipv4.vif_table[vifi]; struct net_device *dev; struct in_device *in_dev; int err; /* Is vif busy ? */ - if (VIF_EXISTS(&init_net, vifi)) + if (VIF_EXISTS(net, vifi)) return -EADDRINUSE; switch (vifc->vifc_flags) { @@ -437,7 +446,7 @@ static int vif_add(struct vifctl *vifc, int mrtsock) * Special Purpose VIF in PIM * All the packets will be sent to the daemon */ - if (init_net.ipv4.mroute_reg_vif_num >= 0) + if (net->ipv4.mroute_reg_vif_num >= 0) return -EADDRINUSE; dev = ipmr_reg_vif(); if (!dev) @@ -451,7 +460,7 @@ static int vif_add(struct vifctl *vifc, int mrtsock) break; #endif case VIFF_TUNNEL: - dev = ipmr_new_tunnel(vifc); + dev = ipmr_new_tunnel(net, vifc); if (!dev) return -ENOBUFS; err = dev_set_allmulti(dev, 1); @@ -462,7 +471,7 @@ static int vif_add(struct vifctl *vifc, int mrtsock) } break; case 0: - dev = ip_dev_find(&init_net, vifc->vifc_lcl_addr.s_addr); + dev = ip_dev_find(net, vifc->vifc_lcl_addr.s_addr); if (!dev) return -EADDRNOTAVAIL; err = dev_set_allmulti(dev, 1); @@ -503,20 +512,22 @@ static int vif_add(struct vifctl *vifc, int mrtsock) v->dev = dev; #ifdef CONFIG_IP_PIMSM if (v->flags&VIFF_REGISTER) - init_net.ipv4.mroute_reg_vif_num = vifi; + net->ipv4.mroute_reg_vif_num = vifi; #endif - if (vifi+1 > init_net.ipv4.maxvif) - init_net.ipv4.maxvif = vifi+1; + if (vifi+1 > net->ipv4.maxvif) + net->ipv4.maxvif = vifi+1; write_unlock_bh(&mrt_lock); return 0; } -static struct mfc_cache *ipmr_cache_find(__be32 origin, __be32 mcastgrp) +static struct mfc_cache *ipmr_cache_find(struct net *net, + __be32 origin, + __be32 mcastgrp) { int line = MFC_HASH(mcastgrp, origin); struct mfc_cache *c; - for (c = init_net.ipv4.mfc_cache_array[line]; c; c = c->next) { + for (c = net->ipv4.mfc_cache_array[line]; c; c = c->next) { if (c->mfc_origin==origin && c->mfc_mcastgrp==mcastgrp) break; } @@ -576,7 +587,7 @@ static void ipmr_cache_resolve(struct mfc_cache *uc, struct mfc_cache *c) memset(&e->msg, 0, sizeof(e->msg)); } - rtnl_unicast(skb, &init_net, NETLINK_CB(skb).pid); + rtnl_unicast(skb, mfc_net(c), NETLINK_CB(skb).pid); } else ip_mr_forward(skb, c, 0); } @@ -589,7 +600,8 @@ static void ipmr_cache_resolve(struct mfc_cache *uc, struct mfc_cache *c) * Called under mrt_lock. */ -static int ipmr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) +static int ipmr_cache_report(struct net *net, + struct sk_buff *pkt, vifi_t vifi, int assert) { struct sk_buff *skb; const int ihl = ip_hdrlen(pkt); @@ -621,7 +633,7 @@ static int ipmr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) memcpy(msg, skb_network_header(pkt), sizeof(struct iphdr)); msg->im_msgtype = IGMPMSG_WHOLEPKT; msg->im_mbz = 0; - msg->im_vif = init_net.ipv4.mroute_reg_vif_num; + msg->im_vif = net->ipv4.mroute_reg_vif_num; ip_hdr(skb)->ihl = sizeof(struct iphdr) >> 2; ip_hdr(skb)->tot_len = htons(ntohs(ip_hdr(pkt)->tot_len) + sizeof(struct iphdr)); @@ -653,7 +665,7 @@ static int ipmr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) skb->transport_header = skb->network_header; } - if (init_net.ipv4.mroute_sk == NULL) { + if (net->ipv4.mroute_sk == NULL) { kfree_skb(skb); return -EINVAL; } @@ -661,7 +673,7 @@ static int ipmr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) /* * Deliver to mrouted */ - ret = sock_queue_rcv_skb(init_net.ipv4.mroute_sk, skb); + ret = sock_queue_rcv_skb(net->ipv4.mroute_sk, skb); if (ret < 0) { if (net_ratelimit()) printk(KERN_WARNING "mroute: pending queue full, dropping entries.