Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6

Documentation/networking/fib_trie.txt

+			LC-trie implementation notes.
+
+Node types
+----------
+leaf 
+	An end node with data. This has a copy of the relevant key, along
+	with 'hlist' with routing table entries sorted by prefix length.
+	See struct leaf and struct leaf_info.
+
+trie node or tnode
+	An internal node, holding an array of child (leaf or tnode) pointers,
+	indexed	through a subset of the key. See Level Compression.
+
+A few concepts explained
+------------------------
+Bits (tnode) 
+	The number of bits in the key segment used for indexing into the
+	child array - the "child index". See Level Compression.
+
+Pos (tnode)
+	The position (in the key) of the key segment used for indexing into
+	the child array. See Path Compression.
+
+Path Compression / skipped bits
+	Any given tnode is linked to from the child array of its parent, using
+	a segment of the key specified by the parent's "pos" and "bits" 
+	In certain cases, this tnode's own "pos" will not be immediately
+	adjacent to the parent (pos+bits), but there will be some bits
+	in the key skipped over because they represent a single path with no
+	deviations. These "skipped bits" constitute Path Compression.
+	Note that the search algorithm will simply skip over these bits when
+	searching, making it necessary to save the keys in the leaves to
+	verify that they actually do match the key we are searching for.
+
+Level Compression / child arrays
+	the trie is kept level balanced moving, under certain conditions, the
+	children of a full child (see "full_children") up one level, so that
+	instead of a pure binary tree, each internal node ("tnode") may
+	contain an arbitrarily large array of links to several children.
+	Conversely, a tnode with a mostly empty	child array (see empty_children)
+	may be "halved", having some of its children moved downwards one level,
+	in order to avoid ever-increasing child arrays.
+
+empty_children
+	the number of positions in the child array of a given tnode that are
+	NULL.
+
+full_children
+	the number of children of a given tnode that aren't path compressed.
+	(in other words, they aren't NULL or leaves and their "pos" is equal
+	to this	tnode's "pos"+"bits").
+
+	(The word "full" here is used more in the sense of "complete" than
+	as the opposite of "empty", which might be a tad confusing.)
+
+Comments
+---------
+
+We have tried to keep the structure of the code as close to fib_hash as 
+possible to allow verification and help up reviewing. 
+
+fib_find_node()
+	A good start for understanding this code. This function implements a
+	straightforward trie lookup.
+
+fib_insert_node()
+	Inserts a new leaf node in the trie. This is bit more complicated than
+	fib_find_node(). Inserting a new node means we might have to run the
+	level compression algorithm on part of the trie.
+
+trie_leaf_remove()
+	Looks up a key, deletes it and runs the level compression algorithm.
+
+trie_rebalance()
+	The key function for the dynamic trie after any change in the trie
+	it is run to optimize and reorganize. Tt will walk the trie upwards 
+	towards the root from a given tnode, doing a resize() at each step 
+	to implement level compression.
+
+resize()
+	Analyzes a tnode and optimizes the child array size by either inflating
+	or shrinking it repeatedly until it fullfills the criteria for optimal
+	level compression. This part follows the original paper pretty closely
+	and there may be some room for experimentation here.
+
+inflate()
+	Doubles the size of the child array within a tnode. Used by resize().
+
+halve()
+	Halves the size of the child array within a tnode - the inverse of
+	inflate(). Used by resize();
+
+fn_trie_insert(), fn_trie_delete(), fn_trie_select_default()
+	The route manipulation functions. Should conform pretty closely to the
+	corresponding functions in fib_hash.
+
+fn_trie_flush()
+	This walks the full trie (using nextleaf()) and searches for empty
+	leaves which have to be removed.
+
+fn_trie_dump()
+	Dumps the routing table ordered by prefix length. This is somewhat
+	slower than the corresponding fib_hash function, as we have to walk the
+	entire trie for each prefix length. In comparison, fib_hash is organized
+	as one "zone"/hash per prefix length.
+
+Locking
+-------
+
+fib_lock is used for an RW-lock in the same way that this is done in fib_hash.
+However, the functions are somewhat separated for other possible locking
+scenarios. It might conceivably be possible to run trie_rebalance via RCU
+to avoid read_lock in the fn_trie_lookup() function.
+
+Main lookup mechanism
+---------------------
+fn_trie_lookup() is the main lookup function.
+
+The lookup is in its simplest form just like fib_find_node(). We descend the
+trie, key segment by key segment, until we find a leaf. check_leaf() does
+the fib_semantic_match in the leaf's sorted prefix hlist.
+
+If we find a match, we are done.
+
+If we don't find a match, we enter prefix matching mode. The prefix length,
+starting out at the same as the key length, is reduced one step at a time,
+and we backtrack upwards through the trie trying to find a longest matching
+prefix. The goal is always to reach a leaf and get a positive result from the
+fib_semantic_match mechanism.
+
+Inside each tnode, the search for longest matching prefix consists of searching
+through the child array, chopping off (zeroing) the least significant "1" of
+the child index until we find a match or the child index consists of nothing but
+zeros.
+
+At this point we backtrack (t->stats.backtrack++) up the trie, continuing to
+chop off part of the key in order to find the longest matching prefix.
+
+At this point we will repeatedly descend subtries to look for a match, and there
+are some optimizations available that can provide us with "shortcuts" to avoid
+descending into dead ends. Look for "HL_OPTIMIZE" sections in the code.
+
+To alleviate any doubts about the correctness of the route selection process,
+a new netlink operation has been added. Look for NETLINK_FIB_LOOKUP, which
+gives userland access to fib_lookup().

drivers/net/shaper.c

@@ -135,10 +135,8 @@ static int shaper_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct shaper *shaper = dev->priv;
  	struct sk_buff *ptr;
-   
-	if (down_trylock(&shaper->sem))
-		return -1;
-
+  
+	spin_lock(&shaper->lock);
  	ptr=shaper->sendq.prev;
  	
  	/*
@@ -232,7 +230,7 @@ static int shaper_start_xmit(struct sk_buff *skb, struct net_device *dev)
                 shaper->stats.collisions++;
  	}
 	shaper_kick(shaper);
-	up(&shaper->sem);
+	spin_unlock(&shaper->lock);
  	return 0;
 }
 
@@ -271,11 +269,9 @@ static void shaper_timer(unsigned long data)
 {
 	struct shaper *shaper = (struct shaper *)data;
 
-	if (!down_trylock(&shaper->sem)) {
-		shaper_kick(shaper);
-		up(&shaper->sem);
-	} else
-		mod_timer(&shaper->timer, jiffies);
+	spin_lock(&shaper->lock);
+	shaper_kick(shaper);
+	spin_unlock(&shaper->lock);
 }
 
 /*
@@ -331,21 +327,6 @@ static void shaper_kick(struct shaper *shaper)
 }
 
 
-/*
- *	Flush the shaper queues on a closedown
- */
- 
-static void shaper_flush(struct shaper *shaper)
-{
-	struct sk_buff *skb;
-
-	down(&shaper->sem);
-	while((skb=skb_dequeue(&shaper->sendq))!=NULL)
-		dev_kfree_skb(skb);
-	shaper_kick(shaper);
-	up(&shaper->sem);
-}
-
 /*
  *	Bring the interface up. We just disallow this until a 
  *	bind.
@@ -375,7 +356,15 @@ static int shaper_open(struct net_device *dev)
 static int shaper_close(struct net_device *dev)
 {
 	struct shaper *shaper=dev->priv;
-	shaper_flush(shaper);
+	struct sk_buff *skb;
+
+	while ((skb = skb_dequeue(&shaper->sendq)) != NULL)
+		dev_kfree_skb(skb);
+
+	spin_lock_bh(&shaper->lock);
+	shaper_kick(shaper);
+	spin_unlock_bh(&shaper->lock);
+
 	del_timer_sync(&shaper->timer);
 	return 0;
 }
@@ -576,6 +565,7 @@ static void shaper_init_priv(struct net_device *dev)
 	init_timer(&sh->timer);
 	sh->timer.function=shaper_timer;
 	sh->timer.data=(unsigned long)sh;
+	spin_lock_init(&sh->lock);
 }
 
 /*

drivers/net/skge.h

@@ -7,6 +7,7 @@
 /* PCI config registers */
 #define PCI_DEV_REG1	0x40
 #define PCI_DEV_REG2	0x44
+#define  PCI_REV_DESC	 0x4
 
 #define PCI_STATUS_ERROR_BITS (PCI_STATUS_DETECTED_PARITY | \
 			       PCI_STATUS_SIG_SYSTEM_ERROR | \

drivers/net/tg3.c

@@ -66,8 +66,8 @@
 
 #define DRV_MODULE_NAME		"tg3"
 #define PFX DRV_MODULE_NAME	": "
-#define DRV_MODULE_VERSION	"3.32"
-#define DRV_MODULE_RELDATE	"June 24, 2005"
+#define DRV_MODULE_VERSION	"3.33"
+#define DRV_MODULE_RELDATE	"July 5, 2005"
 
 #define TG3_DEF_MAC_MODE	0
 #define TG3_DEF_RX_MODE		0
@@ -5117,7 +5117,7 @@ static void tg3_set_bdinfo(struct tg3 *tp, u32 bdinfo_addr,
 }
 
 static void __tg3_set_rx_mode(struct net_device *);
-static void tg3_set_coalesce(struct tg3 *tp, struct ethtool_coalesce *ec)
+static void __tg3_set_coalesce(struct tg3 *tp, struct ethtool_coalesce *ec)
 {
 	tw32(HOSTCC_RXCOL_TICKS, ec->rx_coalesce_usecs);
 	tw32(HOSTCC_TXCOL_TICKS, ec->tx_coalesce_usecs);
@@ -5460,7 +5460,7 @@ static int tg3_reset_hw(struct tg3 *tp)
 		udelay(10);
 	}
 
-	tg3_set_coalesce(tp, &tp->coal);
+	__tg3_set_coalesce(tp, &tp->coal);
 
 	/* set status block DMA address */
 	tw32(HOSTCC_STATUS_BLK_HOST_ADDR + TG3_64BIT_REG_HIGH,
@@ -7821,6 +7821,60 @@ static int tg3_get_coalesce(struct net_device *dev, struct ethtool_coalesce *ec)
 	return 0;
 }
 
