Merge /home/trondmy/scm/kernel/git/torvalds/linux-2.6

7 files changed, 79 insertions, 110 deletions

diff --git a/net/core/datagram.c b/net/core/datagram.c
index da9bf71421a7..81987df536eb 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -211,74 +211,45 @@ void skb_free_datagram(struct sock *sk, struct sk_buff *skb)
 int skb_copy_datagram_iovec(const struct sk_buff *skb, int offset,
                             struct iovec *to, int len)
 {
-        int start = skb_headlen(skb);
-        int i, copy = start - offset;
-
-        /* Copy header. */
-        if (copy > 0) {
-                if (copy > len)
-                        copy = len;
-                if (memcpy_toiovec(to, skb->data + offset, copy))
-                        goto fault;
-                if ((len -= copy) == 0)
-                        return 0;
-                offset += copy;
-        }
+        int i, err, fraglen, end = 0;
+        struct sk_buff *next = skb_shinfo(skb)->frag_list;
+next_skb:
+        fraglen = skb_headlen(skb);
+        i = -1;
 
-        /* Copy paged appendix. Hmm... why does this look so complicated? */
-        for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
-                int end;
-
-                BUG_TRAP(start <= offset + len);
+        while (1) {
+                int start = end;
 
-                end = start + skb_shinfo(skb)->frags[i].size;
-                if ((copy = end - offset) > 0) {
-                        int err;
-                        u8  *vaddr;
-                        skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
-                        struct page *page = frag->page;
+                if ((end += fraglen) > offset) {
+                        int copy = end - offset, o = offset - start;
 
                         if (copy > len)
                                 copy = len;
-                        vaddr = kmap(page);
-                        err = memcpy_toiovec(to, vaddr + frag->page_offset +
-                                             offset - start, copy);
-                        kunmap(page);
+                        if (i == -1)
+                                err = memcpy_toiovec(to, skb->data + o, copy);
+                        else {
+                                skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+                                struct page *page = frag->page;
+                                void *p = kmap(page) + frag->page_offset + o;
+                                err = memcpy_toiovec(to, p, copy);
+                                kunmap(page);
+                        }
                         if (err)
                                 goto fault;
                         if (!(len -= copy))
                                 return 0;
                         offset += copy;
                 }
-                start = end;
+                if (++i >= skb_shinfo(skb)->nr_frags)
+                        break;
+                fraglen = skb_shinfo(skb)->frags[i].size;
         }
-
-        if (skb_shinfo(skb)->frag_list) {
-                struct sk_buff *list = skb_shinfo(skb)->frag_list;
-
-                for (; list; list = list->next) {
-                        int end;
-
-                        BUG_TRAP(start <= offset + len);
-
-                        end = start + list->len;
-                        if ((copy = end - offset) > 0) {
-                                if (copy > len)
-                                        copy = len;
-                                if (skb_copy_datagram_iovec(list,
-                                                            offset - start,
-                                                            to, copy))
-                                        goto fault;
-                                if ((len -= copy) == 0)
-                                        return 0;
-                                offset += copy;
-                        }
-                        start = end;
-                }
+        if (next) {
+                skb = next;
+                BUG_ON(skb_shinfo(skb)->frag_list);
+                next = skb->next;
+                goto next_skb;
         }
-        if (!len)
-                return 0;
-
 fault:
         return -EFAULT;
 }
diff --git a/net/core/dev.c b/net/core/dev.c
index c01511e3d0c1..a44eeef24edf 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -574,6 +574,8 @@ struct net_device *dev_getbyhwaddr(unsigned short type, char *ha)
         return dev;
 }
 
+EXPORT_SYMBOL(dev_getbyhwaddr);
+
 struct net_device *dev_getfirstbyhwtype(unsigned short type)
 {
         struct net_device *dev;
@@ -1130,7 +1132,7 @@ static inline int illegal_highdma(struct net_device *dev, struct sk_buff *skb)
 #endif
 
 /* Keep head the same: replace data */
-int __skb_linearize(struct sk_buff *skb, unsigned int __nocast gfp_mask)
+int __skb_linearize(struct sk_buff *skb, gfp_t gfp_mask)
 {
         unsigned int size;
         u8 *data;
@@ -1257,6 +1259,8 @@ int dev_queue_xmit(struct sk_buff *skb)
                 if (skb_checksum_help(skb, 0))
                         goto out_kfree_skb;
 
+        spin_lock_prefetch(&dev->queue_lock);
+
         /* Disable soft irqs for various locks below. Also 
          * stops preemption for RCU. 
          */
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 39fc55edf691..4128fc76ac3a 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -61,7 +61,9 @@ static int pneigh_ifdown(struct neigh_table *tbl, struct net_device *dev);
 void neigh_changeaddr(struct neigh_table *tbl, struct net_device *dev);
 
 static struct neigh_table *neigh_tables;
+#ifdef CONFIG_PROC_FS
 static struct file_operations neigh_stat_seq_fops;
+#endif
 