\n"); @@ -676,7 +688,7 @@ static int ipmr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) */ static int -ipmr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) +ipmr_cache_unresolved(struct net *net, vifi_t vifi, struct sk_buff *skb) { int err; struct mfc_cache *c; @@ -684,7 +696,7 @@ ipmr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) spin_lock_bh(&mfc_unres_lock); for (c=mfc_unres_queue; c; c=c->next) { - if (net_eq(mfc_net(c), &init_net) && + if (net_eq(mfc_net(c), net) && c->mfc_mcastgrp == iph->daddr && c->mfc_origin == iph->saddr) break; @@ -695,8 +707,8 @@ ipmr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) * Create a new entry if allowable */ - if (atomic_read(&init_net.ipv4.cache_resolve_queue_len) >= 10 || - (c = ipmr_cache_alloc_unres(&init_net)) == NULL) { + if (atomic_read(&net->ipv4.cache_resolve_queue_len) >= 10 || + (c = ipmr_cache_alloc_unres(net)) == NULL) { spin_unlock_bh(&mfc_unres_lock); kfree_skb(skb); @@ -713,7 +725,8 @@ ipmr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) /* * Reflect first query at mrouted. */ - if ((err = ipmr_cache_report(skb, vifi, IGMPMSG_NOCACHE))<0) { + err = ipmr_cache_report(net, skb, vifi, IGMPMSG_NOCACHE); + if (err < 0) { /* If the report failed throw the cache entry out - Brad Parker */ @@ -724,7 +737,7 @@ ipmr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) return err; } - atomic_inc(&init_net.ipv4.cache_resolve_queue_len); + atomic_inc(&net->ipv4.cache_resolve_queue_len); c->next = mfc_unres_queue; mfc_unres_queue = c; @@ -750,14 +763,14 @@ ipmr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) * MFC cache manipulation by user space mroute daemon */ -static int ipmr_mfc_delete(struct mfcctl *mfc) +static int ipmr_mfc_delete(struct net *net, struct mfcctl *mfc) { int line; struct mfc_cache *c, **cp; line = MFC_HASH(mfc->mfcc_mcastgrp.s_addr, mfc->mfcc_origin.s_addr); - for (cp = &init_net.ipv4.mfc_cache_array[line]; + for (cp = &net->ipv4.mfc_cache_array[line]; (c = *cp) != NULL; cp = &c->next) { if (c->mfc_origin == mfc->mfcc_origin.s_addr && c->mfc_mcastgrp == mfc->mfcc_mcastgrp.s_addr) { @@ -772,14 +785,14 @@ static int ipmr_mfc_delete(struct mfcctl *mfc) return -ENOENT; } -static int ipmr_mfc_add(struct mfcctl *mfc, int mrtsock) +static int ipmr_mfc_add(struct net *net, struct mfcctl *mfc, int mrtsock) { int line; struct mfc_cache *uc, *c, **cp; line = MFC_HASH(mfc->mfcc_mcastgrp.s_addr, mfc->mfcc_origin.s_addr); - for (cp = &init_net.ipv4.mfc_cache_array[line]; + for (cp = &net->ipv4.mfc_cache_array[line]; (c = *cp) != NULL; cp = &c->next) { if (c->mfc_origin == mfc->mfcc_origin.s_addr && c->mfc_mcastgrp == mfc->mfcc_mcastgrp.s_addr) @@ -799,7 +812,7 @@ static int ipmr_mfc_add(struct mfcctl *mfc, int mrtsock) if (!ipv4_is_multicast(mfc->mfcc_mcastgrp.s_addr)) return -EINVAL; - c = ipmr_cache_alloc(&init_net); + c = ipmr_cache_alloc(net); if (c == NULL) return -ENOMEM; @@ -811,8 +824,8 @@ static int ipmr_mfc_add(struct mfcctl *mfc, int mrtsock) c->mfc_flags |= MFC_STATIC; write_lock_bh(&mrt_lock); - c->next = init_net.ipv4.mfc_cache_array[line]; - init_net.ipv4.mfc_cache_array[line] = c; + c->next = net->ipv4.mfc_cache_array[line]; + net->ipv4.mfc_cache_array[line] = c; write_unlock_bh(&mrt_lock); /* @@ -822,11 +835,11 @@ static int ipmr_mfc_add(struct mfcctl *mfc, int mrtsock) spin_lock_bh(&mfc_unres_lock); for (cp = &mfc_unres_queue; (uc=*cp) != NULL; cp = &uc->next) { - if (net_eq(mfc_net(uc), &init_net) && + if (net_eq(mfc_net(uc), net) && uc->mfc_origin == c->mfc_origin && uc->mfc_mcastgrp == c->mfc_mcastgrp) { *cp = uc->next; - atomic_dec(&init_net.ipv4.cache_resolve_queue_len); + atomic_dec(&net->ipv4.cache_resolve_queue_len); break; } } @@ -845,16 +858,16 @@ static int ipmr_mfc_add(struct mfcctl *mfc, int mrtsock) * Close the multicast socket, and clear the vif tables etc */ -static void mroute_clean_tables(struct sock *sk) +static void mroute_clean_tables(struct net *net) { int i; /* * Shut down all active vif entries */ - for (i = 0; i < init_net.ipv4.maxvif; i++) { - if (!(init_net.ipv4.vif_table[i].flags&VIFF_STATIC)) - vif_delete(i, 0); + for (i = 0; i < net->ipv4.maxvif; i++) { + if (!(net->ipv4.vif_table[i].flags&VIFF_STATIC)) + vif_delete(net, i, 0); } /* @@ -863,7 +876,7 @@ static void mroute_clean_tables(struct sock *sk) for (i=0; iipv4.mfc_cache_array[i]; while ((c = *cp) != NULL) { if (c->mfc_flags&MFC_STATIC) { cp = &c->next; @@ -877,13 +890,13 @@ static void mroute_clean_tables(struct sock *sk) } } - if (atomic_read(&init_net.ipv4.cache_resolve_queue_len) != 0) { + if (atomic_read(&net->ipv4.cache_resolve_queue_len) != 0) { struct mfc_cache *c, **cp; spin_lock_bh(&mfc_unres_lock); cp = &mfc_unres_queue; while ((c = *cp) != NULL) { - if (!net_eq(mfc_net(c), &init_net)) { + if (!net_eq(mfc_net(c), net)) { cp = &c->next; continue; } @@ -897,15 +910,17 @@ static void mroute_clean_tables(struct sock *sk) static void mrtsock_destruct(struct sock *sk) { + struct net *net = sock_net(sk); + rtnl_lock(); - if (sk == init_net.ipv4.mroute_sk) { - IPV4_DEVCONF_ALL(sock_net(sk), MC_FORWARDING)--; + if (sk == net->ipv4.mroute_sk) { + IPV4_DEVCONF_ALL(net, MC_FORWARDING)--; write_lock_bh(&mrt_lock); - init_net.ipv4.mroute_sk = NULL; + net->ipv4.mroute_sk = NULL; write_unlock_bh(&mrt_lock); - mroute_clean_tables(sk); + mroute_clean_tables(net); } rtnl_unlock(); } @@ -922,9 +937,10 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int int ret; struct vifctl vif; struct mfcctl mfc; + struct net *net = sock_net(sk); if (optname != MRT_INIT) { - if (sk != init_net.ipv4.mroute_sk && !capable(CAP_NET_ADMIN)) + if (sk != net->ipv4.mroute_sk && !capable(CAP_NET_ADMIN)) return -EACCES; } @@ -937,7 +953,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int return -ENOPROTOOPT; rtnl_lock(); - if (init_net.ipv4.mroute_sk) { + if (net->ipv4.mroute_sk) { rtnl_unlock(); return -EADDRINUSE; } @@ -945,15 +961,15 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int ret = ip_ra_control(sk, 1, mrtsock_destruct); if (ret == 0) { write_lock_bh(&mrt_lock); - init_net.ipv4.mroute_sk = sk; + net->ipv4.mroute_sk = sk; write_unlock_bh(&mrt_lock); - IPV4_DEVCONF_ALL(sock_net(sk), MC_FORWARDING)++; + IPV4_DEVCONF_ALL(net, MC_FORWARDING)++; } rtnl_unlock(); return ret; case MRT_DONE: - if (sk != init_net.ipv4.mroute_sk) + if (sk != net->ipv4.mroute_sk) return -EACCES; return ip_ra_control(sk, 0, NULL); case MRT_ADD_VIF: @@ -966,9 +982,9 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int return -ENFILE; rtnl_lock(); if (optname == MRT_ADD_VIF) { - ret = vif_add(&vif, sk == init_net.ipv4.mroute_sk); + ret = vif_add(net, &vif, sk == net->ipv4.mroute_sk); } else { - ret = vif_delete(vif.vifc_vifi, 0); + ret = vif_delete(net, vif.vifc_vifi, 0); } rtnl_unlock(); return ret; @@ -985,9 +1001,9 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int return -EFAULT; rtnl_lock(); if (optname == MRT_DEL_MFC) - ret = ipmr_mfc_delete(&mfc); + ret = ipmr_mfc_delete(net, &mfc); else - ret = ipmr_mfc_add(&mfc, sk == init_net.ipv4.mroute_sk); + ret = ipmr_mfc_add(net, &mfc, sk == net->ipv4.mroute_sk); rtnl_unlock(); return ret; /* @@ -998,7 +1014,7 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int int v; if (get_user(v,(int __user *)optval)) return -EFAULT; - init_net.ipv4.mroute_do_assert = (v) ? 1 : 0; + net->ipv4.mroute_do_assert = (v) ? 1 : 0; return 0; } #ifdef CONFIG_IP_PIMSM @@ -1012,11 +1028,11 @@ int ip_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int rtnl_lock(); ret = 0; - if (v != init_net.ipv4.mroute_do_pim) { - init_net.ipv4.mroute_do_pim = v; - init_net.ipv4.mroute_do_assert = v; + if (v != net->ipv4.mroute_do_pim) { + net->ipv4.mroute_do_pim = v; + net->ipv4.mroute_do_assert = v; #ifdef CONFIG_IP_PIMSM_V2 - if (init_net.ipv4.mroute_do_pim) + if (net->ipv4.mroute_do_pim) ret = inet_add_protocol(&pim_protocol, IPPROTO_PIM); else @@ -1047,6 +1063,7 @@ int ip_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, int { int olr; int val; + struct net *net = sock_net(sk); if (optname != MRT_VERSION && #ifdef CONFIG_IP_PIMSM @@ -1068,10 +1085,10 @@ int ip_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, int val = 0x0305; #ifdef CONFIG_IP_PIMSM else if (optname == MRT_PIM) - val = init_net.ipv4.mroute_do_pim; + val = net->ipv4.mroute_do_pim; #endif else - val = init_net.ipv4.mroute_do_assert; + val = net->ipv4.mroute_do_assert; if (copy_to_user(optval, &val, olr)) return -EFAULT; return 0; @@ -1087,16 +1104,17 @@ int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg) struct sioc_vif_req vr; struct vif_device *vif; struct mfc_cache *c; + struct net *net = sock_net(sk); switch (cmd) { case SIOCGETVIFCNT: if (copy_from_user(&vr, arg, sizeof(vr))) return -EFAULT; - if (vr.vifi >= init_net.ipv4.maxvif) + if (vr.vifi >= net->ipv4.maxvif) return -EINVAL; read_lock(&mrt_lock); - vif = &init_net.ipv4.vif_table[vr.vifi]; - if (VIF_EXISTS(&init_net, vr.vifi)) { + vif = &net->ipv4.vif_table[vr.vifi]; + if (VIF_EXISTS(net, vr.vifi)) { vr.icount = vif->pkt_in; vr.ocount = vif->pkt_out; vr.ibytes = vif->bytes_in; @@ -1114,7 +1132,7 @@ int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg) return -EFAULT; read_lock(&mrt_lock); - c = ipmr_cache_find(sr.src.s_addr, sr.grp.s_addr); + c = ipmr_cache_find(net, sr.src.s_addr, sr.grp.s_addr); if (c) { sr.pktcnt = c->mfc_un.res.pkt; sr.bytecnt = c->mfc_un.res.bytes; @@ -1136,18 +1154,19 @@ int ipmr_ioctl(struct sock *sk, int cmd, void __user *arg) static int ipmr_device_event(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = ptr; + struct net *net = dev_net(dev); struct vif_device *v; int ct; - if (!net_eq(dev_net(dev), &init_net)) + if (!net_eq(dev_net(dev), net)) return NOTIFY_DONE; if (event != NETDEV_UNREGISTER) return NOTIFY_DONE; - v = &init_net.ipv4.vif_table[0]; - for (ct = 0; ct < init_net.ipv4.maxvif; ct++, v++) { + v = &net->ipv4.vif_table[0]; + for (ct = 0; ct < net->ipv4.maxvif; ct++, v++) { if (v->dev == dev) - vif_delete(ct, 1); + vif_delete(net, ct, 1); } return NOTIFY_DONE; } @@ -1207,8 +1226,9 @@ static inline int ipmr_forward_finish(struct sk_buff *skb) static void ipmr_queue_xmit(struct sk_buff *skb, struct mfc_cache *c, int vifi) { + struct net *net = mfc_net(c); const struct iphdr *iph = ip_hdr(skb); - struct vif_device *vif = &init_net.ipv4.vif_table[vifi]; + struct vif_device *vif = &net->ipv4.vif_table[vifi]; struct net_device *dev; struct rtable *rt; int encap = 0; @@ -1222,7 +1242,7 @@ static void ipmr_queue_xmit(struct sk_buff *skb, struct mfc_cache *c, int vifi) vif->bytes_out += skb->len; vif->dev->stats.tx_bytes += skb->len; vif->dev->stats.tx_packets++; - ipmr_cache_report(skb, vifi, IGMPMSG_WHOLEPKT); + ipmr_cache_report(net, skb, vifi, IGMPMSG_WHOLEPKT); kfree_skb(skb); return; } @@ -1235,7 +1255,7 @@ static void ipmr_queue_xmit(struct sk_buff *skb, struct mfc_cache *c, int vifi) .saddr = vif->local, .tos = RT_TOS(iph->tos) } }, .proto = IPPROTO_IPIP }; - if (ip_route_output_key(&init_net, &rt, &fl)) + if (ip_route_output_key(net, &rt, &fl)) goto out_free; encap = sizeof(struct iphdr); } else { @@ -1244,7 +1264,7 @@ static void ipmr_queue_xmit(struct sk_buff *skb, struct mfc_cache *c, int vifi) { .daddr = iph->daddr, .tos = RT_TOS(iph->tos) } }, .proto = IPPROTO_IPIP }; - if (ip_route_output_key(&init_net, &rt, &fl)) + if (ip_route_output_key(net, &rt, &fl)) goto out_free; } @@ -1308,9 +1328,10 @@ out_free: static int ipmr_find_vif(struct net_device *dev) { + struct net *net = dev_net(dev); int ct; - for (ct = init_net.ipv4.maxvif-1; ct >= 0; ct--) { - if (init_net.ipv4.vif_table[ct].dev == dev) + for (ct = net->ipv4.maxvif-1; ct >= 0; ct--) { + if (net->ipv4.vif_table[ct].dev == dev) break; } return ct; @@ -1322,6 +1343,7 @@ static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local { int psend = -1; int vif, ct; + struct net *net = mfc_net(cache); vif = cache->mfc_parent; cache->mfc_un.res.pkt++; @@ -1330,7 +1352,7 @@ static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local /* * Wrong interface: drop packet and (maybe) send PIM assert. */ - if (init_net.ipv4.vif_table[vif].dev != skb->dev) { + if (net->ipv4.vif_table[vif].dev != skb->dev) { int true_vifi; if (skb->rtable->fl.iif == 0) { @@ -1351,24 +1373,24 @@ static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local cache->mfc_un.res.wrong_if++; true_vifi = ipmr_find_vif(skb->dev); - if (true_vifi >= 0 && init_net.ipv4.mroute_do_assert && + if (true_vifi >= 0 && net->ipv4.mroute_do_assert && /* pimsm uses asserts, when switching from RPT to SPT, so that we cannot check that packet arrived on an oif. It is bad, but otherwise we would need to move pretty large chunk of pimd to kernel. Ough... --ANK */ - (init_net.ipv4.mroute_do_pim || + (net->ipv4.mroute_do_pim || cache->mfc_un.res.ttls[true_vifi] < 255) && time_after(jiffies, cache->mfc_un.res.last_assert + MFC_ASSERT_THRESH)) { cache->mfc_un.res.last_assert = jiffies; - ipmr_cache_report(skb, true_vifi, IGMPMSG_WRONGVIF); + ipmr_cache_report(net, skb, true_vifi, IGMPMSG_WRONGVIF); } goto dont_forward; } - init_net.ipv4.vif_table[vif].pkt_in++; - init_net.ipv4.vif_table[vif].bytes_in += skb->len; + net->ipv4.vif_table[vif].pkt_in++; + net->ipv4.vif_table[vif].bytes_in += skb->len; /* * Forward the frame @@ -1408,6 +1430,7 @@ dont_forward: int ip_mr_input(struct sk_buff *skb) { struct mfc_cache *cache; + struct net *net = dev_net(skb->dev); int local = skb->rtable->rt_flags&RTCF_LOCAL; /* Packet is looped back after forward, it should not be @@ -1428,9 +1451,9 @@ int ip_mr_input(struct sk_buff *skb) that we can forward NO IGMP messages. */ read_lock(&mrt_lock); - if (init_net.ipv4.mroute_sk) { + if (net->ipv4.mroute_sk) { nf_reset(skb); - raw_rcv(init_net.ipv4.mroute_sk, skb); + raw_rcv(net->ipv4.mroute_sk, skb); read_unlock(&mrt_lock); return 0; } @@ -1439,7 +1462,7 @@ int ip_mr_input(struct sk_buff *skb) } read_lock(&mrt_lock); - cache = ipmr_cache_find(ip_hdr(skb)->saddr, ip_hdr(skb)->daddr); + cache = ipmr_cache_find(net, ip_hdr(skb)->saddr, ip_hdr(skb)->daddr); /* * No usable cache entry @@ -1459,7 +1482,7 @@ int ip_mr_input(struct sk_buff *skb) vif = ipmr_find_vif(skb->dev); if (vif >= 0) { - int err = ipmr_cache_unresolved(vif, skb); + int err = ipmr_cache_unresolved(net, vif, skb); read_unlock(&mrt_lock); return err; @@ -1490,6 +1513,7 @@ static int __pim_rcv(struct sk_buff *skb, unsigned int pimlen) { struct net_device *reg_dev = NULL; struct iphdr *encap; + struct net *net = dev_net(skb->dev); encap = (struct iphdr *)(skb_transport_header(skb) + pimlen); /* @@ -1504,8 +1528,8 @@ static int __pim_rcv(struct sk_buff *skb, unsigned int