+static int tg3_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec)
+{
+	struct tg3 *tp = netdev_priv(dev);
+	u32 max_rxcoal_tick_int = 0, max_txcoal_tick_int = 0;
+	u32 max_stat_coal_ticks = 0, min_stat_coal_ticks = 0;
+
+	if (!(tp->tg3_flags2 & TG3_FLG2_5705_PLUS)) {
+		max_rxcoal_tick_int = MAX_RXCOAL_TICK_INT;
+		max_txcoal_tick_int = MAX_TXCOAL_TICK_INT;
+		max_stat_coal_ticks = MAX_STAT_COAL_TICKS;
+		min_stat_coal_ticks = MIN_STAT_COAL_TICKS;
+	}
+
+	if ((ec->rx_coalesce_usecs > MAX_RXCOL_TICKS) ||
+	    (ec->tx_coalesce_usecs > MAX_TXCOL_TICKS) ||
+	    (ec->rx_max_coalesced_frames > MAX_RXMAX_FRAMES) ||
+	    (ec->tx_max_coalesced_frames > MAX_TXMAX_FRAMES) ||
+	    (ec->rx_coalesce_usecs_irq > max_rxcoal_tick_int) ||
+	    (ec->tx_coalesce_usecs_irq > max_txcoal_tick_int) ||
+	    (ec->rx_max_coalesced_frames_irq > MAX_RXCOAL_MAXF_INT) ||
+	    (ec->tx_max_coalesced_frames_irq > MAX_TXCOAL_MAXF_INT) ||
+	    (ec->stats_block_coalesce_usecs > max_stat_coal_ticks) ||
+	    (ec->stats_block_coalesce_usecs < min_stat_coal_ticks))
+		return -EINVAL;
+
+	/* No rx interrupts will be generated if both are zero */
+	if ((ec->rx_coalesce_usecs == 0) &&
+	    (ec->rx_max_coalesced_frames == 0))
+		return -EINVAL;
+
+	/* No tx interrupts will be generated if both are zero */
+	if ((ec->tx_coalesce_usecs == 0) &&
+	    (ec->tx_max_coalesced_frames == 0))
+		return -EINVAL;
+
+	/* Only copy relevant parameters, ignore all others. */
+	tp->coal.rx_coalesce_usecs = ec->rx_coalesce_usecs;
+	tp->coal.tx_coalesce_usecs = ec->tx_coalesce_usecs;
+	tp->coal.rx_max_coalesced_frames = ec->rx_max_coalesced_frames;
+	tp->coal.tx_max_coalesced_frames = ec->tx_max_coalesced_frames;
+	tp->coal.rx_coalesce_usecs_irq = ec->rx_coalesce_usecs_irq;
+	tp->coal.tx_coalesce_usecs_irq = ec->tx_coalesce_usecs_irq;
+	tp->coal.rx_max_coalesced_frames_irq = ec->rx_max_coalesced_frames_irq;
+	tp->coal.tx_max_coalesced_frames_irq = ec->tx_max_coalesced_frames_irq;
+	tp->coal.stats_block_coalesce_usecs = ec->stats_block_coalesce_usecs;
+
+	if (netif_running(dev)) {
+		tg3_full_lock(tp, 0);
+		__tg3_set_coalesce(tp, &tp->coal);
+		tg3_full_unlock(tp);
+	}
+	return 0;
+}
+
 static struct ethtool_ops tg3_ethtool_ops = {
 	.get_settings		= tg3_get_settings,
 	.set_settings		= tg3_set_settings,
@@ -7856,6 +7910,7 @@ static struct ethtool_ops tg3_ethtool_ops = {
 	.get_stats_count	= tg3_get_stats_count,
 	.get_ethtool_stats	= tg3_get_ethtool_stats,
 	.get_coalesce		= tg3_get_coalesce,
+	.set_coalesce		= tg3_set_coalesce,
 };
 
 static void __devinit tg3_get_eeprom_size(struct tg3 *tp)
@@ -9800,6 +9855,12 @@ static void __devinit tg3_init_coal(struct tg3 *tp)
 		ec->tx_coalesce_usecs = LOW_TXCOL_TICKS_CLRTCKS;
 		ec->tx_coalesce_usecs_irq = DEFAULT_TXCOAL_TICK_INT_CLRTCKS;
 	}
+
+	if (tp->tg3_flags2 & TG3_FLG2_5705_PLUS) {
+		ec->rx_coalesce_usecs_irq = 0;
+		ec->tx_coalesce_usecs_irq = 0;
+		ec->stats_block_coalesce_usecs = 0;
+	}
 }
 
 static int __devinit tg3_init_one(struct pci_dev *pdev,

drivers/net/tg3.h

@@ -879,31 +879,41 @@
 #define  LOW_RXCOL_TICKS_CLRTCKS	 0x00000014
 #define  DEFAULT_RXCOL_TICKS		 0x00000048
 #define  HIGH_RXCOL_TICKS		 0x00000096
+#define  MAX_RXCOL_TICKS		 0x000003ff
 #define HOSTCC_TXCOL_TICKS		0x00003c0c
 #define  LOW_TXCOL_TICKS		 0x00000096
 #define  LOW_TXCOL_TICKS_CLRTCKS	 0x00000048
 #define  DEFAULT_TXCOL_TICKS		 0x0000012c
 #define  HIGH_TXCOL_TICKS		 0x00000145
+#define  MAX_TXCOL_TICKS		 0x000003ff
 #define HOSTCC_RXMAX_FRAMES		0x00003c10
 #define  LOW_RXMAX_FRAMES		 0x00000005
 #define  DEFAULT_RXMAX_FRAMES		 0x00000008
 #define  HIGH_RXMAX_FRAMES		 0x00000012
+#define  MAX_RXMAX_FRAMES		 0x000000ff
 #define HOSTCC_TXMAX_FRAMES		0x00003c14
 #define  LOW_TXMAX_FRAMES		 0x00000035
 #define  DEFAULT_TXMAX_FRAMES		 0x0000004b
 #define  HIGH_TXMAX_FRAMES		 0x00000052
+#define  MAX_TXMAX_FRAMES		 0x000000ff
 #define HOSTCC_RXCOAL_TICK_INT		0x00003c18
 #define  DEFAULT_RXCOAL_TICK_INT	 0x00000019
 #define  DEFAULT_RXCOAL_TICK_INT_CLRTCKS 0x00000014
+#define  MAX_RXCOAL_TICK_INT		 0x000003ff
 #define HOSTCC_TXCOAL_TICK_INT		0x00003c1c
 #define  DEFAULT_TXCOAL_TICK_INT	 0x00000019
 #define  DEFAULT_TXCOAL_TICK_INT_CLRTCKS 0x00000014
+#define  MAX_TXCOAL_TICK_INT		 0x000003ff
 #define HOSTCC_RXCOAL_MAXF_INT		0x00003c20
 #define  DEFAULT_RXCOAL_MAXF_INT	 0x00000005
+#define  MAX_RXCOAL_MAXF_INT		 0x000000ff
 #define HOSTCC_TXCOAL_MAXF_INT		0x00003c24
 #define  DEFAULT_TXCOAL_MAXF_INT	 0x00000005
+#define  MAX_TXCOAL_MAXF_INT		 0x000000ff
 #define HOSTCC_STAT_COAL_TICKS		0x00003c28
 #define  DEFAULT_STAT_COAL_TICKS	 0x000f4240
+#define  MAX_STAT_COAL_TICKS		 0xd693d400
+#define  MIN_STAT_COAL_TICKS		 0x00000064
 /* 0x3c2c --> 0x3c30 unused */
 #define HOSTCC_STATS_BLK_HOST_ADDR	0x00003c30 /* 64-bit */
 #define HOSTCC_STATUS_BLK_HOST_ADDR	0x00003c38 /* 64-bit */

include/linux/if_shaper.h

@@ -23,7 +23,7 @@ struct shaper
 	__u32 shapeclock;
 	unsigned long recovery;	/* Time we can next clock a packet out on
 				   an empty queue */
-	struct semaphore sem;
+	spinlock_t lock;
         struct net_device_stats stats;
 	struct net_device *dev;
 	int  (*hard_start_xmit) (struct sk_buff *skb,

include/linux/skbuff.h

@@ -183,7 +183,6 @@ struct skb_shared_info {
  *	@priority: Packet queueing priority
  *	@users: User count - see {datagram,tcp}.c
  *	@protocol: Packet protocol from driver
- *	@security: Security level of packet
  *	@truesize: Buffer size 
  *	@head: Head of buffer
  *	@data: Data head pointer
@@ -249,18 +248,18 @@ struct sk_buff {
 				data_len,
 				mac_len,
 				csum;
-	unsigned char		local_df,
-				cloned:1,
-				nohdr:1,
-				pkt_type,
-				ip_summed;
 	__u32			priority;
-	unsigned short		protocol,
-				security;
+	__u8			local_df:1,
+				cloned:1,
+				ip_summed:2,
+				nohdr:1;
+				/* 3 bits spare */
+	__u8			pkt_type;
+	__u16			protocol;
 
 	void			(*destructor)(struct sk_buff *skb);
 #ifdef CONFIG_NETFILTER
-        unsigned long		nfmark;
+	unsigned long		nfmark;
 	__u32			nfcache;
 	__u32			nfctinfo;
 	struct nf_conntrack	*nfct;
@@ -1211,7 +1210,7 @@ static inline void *skb_header_pointer(const struct sk_buff *skb, int offset,
 {
 	int hlen = skb_headlen(skb);
 
-	if (offset + len <= hlen)
+	if (hlen - offset >= len)
 		return skb->data + offset;
 
 	if (skb_copy_bits(skb, offset, buffer, len) < 0)

include/linux/tc_ematch/tc_em_meta.h

@@ -45,7 +45,7 @@ enum
 	TCF_META_ID_REALDEV,
 	TCF_META_ID_PRIORITY,
 	TCF_META_ID_PROTOCOL,
-	TCF_META_ID_SECURITY,
+	TCF_META_ID_SECURITY, /* obsolete */
 	TCF_META_ID_PKTTYPE,
 	TCF_META_ID_PKTLEN,
 	TCF_META_ID_DATALEN,

include/linux/tcp.h

@@ -286,7 +286,7 @@ struct tcp_sock {
 	__u32	max_window;	/* Maximal window ever seen from peer	*/
 	__u32	pmtu_cookie;	/* Last pmtu seen by socket		*/
 	__u32	mss_cache;	/* Cached effective mss, not including SACKS */
-	__u16	mss_cache_std;	/* Like mss_cache, but without TSO */
+	__u16	xmit_size_goal;	/* Goal for segmenting output packets	*/
 	__u16	ext_header_len;	/* Network protocol overhead (IP/IPv6 options) */
 	__u8	ca_state;	/* State of fast-retransmit machine 	*/
 	__u8	retransmits;	/* Number of unrecovered RTO timeouts.	*/

include/net/pkt_sched.h

@@ -13,13 +13,12 @@ struct qdisc_walker
 
 extern rwlock_t qdisc_tree_lock;
 