 /*
    Neighbour hash table buckets are protected with rwlock tbl->lock.
@@ -725,6 +727,13 @@ static __inline__ int neigh_max_probes(struct neighbour *n)
                 p->ucast_probes + p->app_probes + p->mcast_probes);
 }
 
+static inline void neigh_add_timer(struct neighbour *n, unsigned long when)
+{
+        if (unlikely(mod_timer(&n->timer, when))) {
+                printk("NEIGH: BUG, double timer add, state is %x\n",
+                       n->nud_state);
+        }
+}
 
 /* Called when a timer expires for a neighbour entry. */
 
@@ -809,8 +818,7 @@ static void neigh_timer_handler(unsigned long arg)
                 neigh_hold(neigh);
                 if (time_before(next, jiffies + HZ/2))
                         next = jiffies + HZ/2;
-                neigh->timer.expires = next;
-                add_timer(&neigh->timer);
+                neigh_add_timer(neigh, next);
         }
         if (neigh->nud_state & (NUD_INCOMPLETE | NUD_PROBE)) {
                 struct sk_buff *skb = skb_peek(&neigh->arp_queue);
@@ -852,8 +860,7 @@ int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb)
                         atomic_set(&neigh->probes, neigh->parms->ucast_probes);
                         neigh->nud_state     = NUD_INCOMPLETE;
                         neigh_hold(neigh);
-                        neigh->timer.expires = now + 1;
-                        add_timer(&neigh->timer);
+                        neigh_add_timer(neigh, now + 1);
                 } else {
                         neigh->nud_state = NUD_FAILED;
                         write_unlock_bh(&neigh->lock);
@@ -866,8 +873,8 @@ int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb)
                 NEIGH_PRINTK2("neigh %p is delayed.\n", neigh);
                 neigh_hold(neigh);
                 neigh->nud_state = NUD_DELAY;
-                neigh->timer.expires = jiffies + neigh->parms->delay_probe_time;
-                add_timer(&neigh->timer);
+                neigh_add_timer(neigh,
+                                jiffies + neigh->parms->delay_probe_time);
         }
 
         if (neigh->nud_state == NUD_INCOMPLETE) {
@@ -1013,10 +1020,10 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new,
                 neigh_del_timer(neigh);
                 if (new & NUD_IN_TIMER) {
                         neigh_hold(neigh);
-                        neigh->timer.expires = jiffies + 
+                        neigh_add_timer(neigh, (jiffies + 
                                                 ((new & NUD_REACHABLE) ? 
-                                                 neigh->parms->reachable_time : 0);
-                        add_timer(&neigh->timer);
+                                                 neigh->parms->reachable_time :
+                                                 0)));
                 }
                 neigh->nud_state = new;
         }
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 5265dfd69928..802fe11efad0 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -703,7 +703,7 @@ int netpoll_setup(struct netpoll *np)
 
         if (!np->local_ip) {
                 rcu_read_lock();
-                in_dev = __in_dev_get(ndev);
+                in_dev = __in_dev_get_rcu(ndev);
 
                 if (!in_dev || !in_dev->ifa_list) {
                         rcu_read_unlock();
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index ef430b1e8e42..5f043d346694 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -186,7 +186,7 @@
 
 /* Used to help with determining the pkts on receive */
 #define PKTGEN_MAGIC 0xbe9be955
-#define PG_PROC_DIR "pktgen"
+#define PG_PROC_DIR "net/pktgen"
 
 #define MAX_CFLOWS  65536
 
@@ -1476,18 +1476,7 @@ static int proc_thread_write(struct file *file, const char __user *user_buffer,
 
 static int create_proc_dir(void)
 {
-        int     len;
-        /*  does proc_dir already exists */
-        len = strlen(PG_PROC_DIR);
-
-        for (pg_proc_dir = proc_net->subdir; pg_proc_dir; pg_proc_dir=pg_proc_dir->next) {
-                if ((pg_proc_dir->namelen == len) &&
-                    (! memcmp(pg_proc_dir->name, PG_PROC_DIR, len))) 
-                        break;
-        }
-        
-        if (!pg_proc_dir) 
-                pg_proc_dir = create_proc_entry(PG_PROC_DIR, S_IFDIR, proc_net);
+        pg_proc_dir = proc_mkdir(PG_PROC_DIR, NULL);
         
         if (!pg_proc_dir) 
                 return -ENODEV;
@@ -1497,7 +1486,7 @@ static int create_proc_dir(void)
 
 static int remove_proc_dir(void)
 {
-        remove_proc_entry(PG_PROC_DIR, proc_net);
+        remove_proc_entry(PG_PROC_DIR, NULL);
         return 0;
 }
 