pimlen) return 1; read_lock(&mrt_lock); - if (init_net.ipv4.mroute_reg_vif_num >= 0) - reg_dev = init_net.ipv4.vif_table[init_net.ipv4.mroute_reg_vif_num].dev; + if (net->ipv4.mroute_reg_vif_num >= 0) + reg_dev = net->ipv4.vif_table[net->ipv4.mroute_reg_vif_num].dev; if (reg_dev) dev_hold(reg_dev); read_unlock(&mrt_lock); @@ -1540,13 +1564,14 @@ static int __pim_rcv(struct sk_buff *skb, unsigned int pimlen) int pim_rcv_v1(struct sk_buff * skb) { struct igmphdr *pim; + struct net *net = dev_net(skb->dev); if (!pskb_may_pull(skb, sizeof(*pim) + sizeof(struct iphdr))) goto drop; pim = igmp_hdr(skb); - if (!init_net.ipv4.mroute_do_pim || + if (!net->ipv4.mroute_do_pim || pim->group != PIM_V1_VERSION || pim->code != PIM_V1_REGISTER) goto drop; @@ -1586,7 +1611,8 @@ ipmr_fill_mroute(struct sk_buff *skb, struct mfc_cache *c, struct rtmsg *rtm) { int ct; struct rtnexthop *nhp; - struct net_device *dev = init_net.ipv4.vif_table[c->mfc_parent].dev; + struct net *net = mfc_net(c); + struct net_device *dev = net->ipv4.vif_table[c->mfc_parent].dev; u8 *b = skb_tail_pointer(skb); struct rtattr *mp_head; @@ -1602,7 +1628,7 @@ ipmr_fill_mroute(struct sk_buff *skb, struct mfc_cache *c, struct rtmsg *rtm) nhp = (struct rtnexthop *)skb_put(skb, RTA_ALIGN(sizeof(*nhp))); nhp->rtnh_flags = 0; nhp->rtnh_hops = c->mfc_un.res.ttls[ct]; - nhp->rtnh_ifindex = init_net.ipv4.vif_table[ct].dev->ifindex; + nhp->rtnh_ifindex = net->ipv4.vif_table[ct].dev->ifindex; nhp->rtnh_len = sizeof(*nhp); } } @@ -1616,14 +1642,15 @@ rtattr_failure: return -EMSGSIZE; } -int ipmr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait) +int ipmr_get_route(struct net *net, + struct sk_buff *skb, struct rtmsg *rtm, int nowait) { int err; struct mfc_cache *cache; struct rtable *rt = skb->rtable; read_lock(&mrt_lock); - cache = ipmr_cache_find(rt->rt_src, rt->rt_dst); + cache = ipmr_cache_find(net, rt->rt_src, rt->rt_dst); if (cache == NULL) { struct sk_buff *skb2; @@ -1654,7 +1681,7 @@ int ipmr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait) iph->saddr = rt->rt_src; iph->daddr = rt->rt_dst; iph->version = 0; - err = ipmr_cache_unresolved(vif, skb2); + err = ipmr_cache_unresolved(net, vif, skb2); read_unlock(&mrt_lock); return err; } -- cgit v1.2.2 From 69ebbf58f3dff9fb4e5240e472b5869fa869dae1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Fri, 6 Feb 2009 23:46:51 -0800 Subject: ipmr: use goto to common label instead of opencoding MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller --- net/ipv4/ipmr.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net/ipv4/ipmr.c') diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 21a6dc710f20..13e9dd3012b3 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1243,8 +1243,7 @@ static void ipmr_queue_xmit(struct sk_buff *skb, struct mfc_cache *c, int vifi) vif->dev->stats.tx_bytes += skb->len; vif->dev->stats.tx_packets++; ipmr_cache_report(net, skb, vifi, IGMPMSG_WHOLEPKT); - kfree_skb(skb); - return; + goto out_free; } #endif -- cgit v1.2.2