-#define	QDISC_ALIGN		32
-#define	QDISC_ALIGN_CONST	(QDISC_ALIGN - 1)
+#define QDISC_ALIGNTO		32
+#define QDISC_ALIGN(len)	(((len) + QDISC_ALIGNTO-1) & ~(QDISC_ALIGNTO-1))
 
 static inline void *qdisc_priv(struct Qdisc *q)
 {
-	return (char *)q + ((sizeof(struct Qdisc) + QDISC_ALIGN_CONST)
-			      & ~QDISC_ALIGN_CONST);
+	return (char *) q + QDISC_ALIGN(sizeof(struct Qdisc));
 }
 
 /* 
@@ -207,8 +206,6 @@ psched_tod_diff(int delta_sec, int bound)
 
 #endif /* !CONFIG_NET_SCH_CLK_GETTIMEOFDAY */
 
-extern struct Qdisc noop_qdisc;
-extern struct Qdisc_ops noop_qdisc_ops;
 extern struct Qdisc_ops pfifo_qdisc_ops;
 extern struct Qdisc_ops bfifo_qdisc_ops;
 
@@ -216,14 +213,6 @@ extern int register_qdisc(struct Qdisc_ops *qops);
 extern int unregister_qdisc(struct Qdisc_ops *qops);
 extern struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle);
 extern struct Qdisc *qdisc_lookup_class(struct net_device *dev, u32 handle);
-extern void dev_init_scheduler(struct net_device *dev);
-extern void dev_shutdown(struct net_device *dev);
-extern void dev_activate(struct net_device *dev);
-extern void dev_deactivate(struct net_device *dev);
-extern void qdisc_reset(struct Qdisc *qdisc);
-extern void qdisc_destroy(struct Qdisc *qdisc);
-extern struct Qdisc * qdisc_create_dflt(struct net_device *dev,
-	struct Qdisc_ops *ops);
 extern struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r,
 		struct rtattr *tab);
 extern void qdisc_put_rtab(struct qdisc_rate_table *tab);

include/net/sch_generic.h

@@ -164,6 +164,19 @@ extern void qdisc_unlock_tree(struct net_device *dev);
 #define tcf_tree_lock(tp)	qdisc_lock_tree((tp)->q->dev)
 #define tcf_tree_unlock(tp)	qdisc_unlock_tree((tp)->q->dev)
 
+extern struct Qdisc noop_qdisc;
+extern struct Qdisc_ops noop_qdisc_ops;
+
+extern void dev_init_scheduler(struct net_device *dev);
+extern void dev_shutdown(struct net_device *dev);
+extern void dev_activate(struct net_device *dev);
+extern void dev_deactivate(struct net_device *dev);
+extern void qdisc_reset(struct Qdisc *qdisc);
+extern void qdisc_destroy(struct Qdisc *qdisc);
+extern struct Qdisc *qdisc_alloc(struct net_device *dev, struct Qdisc_ops *ops);
+extern struct Qdisc *qdisc_create_dflt(struct net_device *dev,
+				       struct Qdisc_ops *ops);
+
 static inline void
 tcf_destroy(struct tcf_proto *tp)
 {

include/net/slhc_vj.h

@@ -170,19 +170,14 @@ struct slcompress {
 };
 #define NULLSLCOMPR	(struct slcompress *)0
 
-#define __ARGS(x) x
-
 /* In slhc.c: */
-struct slcompress *slhc_init __ARGS((int rslots, int tslots));
-void slhc_free __ARGS((struct slcompress *comp));
-
-int slhc_compress __ARGS((struct slcompress *comp, unsigned char *icp,
-			  int isize, unsigned char *ocp, unsigned char **cpp,
-			  int compress_cid));
-int slhc_uncompress __ARGS((struct slcompress *comp, unsigned char *icp,
-			    int isize));
-int slhc_remember __ARGS((struct slcompress *comp, unsigned char *icp,
-			  int isize));
-int slhc_toss __ARGS((struct slcompress *comp));
+struct slcompress *slhc_init(int rslots, int tslots);
+void slhc_free(struct slcompress *comp);
+
+int slhc_compress(struct slcompress *comp, unsigned char *icp, int isize,
+		  unsigned char *ocp, unsigned char **cpp, int compress_cid);
+int slhc_uncompress(struct slcompress *comp, unsigned char *icp, int isize);
+int slhc_remember(struct slcompress *comp, unsigned char *icp, int isize);
+int slhc_toss(struct slcompress *comp);
 
 #endif	/* _SLHC_H */

include/net/sock.h

@@ -1134,13 +1134,16 @@ static inline void sk_stream_moderate_sndbuf(struct sock *sk)
 static inline struct sk_buff *sk_stream_alloc_pskb(struct sock *sk,
 						   int size, int mem, int gfp)
 {
-	struct sk_buff *skb = alloc_skb(size + sk->sk_prot->max_header, gfp);
+	struct sk_buff *skb;
+	int hdr_len;
 
+	hdr_len = SKB_DATA_ALIGN(sk->sk_prot->max_header);
+	skb = alloc_skb(size + hdr_len, gfp);
 	if (skb) {
 		skb->truesize += mem;
 		if (sk->sk_forward_alloc >= (int)skb->truesize ||
 		    sk_stream_mem_schedule(sk, skb->truesize, 0)) {
-			skb_reserve(skb, sk->sk_prot->max_header);
+			skb_reserve(skb, hdr_len);
 			return skb;
 		}
 		__kfree_skb(skb);

include/net/tcp.h

@@ -721,11 +721,16 @@ static inline int tcp_ack_scheduled(struct tcp_sock *tp)
 	return tp->ack.pending&TCP_ACK_SCHED;
 }
 
-static __inline__ void tcp_dec_quickack_mode(struct tcp_sock *tp)
+static __inline__ void tcp_dec_quickack_mode(struct tcp_sock *tp, unsigned int pkts)
 {
-	if (tp->ack.quick && --tp->ack.quick == 0) {
-		/* Leaving quickack mode we deflate ATO. */
-		tp->ack.ato = TCP_ATO_MIN;
+	if (tp->ack.quick) {
+		if (pkts >= tp->ack.quick) {
+			tp->ack.quick = 0;
+
+			/* Leaving quickack mode we deflate ATO. */
+			tp->ack.ato = TCP_ATO_MIN;
+		} else
+			tp->ack.quick -= pkts;
 	}
 }
 
@@ -843,7 +848,9 @@ extern __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb,
 
 /* tcp_output.c */
 
-extern int tcp_write_xmit(struct sock *, int nonagle);
+extern void __tcp_push_pending_frames(struct sock *sk, struct tcp_sock *tp,
+				      unsigned int cur_mss, int nonagle);
+extern int tcp_may_send_now(struct sock *sk, struct tcp_sock *tp);
 extern int tcp_retransmit_skb(struct sock *, struct sk_buff *);
 extern void tcp_xmit_retransmit_queue(struct sock *);
 extern void tcp_simple_retransmit(struct sock *);
@@ -855,10 +862,13 @@ extern int  tcp_write_wakeup(struct sock *);
 extern void tcp_send_fin(struct sock *sk);
 extern void tcp_send_active_reset(struct sock *sk, int priority);
 extern int  tcp_send_synack(struct sock *);
-extern void tcp_push_one(struct sock *, unsigned mss_now);
+extern void tcp_push_one(struct sock *, unsigned int mss_now);
 extern void tcp_send_ack(struct sock *sk);
 extern void tcp_send_delayed_ack(struct sock *sk);
 
+/* tcp_input.c */
+extern void tcp_cwnd_application_limited(struct sock *sk);
+
 /* tcp_timer.c */
 extern void tcp_init_xmit_timers(struct sock *);
 extern void tcp_clear_xmit_timers(struct sock *);
@@ -958,7 +968,7 @@ static inline void tcp_reset_xmit_timer(struct sock *sk, int what, unsigned long
 static inline void tcp_initialize_rcv_mss(struct sock *sk)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
-	unsigned int hint = min(tp->advmss, tp->mss_cache_std);
+	unsigned int hint = min_t(unsigned int, tp->advmss, tp->mss_cache);
 
 	hint = min(hint, tp->rcv_wnd/2);
 	hint = min(hint, TCP_MIN_RCVMSS);
@@ -1225,28 +1235,6 @@ static inline void tcp_sync_left_out(struct tcp_sock *tp)
 	tp->left_out = tp->sacked_out + tp->lost_out;
 }
 
-extern void tcp_cwnd_application_limited(struct sock *sk);
-
-/* Congestion window validation. (RFC2861) */
-
-static inline void tcp_cwnd_validate(struct sock *sk, struct tcp_sock *tp)
-{
-	__u32 packets_out = tp->packets_out;
-
-	if (packets_out >= tp->snd_cwnd) {
-		/* Network is feed fully. */
-		tp->snd_cwnd_used = 0;
-		tp->snd_cwnd_stamp = tcp_time_stamp;
-	} else {
-		/* Network starves. */
-		if (tp->packets_out > tp->snd_cwnd_used)
-			tp->snd_cwnd_used = tp->packets_out;
-
-		if ((s32)(tcp_time_stamp - tp->snd_cwnd_stamp) >= tp->rto)
-			tcp_cwnd_application_limited(sk);
-	}
-}
-
 /* Set slow start threshould and cwnd not falling to slow start */
 static inline void __tcp_enter_cwr(struct tcp_sock *tp)
 {
@@ -1279,12 +1267,6 @@ static __inline__ __u32 tcp_max_burst(const struct tcp_sock *tp)
 	return 3;
 }
 
-static __inline__ int tcp_minshall_check(const struct tcp_sock *tp)
-{
-	return after(tp->snd_sml,tp->snd_una) &&
-		!after(tp->snd_sml, tp->snd_nxt);
-}
-
 static __inline__ void tcp_minshall_update(struct tcp_sock *tp, int mss, 
 					   const struct sk_buff *skb)
 {
@@ -1292,122 +1274,18 @@ static __inline__ void tcp_minshall_update(struct tcp_sock *tp, int mss,
 		tp->snd_sml = TCP_SKB_CB(skb)->end_seq;
 }
 