@@ -1678,13 +1667,12 @@ static void pktgen_setup_inject(struct pktgen_dev *pkt_dev)
                         struct in_device *in_dev; 
 
                         rcu_read_lock();
-                        in_dev = __in_dev_get(pkt_dev->odev);
+                        in_dev = __in_dev_get_rcu(pkt_dev->odev);
                         if (in_dev) {
                                 if (in_dev->ifa_list) {
                                         pkt_dev->saddr_min = in_dev->ifa_list->ifa_address;
                                         pkt_dev->saddr_max = pkt_dev->saddr_min;
                                 }
-                                __in_dev_put(in_dev);   
                         }
                         rcu_read_unlock();
                 }
@@ -2908,7 +2896,7 @@ static int pktgen_add_device(struct pktgen_thread *t, const char* ifname)
                 pkt_dev->udp_dst_max = 9;
 
                 strncpy(pkt_dev->ifname, ifname, 31);
-                sprintf(pkt_dev->fname, "net/%s/%s", PG_PROC_DIR, ifname);
+                sprintf(pkt_dev->fname, "%s/%s", PG_PROC_DIR, ifname);
 
                 if (! pktgen_setup_dev(pkt_dev)) {
                         printk("pktgen: ERROR: pktgen_setup_dev failed.\n");
@@ -2981,7 +2969,7 @@ static int pktgen_create_thread(const char* name, int cpu)
         spin_lock_init(&t->if_lock);
         t->cpu = cpu;
         
-        sprintf(t->fname, "net/%s/%s", PG_PROC_DIR, t->name);
+        sprintf(t->fname, "%s/%s", PG_PROC_DIR, t->name);
         t->proc_ent = create_proc_entry(t->fname, 0600, NULL);
         if (!t->proc_ent) {
                 printk("pktgen: cannot create %s procfs entry.\n", t->fname);
@@ -3064,7 +3052,7 @@ static int __init pg_init(void)
 
         create_proc_dir();
 
-        sprintf(module_fname, "net/%s/pgctrl", PG_PROC_DIR);
+        sprintf(module_fname, "%s/pgctrl", PG_PROC_DIR);
         module_proc_ent = create_proc_entry(module_fname, 0600, NULL);
         if (!module_proc_ent) {
                 printk("pktgen: ERROR: cannot create %s procfs entry.\n", module_fname);
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index f80a28785610..af9b1516e21f 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -71,8 +71,6 @@
 static kmem_cache_t *skbuff_head_cache __read_mostly;
 static kmem_cache_t *skbuff_fclone_cache __read_mostly;
 
-struct timeval __read_mostly skb_tv_base;
-
 /*
  *      Keep out-of-line to prevent kernel bloat.
  *      __builtin_return_address is not used because it is not always
@@ -132,7 +130,7 @@ void skb_under_panic(struct sk_buff *skb, int sz, void *here)
  *      Buffers may only be allocated from interrupts using a @gfp_mask of
  *      %GFP_ATOMIC.
  */
-struct sk_buff *__alloc_skb(unsigned int size, unsigned int __nocast gfp_mask,
+struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
                             int fclone)
 {
         struct sk_buff *skb;
@@ -200,7 +198,7 @@ nodata:
  */
 struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp,
                                      unsigned int size,
-                                     unsigned int __nocast gfp_mask)
+                                     gfp_t gfp_mask)
 {
         struct sk_buff *skb;
         u8 *data;
@@ -363,7 +361,7 @@ void __kfree_skb(struct sk_buff *skb)
  *      %GFP_ATOMIC.
  */
 
-struct sk_buff *skb_clone(struct sk_buff *skb, unsigned int __nocast gfp_mask)
+struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask)
 {
         struct sk_buff *n;
 
@@ -502,7 +500,7 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
  *      header is going to be modified. Use pskb_copy() instead.
  */
 
-struct sk_buff *skb_copy(const struct sk_buff *skb, unsigned int __nocast gfp_mask)
+struct sk_buff *skb_copy(const struct sk_buff *skb, gfp_t gfp_mask)
 {
         int headerlen = skb->data - skb->head;
         /*
@@ -541,7 +539,7 @@ struct sk_buff *skb_copy(const struct sk_buff *skb, unsigned int __nocast gfp_ma
  *      The returned buffer has a reference count of 1.
  */
 