-/* Return 0, if packet can be sent now without violation Nagle's rules:
-   1. It is full sized.
-   2. Or it contains FIN.
-   3. Or TCP_NODELAY was set.
-   4. Or TCP_CORK is not set, and all sent packets are ACKed.
-      With Minshall's modification: all sent small packets are ACKed.
- */
-
-static __inline__ int
-tcp_nagle_check(const struct tcp_sock *tp, const struct sk_buff *skb, 
-		unsigned mss_now, int nonagle)
-{
-	return (skb->len < mss_now &&
-		!(TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN) &&
-		((nonagle&TCP_NAGLE_CORK) ||
-		 (!nonagle &&
-		  tp->packets_out &&
-		  tcp_minshall_check(tp))));
-}
-
-extern void tcp_set_skb_tso_segs(struct sock *, struct sk_buff *);
-
-/* This checks if the data bearing packet SKB (usually sk->sk_send_head)
- * should be put on the wire right now.
- */
-static __inline__ int tcp_snd_test(struct sock *sk,
-				   struct sk_buff *skb,
-				   unsigned cur_mss, int nonagle)
-{
-	struct tcp_sock *tp = tcp_sk(sk);
-	int pkts = tcp_skb_pcount(skb);
-
-	if (!pkts) {
-		tcp_set_skb_tso_segs(sk, skb);
-		pkts = tcp_skb_pcount(skb);
-	}
-
-	/*	RFC 1122 - section 4.2.3.4
-	 *
-	 *	We must queue if
-	 *
-	 *	a) The right edge of this frame exceeds the window
-	 *	b) There are packets in flight and we have a small segment
-	 *	   [SWS avoidance and Nagle algorithm]
-	 *	   (part of SWS is done on packetization)
-	 *	   Minshall version sounds: there are no _small_
-	 *	   segments in flight. (tcp_nagle_check)
-	 *	c) We have too many packets 'in flight'
-	 *
-	 * 	Don't use the nagle rule for urgent data (or
-	 *	for the final FIN -DaveM).
-	 *
-	 *	Also, Nagle rule does not apply to frames, which
-	 *	sit in the middle of queue (they have no chances
-	 *	to get new data) and if room at tail of skb is
-	 *	not enough to save something seriously (<32 for now).
-	 */
-
-	/* Don't be strict about the congestion window for the
-	 * final FIN frame.  -DaveM
-	 */
-	return (((nonagle&TCP_NAGLE_PUSH) || tp->urg_mode
-		 || !tcp_nagle_check(tp, skb, cur_mss, nonagle)) &&
-		(((tcp_packets_in_flight(tp) + (pkts-1)) < tp->snd_cwnd) ||
-		 (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN)) &&
-		!after(TCP_SKB_CB(skb)->end_seq, tp->snd_una + tp->snd_wnd));
-}
-
 static __inline__ void tcp_check_probe_timer(struct sock *sk, struct tcp_sock *tp)
 {
 	if (!tp->packets_out && !tp->pending)
 		tcp_reset_xmit_timer(sk, TCP_TIME_PROBE0, tp->rto);
 }
 
-static __inline__ int tcp_skb_is_last(const struct sock *sk, 
-				      const struct sk_buff *skb)
-{
-	return skb->next == (struct sk_buff *)&sk->sk_write_queue;
-}
-
-/* Push out any pending frames which were held back due to
- * TCP_CORK or attempt at coalescing tiny packets.
- * The socket must be locked by the caller.
- */
-static __inline__ void __tcp_push_pending_frames(struct sock *sk,
-						 struct tcp_sock *tp,
-						 unsigned cur_mss,
-						 int nonagle)
-{
-	struct sk_buff *skb = sk->sk_send_head;
-
-	if (skb) {
-		if (!tcp_skb_is_last(sk, skb))
-			nonagle = TCP_NAGLE_PUSH;
-		if (!tcp_snd_test(sk, skb, cur_mss, nonagle) ||
-		    tcp_write_xmit(sk, nonagle))
-			tcp_check_probe_timer(sk, tp);
-	}
-	tcp_cwnd_validate(sk, tp);
-}
-
 static __inline__ void tcp_push_pending_frames(struct sock *sk,
 					       struct tcp_sock *tp)
 {
 	__tcp_push_pending_frames(sk, tp, tcp_current_mss(sk, 1), tp->nonagle);
 }
 
-static __inline__ int tcp_may_send_now(struct sock *sk, struct tcp_sock *tp)
-{
-	struct sk_buff *skb = sk->sk_send_head;
-
-	return (skb &&
-		tcp_snd_test(sk, skb, tcp_current_mss(sk, 1),
-			     tcp_skb_is_last(sk, skb) ? TCP_NAGLE_PUSH : tp->nonagle));
-}
-
 static __inline__ void tcp_init_wl(struct tcp_sock *tp, u32 ack, u32 seq)
 {
 	tp->snd_wl1 = seq;

net/core/dev.c

@@ -2089,10 +2089,11 @@ void dev_set_promiscuity(struct net_device *dev, int inc)
 {
 	unsigned short old_flags = dev->flags;
 
-	dev->flags |= IFF_PROMISC;
 	if ((dev->promiscuity += inc) == 0)
 		dev->flags &= ~IFF_PROMISC;
-	if (dev->flags ^ old_flags) {
+	else
+		dev->flags |= IFF_PROMISC;
+	if (dev->flags != old_flags) {
 		dev_mc_upload(dev);
 		printk(KERN_INFO "device %s %s promiscuous mode\n",
 		       dev->name, (dev->flags & IFF_PROMISC) ? "entered" :

net/core/filter.c

@@ -36,7 +36,7 @@
 #include <linux/filter.h>
 
 /* No hurry in this branch */
-static u8 *load_pointer(struct sk_buff *skb, int k)
+static void *__load_pointer(struct sk_buff *skb, int k)
 {
 	u8 *ptr = NULL;
 
@@ -50,6 +50,18 @@ static u8 *load_pointer(struct sk_buff *skb, int k)
 	return NULL;
 }
 
+static inline void *load_pointer(struct sk_buff *skb, int k,
+                                 unsigned int size, void *buffer)
+{
+	if (k >= 0)
+		return skb_header_pointer(skb, k, size, buffer);
+	else {
+		if (k >= SKF_AD_OFF)
+			return NULL;
+		return __load_pointer(skb, k);
+	}
+}
+
 /**
  *	sk_run_filter	- 	run a filter on a socket
  *	@skb: buffer to run the filter on
@@ -64,15 +76,12 @@ static u8 *load_pointer(struct sk_buff *skb, int k)
  
 int sk_run_filter(struct sk_buff *skb, struct sock_filter *filter, int flen)
 {
-	unsigned char *data = skb->data;
-	/* len is UNSIGNED. Byte wide insns relies only on implicit
-	   type casts to prevent reading arbitrary memory locations.
-	 */
-	unsigned int len = skb->len-skb->data_len;
 	struct sock_filter *fentry;	/* We walk down these */
+	void *ptr;
 	u32 A = 0;	   		/* Accumulator */
 	u32 X = 0;   			/* Index Register */
 	u32 mem[BPF_MEMWORDS];		/* Scratch Memory Store */
+	u32 tmp;
 	int k;
 	int pc;
 
@@ -168,86 +177,35 @@ int sk_run_filter(struct sk_buff *skb, struct sock_filter *filter, int flen)
 		case BPF_LD|BPF_W|BPF_ABS:
 			k = fentry->k;
  load_w:
-			if (k >= 0 && (unsigned int)(k+sizeof(u32)) <= len) {
-				A = ntohl(*(u32*)&data[k]);
+			ptr = load_pointer(skb, k, 4, &tmp);
+			if (ptr != NULL) {
+				A = ntohl(*(u32 *)ptr);
 				continue;
 			}
-			if (k < 0) {
-				u8 *ptr;
-
-				if (k >= SKF_AD_OFF)
-					break;
-				ptr = load_pointer(skb, k);
-				if (ptr) {
-					A = ntohl(*(u32*)ptr);
-					continue;
-				}
-			} else {
-				u32 _tmp, *p;
-				p = skb_header_pointer(skb, k, 4, &_tmp);
-				if (p != NULL) {
-					A = ntohl(*p);
-					continue;
-				}
-			}
 			return 0;
 		case BPF_LD|BPF_H|BPF_ABS:
 			k = fentry->k;
  load_h:
-			if (k >= 0 && (unsigned int)(k + sizeof(u16)) <= len) {
-				A = ntohs(*(u16*)&data[k]);
+			ptr = load_pointer(skb, k, 2, &tmp);
+			if (ptr != NULL) {
+				A = ntohs(*(u16 *)ptr);
 				continue;
 			}
-			if (k < 0) {
-				u8 *ptr;
-
-				if (k >= SKF_AD_OFF)
-					break;
-				ptr = load_pointer(skb, k);
-				if (ptr) {
-					A = ntohs(*(u16*)ptr);
-					continue;
-				}
-			} else {
-				u16 _tmp, *p;
-				p = skb_header_pointer(skb, k, 2, &_tmp);
-				if (p != NULL) {
-					A = ntohs(*p);
-					continue;
-				}
-			}
 			return 0;
 		case BPF_LD|BPF_B|BPF_ABS:
 			k = fentry->k;
 load_b:
-			if (k >= 0 && (unsigned int)k < len) {
-				A = data[k];
+			ptr = load_pointer(skb, k, 1, &tmp);
+			if (ptr != NULL) {
+				A = *(u8 *)ptr;
 				continue;
 			}
-			if (k < 0) {
-				u8 *ptr;
-
-				if (k >= SKF_AD_OFF)
-					break;
-				ptr = load_pointer(skb, k);
-				if (ptr) {
-					A = *ptr;
-					continue;
-				}
-			} else {
-				u8 _tmp, *p;
-				p = skb_header_pointer(skb, k, 1, &_tmp);
-				if (p != NULL) {
-					A = *p;
-					continue;
-				}
-			}
 			return 0;
 		case BPF_LD|BPF_W|BPF_LEN:
-			A = len;
+			A = skb->len;
 			continue;
 		case BPF_LDX|BPF_W|BPF_LEN:
-			X = len;
+			X = skb->len;
 			continue;
 		case BPF_LD|BPF_W|BPF_IND:
 			k = X + fentry->k;
@@ -259,10 +217,12 @@ int sk_run_filter(struct sk_buff *skb, struct sock_filter *filter, int flen)
 			k = X + fentry->k;
 			goto load_b;
 		case BPF_LDX|BPF_B|BPF_MSH:
-			if (fentry->k >= len)
-				return 0;
-			X = (data[fentry->k] & 0xf) << 2;
-			continue;
+			ptr = load_pointer(skb, fentry->k, 1, &tmp);
+			if (ptr != NULL) {
+				X = (*(u8 *)ptr & 0xf) << 2;
+				continue;
+			}
+			return 0;
 		case BPF_LD|BPF_IMM:
 			A = fentry->k;
 			continue;

net/core/skbuff.c

@@ -357,7 +357,6 @@ struct sk_buff *skb_clone(struct sk_buff *skb, int gfp_mask)
 	C(ip_summed);
 	C(priority);
 	C(protocol);
-	C(security);
 	n->destructor = NULL;
 #ifdef CONFIG_NETFILTER
 	C(nfmark);
@@ -422,7 +421,6 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
 	new->pkt_type	= old->pkt_type;
 	new->stamp	= old->stamp;
 	new->destructor = NULL;
-	new->security	= old->security;
 #ifdef CONFIG_NETFILTER
 	new->nfmark	= old->nfmark;
 	new->nfcache	= old->nfcache;

net/decnet/dn_fib.c

@@ -551,7 +551,8 @@ int dn_fib_dump(struct sk_buff *skb, struct netlink_callback *cb)
 		if (t < s_t)
 			continue;
 		if (t > s_t)
-			memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(int));
+			memset(&cb->args[1], 0,
+			       sizeof(cb->args) - sizeof(cb->args[0]));
 		tb = dn_fib_get_table(t, 0);
 		if (tb == NULL)
 			continue;

net/ipv4/af_inet.c

@@ -1009,6 +1009,15 @@ static int __init init_ipv4_mibs(void)
 static int ipv4_proc_init(void);
 extern void ipfrag_init(void);
 
+/*
+ *	IP protocol layer initialiser
+ */
+
+static struct packet_type ip_packet_type = {
+	.type = __constant_htons(ETH_P_IP),
+	.func = ip_rcv,
+};
+
 static int __init inet_init(void)
 {
 	struct sk_buff *dummy_skb;
@@ -1102,6 +1111,8 @@ static int __init inet_init(void)
 
 	ipfrag_init();
 
+	dev_add_pack(&ip_packet_type);
+
 	rc = 0;
 out:
 	return rc;

net/ipv4/fib_trie.c

@@ -43,7 +43,7 @@
  *		2 of the License, or (at your option) any later version.
  */
 
-#define VERSION "0.324"
+#define VERSION "0.325"
 
 #include <linux/config.h>
 #include <asm/uaccess.h>
@@ -136,6 +136,7 @@ struct trie_use_stats {
 	unsigned int semantic_match_passed;
 	unsigned int semantic_match_miss;
 	unsigned int null_node_hit;
+	unsigned int resize_node_skipped;
 };
 #endif
 
@@ -164,8 +165,8 @@ static void put_child(struct trie *t, struct tnode *tn, int i, struct node *n);
 static void tnode_put_child_reorg(struct tnode *tn, int i, struct node *n, int wasfull);
 static int tnode_child_length(struct tnode *tn);
 static struct node *resize(struct trie *t, struct tnode *tn);
-static struct tnode *inflate(struct trie *t, struct tnode *tn);
-static struct tnode *halve(struct trie *t, struct tnode *tn);
+static struct tnode *inflate(struct trie *t, struct tnode *tn, int *err);
+static struct tnode *halve(struct trie *t, struct tnode *tn, int *err);
 static void tnode_free(struct tnode *tn);
 static void trie_dump_seq(struct seq_file *seq, struct trie *t);
 extern struct fib_alias *fib_find_alias(struct list_head *fah, u8 tos, u32 prio);
@@ -358,11 +359,32 @@ static inline void free_leaf_info(struct leaf_info *li)
 	kfree(li);
 }
 
+static struct tnode *tnode_alloc(unsigned int size)
+{
+	if (size <= PAGE_SIZE) {
+		return kmalloc(size, GFP_KERNEL);
+	} else {
+		return (struct tnode *)
+		       __get_free_pages(GFP_KERNEL, get_order(size));
+	}
+}
+
+static void __tnode_free(struct tnode *tn)
+{
+	unsigned int size = sizeof(struct tnode) +
+	                    (1<<tn->bits) * sizeof(struct node *);
+
+	if (size <= PAGE_SIZE)
+		kfree(tn);
+	else
+		free_pages((unsigned long)tn, get_order(size));
+}
+
 static struct tnode* tnode_new(t_key key, int pos, int bits)
 {
 	int nchildren = 1<<bits;
 	int sz = sizeof(struct tnode) + nchildren * sizeof(struct node *);
-	struct tnode *tn = kmalloc(sz,  GFP_KERNEL);
+	struct tnode *tn = tnode_alloc(sz);
 
 	if(tn)  {
 		memset(tn, 0, sz);
@@ -390,7 +412,7 @@ static void tnode_free(struct tnode *tn)
 			printk("FL %p \n", tn);
 	}
 	else if(IS_TNODE(tn)) { 
-		kfree(tn);
+		__tnode_free(tn);
 		if(trie_debug > 0 ) 
 			printk("FT %p \n", tn);
 	}
@@ -460,6 +482,7 @@ static void tnode_put_child_reorg(struct tnode *tn, int i, struct node *n, int w
 static struct node *resize(struct trie *t, struct tnode *tn) 
 {
 	int i;
+	int err = 0;
 
  	if (!tn)
 		return NULL;
@@ -556,12 +579,20 @@ static struct node *resize(struct trie *t, struct tnode *tn)
 	 */
 
 	check_tnode(tn);
-
+	
+	err = 0;
 	while ((tn->full_children > 0 &&
 	       50 * (tn->full_children + tnode_child_length(tn) - tn->empty_children) >=
 				inflate_threshold * tnode_child_length(tn))) {
 
-		tn = inflate(t, tn);
+		tn = inflate(t, tn, &err);
+
+		if(err) {
+#ifdef CONFIG_IP_FIB_TRIE_STATS
+			t->stats.resize_node_skipped++;
+#endif
+			break;
+		}
 	}
 
 	check_tnode(tn);
@@ -570,11 +601,22 @@ static struct node *resize(struct trie *t, struct tnode *tn)
 	 * Halve as long as the number of empty children in this
 	 * node is above threshold.
 	 */
+
+	err = 0;
 	while (tn->bits > 1 &&
 	       100 * (tnode_child_length(tn) - tn->empty_children) <
-	       halve_threshold * tnode_child_length(tn))
+	       halve_threshold * tnode_child_length(tn)) {
+
+		tn = halve(t, tn, &err);
+
+		if(err) {
+#ifdef CONFIG_IP_FIB_TRIE_STATS
+			t->stats.resize_node_skipped++;
+#endif
+			break;
+		}
+	}
 
-		tn = halve(t, tn);
   
 	/* Only one child remains */
 
@@ -599,7 +641,7 @@ static struct node *resize(struct trie *t, struct tnode *tn)
 	return (struct node *) tn;
 }
 
-static struct tnode *inflate(struct trie *t, struct tnode *tn)
+static struct tnode *inflate(struct trie *t, struct tnode *tn, int *err)
 {
 	struct tnode *inode;
 	struct tnode *oldtnode = tn;
@@ -611,8 +653,63 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn)
 
 	tn = tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits + 1);
 
-	if (!tn)
-		trie_bug("tnode_new failed");
+	if (!tn) {
+		*err = -ENOMEM;
+		return oldtnode;
+	}
+
+	/*
+	 * Preallocate and store tnodes before the actual work so we 
+	 * don't get into an inconsistent state if memory allocation 
+	 * fails. In case of failure we return the oldnode and  inflate 
+	 * of tnode is ignored.
+	 */
+			
+	for(i = 0; i < olen; i++) {
+		struct tnode *inode = (struct tnode *) tnode_get_child(oldtnode, i);
+
+		if (inode &&
+		    IS_TNODE(inode) &&
+		    inode->pos == oldtnode->pos + oldtnode->bits &&
+		    inode->bits > 1) {
+			struct tnode *left, *right;
+
+			t_key m = TKEY_GET_MASK(inode->pos, 1);
+ 
+			left = tnode_new(inode->key&(~m), inode->pos + 1,
+					 inode->bits - 1);
+
+			if(!left) {
+				*err = -ENOMEM; 
+				break;
+			}
+			
+			right = tnode_new(inode->key|m, inode->pos + 1,
+					  inode->bits - 1);
+
+			if(!right) {
+				*err = -ENOMEM; 
+				break;
+			}
+
+			put_child(t, tn, 2*i, (struct node *) left);
+			put_child(t, tn, 2*i+1, (struct node *) right);
+		}
+	}
+
+	if(*err) {
+		int size = tnode_child_length(tn);
+		int j;
+
+		for(j = 0; j < size; j++) 
+			if( tn->child[j])
+				tnode_free((struct tnode *)tn->child[j]);
+
+		tnode_free(tn);
+		
+		*err = -ENOMEM;
+		return oldtnode;
+	}
 
 	for(i = 0; i < olen; i++) {
 		struct node *node = tnode_get_child(oldtnode, i);
@@ -625,7 +722,7 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn)
 
 		if(IS_LEAF(node) || ((struct tnode *) node)->pos >
 		   tn->pos + tn->bits - 1) {
-			if(tkey_extract_bits(node->key, tn->pos + tn->bits - 1,
+			if(tkey_extract_bits(node->key, oldtnode->pos + oldtnode->bits,
 					     1) == 0)
 				put_child(t, tn, 2*i, node);
 			else
@@ -665,27 +762,22 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn)
 			 * the position (inode->pos)
 			 */
 
-			t_key m = TKEY_GET_MASK(inode->pos, 1);
- 
 			/* Use the old key, but set the new significant 
 			 *   bit to zero. 
 			 */
-			left = tnode_new(inode->key&(~m), inode->pos + 1,
-					 inode->bits - 1);
 
-			if(!left) 
-				trie_bug("tnode_new failed");
-			
-			
-			/* Use the old key, but set the new significant 
-			 * bit to one. 
-			 */
-			right = tnode_new(inode->key|m, inode->pos + 1,
-					  inode->bits - 1);
+			left = (struct tnode *) tnode_get_child(tn, 2*i);
+			put_child(t, tn, 2*i, NULL);
+
+			if(!left)
+				BUG();
+
+			right = (struct tnode *) tnode_get_child(tn, 2*i+1);
+			put_child(t, tn, 2*i+1, NULL);
+
+			if(!right)
+				BUG();
 
-			if(!right) 
-				trie_bug("tnode_new failed");
-			
 			size = tnode_child_length(left);
 			for(j = 0; j < size; j++) {
 				put_child(t, left, j, inode->child[j]);
@@ -701,7 +793,7 @@ static struct tnode *inflate(struct trie *t, struct tnode *tn)
 	return tn;
 }
 
-static struct tnode *halve(struct trie *t, struct tnode *tn)
+static struct tnode *halve(struct trie *t, struct tnode *tn, int *err)
 {
 	struct tnode *oldtnode = tn;
 	struct node *left, *right;
@@ -712,8 +804,48 @@ static struct tnode *halve(struct trie *t, struct tnode *tn)
   
 	tn=tnode_new(oldtnode->key, oldtnode->pos, oldtnode->bits - 1);
 