-struct sk_buff *pskb_copy(struct sk_buff *skb, unsigned int __nocast gfp_mask)
+struct sk_buff *pskb_copy(struct sk_buff *skb, gfp_t gfp_mask)
 {
         /*
          *      Allocate the copy buffer
@@ -600,7 +598,7 @@ out:
  */
 
 int pskb_expand_head(struct sk_buff *skb, int nhead, int ntail,
-                     unsigned int __nocast gfp_mask)
+                     gfp_t gfp_mask)
 {
         int i;
         u8 *data;
@@ -691,7 +689,7 @@ struct sk_buff *skb_realloc_headroom(struct sk_buff *skb, unsigned int headroom)
  */
 struct sk_buff *skb_copy_expand(const struct sk_buff *skb,
                                 int newheadroom, int newtailroom,
-                                unsigned int __nocast gfp_mask)
+                                gfp_t gfp_mask)
 {
         /*
          *      Allocate the copy buffer
@@ -1708,8 +1706,6 @@ void __init skb_init(void)
                                                 NULL, NULL);
         if (!skbuff_fclone_cache)
                 panic("cannot create skbuff cache");
-
-        do_gettimeofday(&skb_tv_base);
 }
 
 EXPORT_SYMBOL(___pskb_trim);
@@ -1743,4 +1739,3 @@ EXPORT_SYMBOL(skb_prepare_seq_read);
 EXPORT_SYMBOL(skb_seq_read);
 EXPORT_SYMBOL(skb_abort_seq_read);
 EXPORT_SYMBOL(skb_find_text);
-EXPORT_SYMBOL(skb_tv_base);
diff --git a/net/core/sock.c b/net/core/sock.c
index ac63b56e23b2..1c52fe809eda 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -637,7 +637,7 @@ lenout:
  *      @prot: struct proto associated with this new sock instance
  *      @zero_it: if we should zero the newly allocated sock
  */
-struct sock *sk_alloc(int family, unsigned int __nocast priority,
+struct sock *sk_alloc(int family, gfp_t priority,
                       struct proto *prot, int zero_it)
 {
         struct sock *sk = NULL;
@@ -660,16 +660,20 @@ struct sock *sk_alloc(int family, unsigned int __nocast priority,
                         sock_lock_init(sk);
                 }
                 
-                if (security_sk_alloc(sk, family, priority)) {
-                        if (slab != NULL)
-                                kmem_cache_free(slab, sk);
-                        else
-                                kfree(sk);
-                        sk = NULL;
-                } else
-                        __module_get(prot->owner);
+                if (security_sk_alloc(sk, family, priority))
+                        goto out_free;
+
+                if (!try_module_get(prot->owner))
+                        goto out_free;
         }
         return sk;
+
+out_free:
+        if (slab != NULL)
+                kmem_cache_free(slab, sk);
+        else
+                kfree(sk);
+        return NULL;
 }
 
 void sk_free(struct sock *sk)
@@ -700,7 +704,7 @@ void sk_free(struct sock *sk)
         module_put(owner);
 }
 
-struct sock *sk_clone(const struct sock *sk, const unsigned int __nocast priority)
+struct sock *sk_clone(const struct sock *sk, const gfp_t priority)
 {
         struct sock *newsk = sk_alloc(sk->sk_family, priority, sk->sk_prot, 0);
 
@@ -841,7 +845,7 @@ unsigned long sock_i_ino(struct sock *sk)
  * Allocate a skb from the socket's send buffer.
  */
 struct sk_buff *sock_wmalloc(struct sock *sk, unsigned long size, int force,
-                             unsigned int __nocast priority)
+                             gfp_t priority)
 {
         if (force || atomic_read(&sk->sk_wmem_alloc) < sk->sk_sndbuf) {
                 struct sk_buff * skb = alloc_skb(size, priority);
@@ -857,7 +861,7 @@ struct sk_buff *sock_wmalloc(struct sock *sk, unsigned long size, int force,
  * Allocate a skb from the socket's receive buffer.
  */ 
 struct sk_buff *sock_rmalloc(struct sock *sk, unsigned long size, int force,
-                             unsigned int __nocast priority)
+                             gfp_t priority)
 {
         if (force || atomic_read(&sk->sk_rmem_alloc) < sk->sk_rcvbuf) {
                 struct sk_buff *skb = alloc_skb(size, priority);
@@ -872,7 +876,7 @@ struct sk_buff *sock_rmalloc(struct sock *sk, unsigned long size, int force,
 /* 
  * Allocate a memory block from the socket's option memory buffer.
  */ 
-void *sock_kmalloc(struct sock *sk, int size, unsigned int __nocast priority)
+void *sock_kmalloc(struct sock *sk, int size, gfp_t priority)
 {
         if ((unsigned)size <= sysctl_optmem_max &&
             atomic_read(&sk->sk_omem_alloc) + size < sysctl_optmem_max) {