-	if(!tn) 
-		trie_bug("tnode_new failed");
+	if (!tn) {
+		*err = -ENOMEM;
+		return oldtnode;
+	}
+
+	/*
+	 * Preallocate and store tnodes before the actual work so we 
+	 * don't get into an inconsistent state if memory allocation 
+	 * fails. In case of failure we return the oldnode and halve 
+	 * of tnode is ignored.
+	 */
+
+	for(i = 0; i < olen; i += 2) {
+		left = tnode_get_child(oldtnode, i);
+		right = tnode_get_child(oldtnode, i+1);
+    
+		/* Two nonempty children */
+		if( left && right)  {
+			struct tnode *newBinNode =
+				tnode_new(left->key, tn->pos + tn->bits, 1);
+
+			if(!newBinNode) {
+				*err = -ENOMEM; 
+				break;
+			}
+			put_child(t, tn, i/2, (struct node *)newBinNode);
+		}
+	}
+
+	if(*err) {
+		int size = tnode_child_length(tn);
+		int j;
+
+		for(j = 0; j < size; j++) 
+			if( tn->child[j])
+				tnode_free((struct tnode *)tn->child[j]);
+
+		tnode_free(tn);
+		
+		*err = -ENOMEM;
+		return oldtnode;
+	}
 
 	for(i = 0; i < olen; i += 2) {
 		left = tnode_get_child(oldtnode, i);
@@ -730,10 +862,11 @@ static struct tnode *halve(struct trie *t, struct tnode *tn)
 		/* Two nonempty children */
 		else {
 			struct tnode *newBinNode =
-				tnode_new(left->key, tn->pos + tn->bits, 1);
+				(struct tnode *) tnode_get_child(tn, i/2);
+			put_child(t, tn, i/2, NULL);
 
 			if(!newBinNode) 
-				trie_bug("tnode_new failed");
+				BUG();
 
 			put_child(t, newBinNode, 0, left);
 			put_child(t, newBinNode, 1, right);
@@ -2301,6 +2434,7 @@ static void collect_and_show(struct trie *t, struct seq_file *seq)
 	seq_printf(seq,"semantic match passed = %d\n", t->stats.semantic_match_passed);
 	seq_printf(seq,"semantic match miss = %d\n", t->stats.semantic_match_miss);
 	seq_printf(seq,"null node hit= %d\n", t->stats.null_node_hit);
+	seq_printf(seq,"skipped node resize = %d\n", t->stats.resize_node_skipped);
 #ifdef CLEAR_STATS
 	memset(&(t->stats), 0, sizeof(t->stats));
 #endif

net/ipv4/ip_output.c

@@ -389,7 +389,6 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from)
 	to->pkt_type = from->pkt_type;
 	to->priority = from->priority;
 	to->protocol = from->protocol;
-	to->security = from->security;
 	dst_release(to->dst);
 	to->dst = dst_clone(from->dst);
 	to->dev = from->dev;
@@ -1329,23 +1328,8 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar
 	ip_rt_put(rt);
 }
 
-/*
- *	IP protocol layer initialiser
- */
-
-static struct packet_type ip_packet_type = {
-	.type = __constant_htons(ETH_P_IP),
-	.func = ip_rcv,
-};
-
-/*
- *	IP registers the packet type and then calls the subprotocol initialisers
- */
-
 void __init ip_init(void)
 {
-	dev_add_pack(&ip_packet_type);
-
 	ip_rt_init();
 	inet_initpeers();
 

net/ipv4/route.c

@@ -54,6 +54,7 @@
  *		Marc Boucher	:	routing by fwmark
  *	Robert Olsson		:	Added rt_cache statistics
  *	Arnaldo C. Melo		:	Convert proc stuff to seq_file
+ *	Eric Dumazet		:	hashed spinlocks and rt_check_expire() fixes.
  *
  *		This program is free software; you can redistribute it and/or
  *		modify it under the terms of the GNU General Public License
@@ -70,6 +71,7 @@
 #include <linux/kernel.h>
 #include <linux/sched.h>
 #include <linux/mm.h>
+#include <linux/bootmem.h>
 #include <linux/string.h>
 #include <linux/socket.h>
 #include <linux/sockios.h>
@@ -201,8 +203,37 @@ __u8 ip_tos2prio[16] = {
 
 struct rt_hash_bucket {
 	struct rtable	*chain;
-	spinlock_t	lock;
-} __attribute__((__aligned__(8)));
+};
+#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
+/*
+ * Instead of using one spinlock for each rt_hash_bucket, we use a table of spinlocks
+ * The size of this table is a power of two and depends on the number of CPUS.
+ */
+#if NR_CPUS >= 32
+#define RT_HASH_LOCK_SZ	4096
+#elif NR_CPUS >= 16
+#define RT_HASH_LOCK_SZ	2048
+#elif NR_CPUS >= 8
+#define RT_HASH_LOCK_SZ	1024
+#elif NR_CPUS >= 4
+#define RT_HASH_LOCK_SZ	512
+#else
+#define RT_HASH_LOCK_SZ	256
+#endif
+
+static spinlock_t	*rt_hash_locks;
+# define rt_hash_lock_addr(slot) &rt_hash_locks[(slot) & (RT_HASH_LOCK_SZ - 1)]
+# define rt_hash_lock_init()	{ \
+		int i; \
+		rt_hash_locks = kmalloc(sizeof(spinlock_t) * RT_HASH_LOCK_SZ, GFP_KERNEL); \
+		if (!rt_hash_locks) panic("IP: failed to allocate rt_hash_locks\n"); \
+		for (i = 0; i < RT_HASH_LOCK_SZ; i++) \
+			spin_lock_init(&rt_hash_locks[i]); \
+		}
+#else
+# define rt_hash_lock_addr(slot) NULL
+# define rt_hash_lock_init()
+#endif
 
 static struct rt_hash_bucket 	*rt_hash_table;
 static unsigned			rt_hash_mask;
@@ -575,19 +606,26 @@ static struct rtable **rt_remove_balanced_route(struct rtable **chain_head,
 /* This runs via a timer and thus is always in BH context. */
 static void rt_check_expire(unsigned long dummy)
 {
-	static int rover;
-	int i = rover, t;
+	static unsigned int rover;
+	unsigned int i = rover, goal;
 	struct rtable *rth, **rthp;
 	unsigned long now = jiffies;
-
-	for (t = ip_rt_gc_interval << rt_hash_log; t >= 0;
-	     t -= ip_rt_gc_timeout) {
+	u64 mult;
+
+	mult = ((u64)ip_rt_gc_interval) << rt_hash_log;
+	if (ip_rt_gc_timeout > 1)
+		do_div(mult, ip_rt_gc_timeout);
+	goal = (unsigned int)mult;
+	if (goal > rt_hash_mask) goal = rt_hash_mask + 1;
+	for (; goal > 0; goal--) {
 		unsigned long tmo = ip_rt_gc_timeout;
 
 		i = (i + 1) & rt_hash_mask;
 		rthp = &rt_hash_table[i].chain;
 
-		spin_lock(&rt_hash_table[i].lock);
+		if (*rthp == 0)
+			continue;
+		spin_lock(rt_hash_lock_addr(i));
 		while ((rth = *rthp) != NULL) {
 			if (rth->u.dst.expires) {
 				/* Entry is expired even if it is in use */
@@ -620,14 +658,14 @@ static void rt_check_expire(unsigned long dummy)
  			rt_free(rth);
 #endif /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */
 		}
-		spin_unlock(&rt_hash_table[i].lock);
+		spin_unlock(rt_hash_lock_addr(i));
 
 		/* Fallback loop breaker. */
 		if (time_after(jiffies, now))
 			break;
 	}
 	rover = i;
-	mod_timer(&rt_periodic_timer, now + ip_rt_gc_interval);
+	mod_timer(&rt_periodic_timer, jiffies + ip_rt_gc_interval);
 }
 
 /* This can run from both BH and non-BH contexts, the latter
@@ -643,11 +681,11 @@ static void rt_run_flush(unsigned long dummy)
 	get_random_bytes(&rt_hash_rnd, 4);
 
 	for (i = rt_hash_mask; i >= 0; i--) {
-		spin_lock_bh(&rt_hash_table[i].lock);
+		spin_lock_bh(rt_hash_lock_addr(i));
 		rth = rt_hash_table[i].chain;
 		if (rth)
 			rt_hash_table[i].chain = NULL;
-		spin_unlock_bh(&rt_hash_table[i].lock);
+		spin_unlock_bh(rt_hash_lock_addr(i));
 
 		for (; rth; rth = next) {
 			next = rth->u.rt_next;
@@ -780,7 +818,7 @@ static int rt_garbage_collect(void)
 
 			k = (k + 1) & rt_hash_mask;
 			rthp = &rt_hash_table[k].chain;
-			spin_lock_bh(&rt_hash_table[k].lock);
+			spin_lock_bh(rt_hash_lock_addr(k));
 			while ((rth = *rthp) != NULL) {
 				if (!rt_may_expire(rth, tmo, expire)) {
 					tmo >>= 1;
@@ -812,7 +850,7 @@ static int rt_garbage_collect(void)
 				goal--;
 #endif /* CONFIG_IP_ROUTE_MULTIPATH_CACHED */
 			}
-			spin_unlock_bh(&rt_hash_table[k].lock);
+			spin_unlock_bh(rt_hash_lock_addr(k));
 			if (goal <= 0)
 				break;
 		}
@@ -882,7 +920,7 @@ static int rt_intern_hash(unsigned hash, struct rtable *rt, struct rtable **rp)
 
 	rthp = &rt_hash_table[hash].chain;
 
-	spin_lock_bh(&rt_hash_table[hash].lock);
+	spin_lock_bh(rt_hash_lock_addr(hash));
 	while ((rth = *rthp) != NULL) {
 #ifdef CONFIG_IP_ROUTE_MULTIPATH_CACHED
 		if (!(rth->u.dst.flags & DST_BALANCED) &&
@@ -908,7 +946,7 @@ static int rt_intern_hash(unsigned hash, struct rtable *rt, struct rtable **rp)
 			rth->u.dst.__use++;
 			dst_hold(&rth->u.dst);
 			rth->u.dst.lastuse = now;
-			spin_unlock_bh(&rt_hash_table[hash].lock);
+			spin_unlock_bh(rt_hash_lock_addr(hash));
 
 			rt_drop(rt);
 			*rp = rth;
@@ -949,7 +987,7 @@ static int rt_intern_hash(unsigned hash, struct rtable *rt, struct rtable **rp)
 	if (rt->rt_type == RTN_UNICAST || rt->fl.iif == 0) {
 		int err = arp_bind_neighbour(&rt->u.dst);
 		if (err) {
-			spin_unlock_bh(&rt_hash_table[hash].lock);
+			spin_unlock_bh(rt_hash_lock_addr(hash));
 
 			if (err != -ENOBUFS) {
 				rt_drop(rt);
@@ -990,7 +1028,7 @@ static int rt_intern_hash(unsigned hash, struct rtable *rt, struct rtable **rp)
 	}
 #endif
 	rt_hash_table[hash].chain = rt;
-	spin_unlock_bh(&rt_hash_table[hash].lock);
+	spin_unlock_bh(rt_hash_lock_addr(hash));
 	*rp = rt;
 	return 0;
 }
@@ -1058,7 +1096,7 @@ static void rt_del(unsigned hash, struct rtable *rt)
 {
 	struct rtable **rthp;
 
-	spin_lock_bh(&rt_hash_table[hash].lock);
+	spin_lock_bh(rt_hash_lock_addr(hash));
 	ip_rt_put(rt);
 	for (rthp = &rt_hash_table[hash].chain; *rthp;
 	     rthp = &(*rthp)->u.rt_next)
@@ -1067,7 +1105,7 @@ static void rt_del(unsigned hash, struct rtable *rt)
 			rt_free(rt);
 			break;
 		}
-	spin_unlock_bh(&rt_hash_table[hash].lock);
+	spin_unlock_bh(rt_hash_lock_addr(hash));
 }
 
 void ip_rt_redirect(u32 old_gw, u32 daddr, u32 new_gw,
@@ -3073,12 +3111,14 @@ __setup("rhash_entries=", set_rhash_entries);
 
 int __init ip_rt_init(void)
 {
-	int i, order, goal, rc = 0;
+	int rc = 0;
 
 	rt_hash_rnd = (int) ((num_physpages ^ (num_physpages>>8)) ^
 			     (jiffies ^ (jiffies >> 7)));
 
 #ifdef CONFIG_NET_CLS_ROUTE
+	{
+	int order;
 	for (order = 0;
 	     (PAGE_SIZE << order) < 256 * sizeof(struct ip_rt_acct) * NR_CPUS; order++)
 		/* NOTHING */;
@@ -3086,6 +3126,7 @@ int __init ip_rt_init(void)
 	if (!ip_rt_acct)
 		panic("IP: failed to allocate ip_rt_acct\n");
 	memset(ip_rt_acct, 0, PAGE_SIZE << order);
+	}
 #endif
 
 	ipv4_dst_ops.kmem_cachep = kmem_cache_create("ip_dst_cache",
@@ -3096,36 +3137,19 @@ int __init ip_rt_init(void)
 	if (!ipv4_dst_ops.kmem_cachep)
 		panic("IP: failed to allocate ip_dst_cache\n");
 
-	goal = num_physpages >> (26 - PAGE_SHIFT);
-	if (rhash_entries)
-		goal = (rhash_entries * sizeof(struct rt_hash_bucket)) >> PAGE_SHIFT;
-	for (order = 0; (1UL << order) < goal; order++)
-		/* NOTHING */;
-
-	do {
-		rt_hash_mask = (1UL << order) * PAGE_SIZE /
-			sizeof(struct rt_hash_bucket);
-		while (rt_hash_mask & (rt_hash_mask - 1))
-			rt_hash_mask--;
-		rt_hash_table = (struct rt_hash_bucket *)
-			__get_free_pages(GFP_ATOMIC, order);
-	} while (rt_hash_table == NULL && --order > 0);
-
-	if (!rt_hash_table)
-		panic("Failed to allocate IP route cache hash table\n");
-
-	printk(KERN_INFO "IP: routing cache hash table of %u buckets, %ldKbytes\n",
-	       rt_hash_mask,
-	       (long) (rt_hash_mask * sizeof(struct rt_hash_bucket)) / 1024);
-
-	for (rt_hash_log = 0; (1 << rt_hash_log) != rt_hash_mask; rt_hash_log++)
-		/* NOTHING */;
-
-	rt_hash_mask--;
-	for (i = 0; i <= rt_hash_mask; i++) {
-		spin_lock_init(&rt_hash_table[i].lock);
-		rt_hash_table[i].chain = NULL;
-	}
+	rt_hash_table = (struct rt_hash_bucket *)
+		alloc_large_system_hash("IP route cache",
+					sizeof(struct rt_hash_bucket),
+					rhash_entries,
+					(num_physpages >= 128 * 1024) ?
+						(27 - PAGE_SHIFT) :
+						(29 - PAGE_SHIFT),
+					HASH_HIGHMEM,
+					&rt_hash_log,
+					&rt_hash_mask,
+					0);
+	memset(rt_hash_table, 0, (rt_hash_mask + 1) * sizeof(struct rt_hash_bucket));
+	rt_hash_lock_init();
 
 	ipv4_dst_ops.gc_thresh = (rt_hash_mask + 1);
 	ip_rt_max_size = (rt_hash_mask + 1) * 16;

net/ipv4/tcp.c

@@ -615,7 +615,7 @@ static ssize_t do_tcp_sendpages(struct sock *sk, struct page **pages, int poffse
 			 size_t psize, int flags)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
-	int mss_now;
+	int mss_now, size_goal;
 	int err;
 	ssize_t copied;
 	long timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);
@@ -628,6 +628,7 @@ static ssize_t do_tcp_sendpages(struct sock *sk, struct page **pages, int poffse
 	clear_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags);
 
 	mss_now = tcp_current_mss(sk, !(flags&MSG_OOB));
+	size_goal = tp->xmit_size_goal;
 	copied = 0;
 
 	err = -EPIPE;
@@ -641,7 +642,7 @@ static ssize_t do_tcp_sendpages(struct sock *sk, struct page **pages, int poffse
 		int offset = poffset % PAGE_SIZE;
 		int size = min_t(size_t, psize, PAGE_SIZE - offset);
 
-		if (!sk->sk_send_head || (copy = mss_now - skb->len) <= 0) {
+		if (!sk->sk_send_head || (copy = size_goal - skb->len) <= 0) {
 new_segment:
 			if (!sk_stream_memory_free(sk))
 				goto wait_for_sndbuf;
@@ -652,7 +653,7 @@ static ssize_t do_tcp_sendpages(struct sock *sk, struct page **pages, int poffse
 				goto wait_for_memory;
 
 			skb_entail(sk, tp, skb);
-			copy = mss_now;
+			copy = size_goal;
 		}
 
 		if (copy > size)
@@ -693,7 +694,7 @@ static ssize_t do_tcp_sendpages(struct sock *sk, struct page **pages, int poffse
 		if (!(psize -= copy))
 			goto out;
 
-		if (skb->len != mss_now || (flags & MSG_OOB))
+		if (skb->len < mss_now || (flags & MSG_OOB))
 			continue;
 
 		if (forced_push(tp)) {
@@ -713,6 +714,7 @@ static ssize_t do_tcp_sendpages(struct sock *sk, struct page **pages, int poffse
 			goto do_error;
 
 		mss_now = tcp_current_mss(sk, !(flags&MSG_OOB));
+		size_goal = tp->xmit_size_goal;
 	}
 
 out:
@@ -754,15 +756,20 @@ ssize_t tcp_sendpage(struct socket *sock, struct page *page, int offset,
 
 static inline int select_size(struct sock *sk, struct tcp_sock *tp)
 {
-	int tmp = tp->mss_cache_std;
+	int tmp = tp->mss_cache;
 
 	if (sk->sk_route_caps & NETIF_F_SG) {
-		int pgbreak = SKB_MAX_HEAD(MAX_TCP_HEADER);
+		if (sk->sk_route_caps & NETIF_F_TSO)
+			tmp = 0;
+		else {
+			int pgbreak = SKB_MAX_HEAD(MAX_TCP_HEADER);
 
-		if (tmp >= pgbreak &&
-		    tmp <= pgbreak + (MAX_SKB_FRAGS - 1) * PAGE_SIZE)
-			tmp = pgbreak;
+			if (tmp >= pgbreak &&
+			    tmp <= pgbreak + (MAX_SKB_FRAGS - 1) * PAGE_SIZE)
+				tmp = pgbreak;
+		}
 	}
+
 	return tmp;
 }
 
@@ -773,7 +780,7 @@ int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *skb;
 	int iovlen, flags;
-	int mss_now;
+	int mss_now, size_goal;
 	int err, copied;
 	long timeo;
 
@@ -792,6 +799,7 @@ int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 	clear_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags);
 
 	mss_now = tcp_current_mss(sk, !(flags&MSG_OOB));
+	size_goal = tp->xmit_size_goal;
 
 	/* Ok commence sending. */
 	iovlen = msg->msg_iovlen;
@@ -814,7 +822,7 @@ int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 			skb = sk->sk_write_queue.prev;
 
 			if (!sk->sk_send_head ||
-			    (copy = mss_now - skb->len) <= 0) {
+			    (copy = size_goal - skb->len) <= 0) {
 
 new_segment:
 				/* Allocate new segment. If the interface is SG,
@@ -837,7 +845,7 @@ int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 					skb->ip_summed = CHECKSUM_HW;
 
 				skb_entail(sk, tp, skb);
-				copy = mss_now;
+				copy = size_goal;
 			}
 
 			/* Try to append data to the end of skb. */
@@ -872,11 +880,6 @@ int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 					tcp_mark_push(tp, skb);
 					goto new_segment;
 				} else if (page) {
-					/* If page is cached, align
-					 * offset to L1 cache boundary
-					 */
-					off = (off + L1_CACHE_BYTES - 1) &
-					      ~(L1_CACHE_BYTES - 1);
 					if (off == PAGE_SIZE) {
 						put_page(page);
 						TCP_PAGE(sk) = page = NULL;
@@ -937,7 +940,7 @@ int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 			if ((seglen -= copy) == 0 && iovlen == 0)
 				goto out;
 
-			if (skb->len != mss_now || (flags & MSG_OOB))
+			if (skb->len < mss_now || (flags & MSG_OOB))
 				continue;
 
 			if (forced_push(tp)) {
@@ -957,6 +960,7 @@ int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 				goto do_error;
 
 			mss_now = tcp_current_mss(sk, !(flags&MSG_OOB));
+			size_goal = tp->xmit_size_goal;
 		}
 	}
 
@@ -2128,7 +2132,7 @@ void tcp_get_info(struct sock *sk, struct tcp_info *info)
 
 	info->tcpi_rto = jiffies_to_usecs(tp->rto);
 	info->tcpi_ato = jiffies_to_usecs(tp->ack.ato);
-	info->tcpi_snd_mss = tp->mss_cache_std;
+	info->tcpi_snd_mss = tp->mss_cache;
 	info->tcpi_rcv_mss = tp->ack.rcv_mss;
 
 	info->tcpi_unacked = tp->packets_out;
@@ -2178,7 +2182,7 @@ int tcp_getsockopt(struct sock *sk, int level, int optname, char __user *optval,
 
 	switch (optname) {
 	case TCP_MAXSEG:
-		val = tp->mss_cache_std;
+		val = tp->mss_cache;
 		if (!val && ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)))
 			val = tp->rx_opt.user_mss;
 		break;

net/ipv4/tcp_input.c

@@ -740,10 +740,10 @@ __u32 tcp_init_cwnd(struct tcp_sock *tp, struct dst_entry *dst)
 	__u32 cwnd = (dst ? dst_metric(dst, RTAX_INITCWND) : 0);
 
 	if (!cwnd) {
-		if (tp->mss_cache_std > 1460)
+		if (tp->mss_cache > 1460)
 			cwnd = 2;
 		else
-			cwnd = (tp->mss_cache_std > 1095) ? 3 : 4;
+			cwnd = (tp->mss_cache > 1095) ? 3 : 4;
 	}
 	return min_t(__u32, cwnd, tp->snd_cwnd_clamp);
 }
@@ -914,7 +914,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 	if (sk->sk_route_caps & NETIF_F_TSO) {
 		sk->sk_route_caps &= ~NETIF_F_TSO;
 		sock_set_flag(sk, SOCK_NO_LARGESEND);
-		tp->mss_cache = tp->mss_cache_std;
+		tp->mss_cache = tp->mss_cache;
 	}
 
 	if (!tp->sacked_out)
@@ -1077,7 +1077,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_
 			    (IsFack(tp) ||
 			     !before(lost_retrans,
 				     TCP_SKB_CB(skb)->ack_seq + tp->reordering *
-				     tp->mss_cache_std))) {
+				     tp->mss_cache))) {
 				TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
 				tp->retrans_out -= tcp_skb_pcount(skb);
 
@@ -1957,15 +1957,6 @@ static inline void tcp_ack_packets_out(struct sock *sk, struct tcp_sock *tp)
 	}
 }
 
-/* There is one downside to this scheme.  Although we keep the
- * ACK clock ticking, adjusting packet counters and advancing
- * congestion window, we do not liberate socket send buffer
- * space.
- *
- * Mucking with skb->truesize and sk->sk_wmem_alloc et al.
- * then making a write space wakeup callback is a possible
- * future enhancement.  WARNING: it is not trivial to make.
- */
 static int tcp_tso_acked(struct sock *sk, struct sk_buff *skb,
 			 __u32 now, __s32 *seq_rtt)
 {
@@ -2047,7 +2038,8 @@ static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p, s32 *seq_usrtt
 		 * the other end.
 		 */
 		if (after(scb->end_seq, tp->snd_una)) {
-			if (tcp_skb_pcount(skb) > 1)
+			if (tcp_skb_pcount(skb) > 1 &&
+			    after(tp->snd_una, scb->seq))
 				acked |= tcp_tso_acked(sk, skb,
 						       now, &seq_rtt);
 			break;
@@ -3308,6 +3300,28 @@ void tcp_cwnd_application_limited(struct sock *sk)
 	tp->snd_cwnd_stamp = tcp_time_stamp;
 }
 
+static inline int tcp_should_expand_sndbuf(struct sock *sk, struct tcp_sock *tp)
+{
+	/* If the user specified a specific send buffer setting, do
+	 * not modify it.
+	 */
+	if (sk->sk_userlocks & SOCK_SNDBUF_LOCK)
+		return 0;
+
+	/* If we are under global TCP memory pressure, do not expand.  */
+	if (tcp_memory_pressure)
+		return 0;
+
+	/* If we are under soft global TCP memory pressure, do not expand.  */
+	if (atomic_read(&tcp_memory_allocated) >= sysctl_tcp_mem[0])
+		return 0;
+
+	/* If we filled the congestion window, do not expand.  */
+	if (tp->packets_out >= tp->snd_cwnd)
+		return 0;
+
+	return 1;
+}
 
 /* When incoming ACK allowed to free some skb from write_queue,
  * we remember this event in flag SOCK_QUEUE_SHRUNK and wake up socket
@@ -3319,11 +3333,8 @@ static void tcp_new_space(struct sock *sk)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 
-	if (tp->packets_out < tp->snd_cwnd &&
-	    !(sk->sk_userlocks & SOCK_SNDBUF_LOCK) &&
-	    !tcp_memory_pressure &&
-	    atomic_read(&tcp_memory_allocated) < sysctl_tcp_mem[0]) {
- 		int sndmem = max_t(u32, tp->rx_opt.mss_clamp, tp->mss_cache_std) +
+	if (tcp_should_expand_sndbuf(sk, tp)) {
+ 		int sndmem = max_t(u32, tp->rx_opt.mss_clamp, tp->mss_cache) +
 			MAX_TCP_HEADER + 16 + sizeof(struct sk_buff),
 		    demanded = max_t(unsigned int, tp->snd_cwnd,
 						   tp->reordering + 1);
@@ -3346,22 +3357,9 @@ static inline void tcp_check_space(struct sock *sk)
 	}
 }
 
-static void __tcp_data_snd_check(struct sock *sk, struct sk_buff *skb)
-{
-	struct tcp_sock *tp = tcp_sk(sk);
-
-	if (after(TCP_SKB_CB(skb)->end_seq, tp->snd_una + tp->snd_wnd) ||
-	    tcp_packets_in_flight(tp) >= tp->snd_cwnd ||
-	    tcp_write_xmit(sk, tp->nonagle))
-		tcp_check_probe_timer(sk, tp);
-}
-
-static __inline__ void tcp_data_snd_check(struct sock *sk)
+static __inline__ void tcp_data_snd_check(struct sock *sk, struct tcp_sock *tp)
 {
-	struct sk_buff *skb = sk->sk_send_head;
-
-	if (skb != NULL)
-		__tcp_data_snd_check(sk, skb);
+	tcp_push_pending_frames(sk, tp);
 	tcp_check_space(sk);
 }
 
@@ -3655,7 +3653,7 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
 				 */
 				tcp_ack(sk, skb, 0);
 				__kfree_skb(skb); 
-				tcp_data_snd_check(sk);
+				tcp_data_snd_check(sk, tp);
 				return 0;
 			} else { /* Header too small */
 				TCP_INC_STATS_BH(TCP_MIB_INERRS);
@@ -3721,7 +3719,7 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
 			if (TCP_SKB_CB(skb)->ack_seq != tp->snd_una) {
 				/* Well, only one small jumplet in fast path... */
 				tcp_ack(sk, skb, FLAG_DATA);
-				tcp_data_snd_check(sk);
+				tcp_data_snd_check(sk, tp);
 				if (!tcp_ack_scheduled(tp))
 					goto no_ack;
 			}
@@ -3799,7 +3797,7 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
 	/* step 7: process the segment text */
 	tcp_data_queue(sk, skb);
 
-	tcp_data_snd_check(sk);
+	tcp_data_snd_check(sk, tp);
 	tcp_ack_snd_check(sk);
 	return 0;
 
@@ -4109,7 +4107,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 		/* Do step6 onward by hand. */
 		tcp_urg(sk, skb, th);
 		__kfree_skb(skb);
-		tcp_data_snd_check(sk);
+		tcp_data_snd_check(sk, tp);
 		return 0;
 	}
 
@@ -4300,7 +4298,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
 
 	/* tcp_data could move socket to TIME-WAIT */
 	if (sk->sk_state != TCP_CLOSE) {
-		tcp_data_snd_check(sk);
+		tcp_data_snd_check(sk, tp);
 		tcp_ack_snd_check(sk);
 	}
 

net/ipv4/tcp_ipv4.c

@@ -2045,7 +2045,7 @@ static int tcp_v4_init_sock(struct sock *sk)
 	 */
 	tp->snd_ssthresh = 0x7fffffff;	/* Infinity */
 	tp->snd_cwnd_clamp = ~0;
-	tp->mss_cache_std = tp->mss_cache = 536;
+	tp->mss_cache = 536;
 
 	tp->reordering = sysctl_tcp_reordering;
 	tp->ca_ops = &tcp_init_congestion_ops;

net/ipv4/tcp_output.c

@@ -49,7 +49,7 @@ int sysctl_tcp_retrans_collapse = 1;
  * will allow a single TSO frame to consume.  Building TSO frames
  * which are too large can cause TCP streams to be bursty.
  */
-int sysctl_tcp_tso_win_divisor = 8;
+int sysctl_tcp_tso_win_divisor = 3;
 
 static inline void update_send_head(struct sock *sk, struct tcp_sock *tp,
 				    struct sk_buff *skb)
@@ -140,11 +140,11 @@ static inline void tcp_event_data_sent(struct tcp_sock *tp,
 		tp->ack.pingpong = 1;
 }
 
-static __inline__ void tcp_event_ack_sent(struct sock *sk)
+static __inline__ void tcp_event_ack_sent(struct sock *sk, unsigned int pkts)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 
-	tcp_dec_quickack_mode(tp);
+	tcp_dec_quickack_mode(tp, pkts);
 	tcp_clear_xmit_timer(sk, TCP_TIME_DACK);
 }
 
@@ -355,7 +355,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb)
 		tp->af_specific->send_check(sk, th, skb->len, skb);
 
 		if (tcb->flags & TCPCB_FLAG_ACK)
-			tcp_event_ack_sent(sk);
+			tcp_event_ack_sent(sk, tcp_skb_pcount(skb));
 
 		if (skb->len != tcp_header_size)
 			tcp_event_data_sent(tp, skb, sk);
@@ -403,42 +403,11 @@ static void tcp_queue_skb(struct sock *sk, struct sk_buff *skb)
 		sk->sk_send_head = skb;
 }
 
-static inline void tcp_tso_set_push(struct sk_buff *skb)
-{
-	/* Force push to be on for any TSO frames to workaround
-	 * problems with busted implementations like Mac OS-X that
-	 * hold off socket receive wakeups until push is seen.
-	 */
-	if (tcp_skb_pcount(skb) > 1)
-		TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH;
-}
-
-/* Send _single_ skb sitting at the send head. This function requires
- * true push pending frames to setup probe timer etc.
- */
-void tcp_push_one(struct sock *sk, unsigned cur_mss)
+static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
-	struct sk_buff *skb = sk->sk_send_head;
 
-	if (tcp_snd_test(sk, skb, cur_mss, TCP_NAGLE_PUSH)) {
-		/* Send it out now. */
-		TCP_SKB_CB(skb)->when = tcp_time_stamp;
-		tcp_tso_set_push(skb);
-		if (!tcp_transmit_skb(sk, skb_clone(skb, sk->sk_allocation))) {
-			sk->sk_send_head = NULL;
-			tp->snd_nxt = TCP_SKB_CB(skb)->end_seq;
-			tcp_packets_out_inc(sk, tp, skb);
-			return;
-		}
-	}
-}
-
-void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb)
-{
-	struct tcp_sock *tp = tcp_sk(sk);
-
-	if (skb->len <= tp->mss_cache_std ||
+	if (skb->len <= tp->mss_cache ||
 	    !(sk->sk_route_caps & NETIF_F_TSO)) {
 		/* Avoid the costly divide in the normal
 		 * non-TSO case.
@@ -448,10 +417,10 @@ void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb)
 	} else {
 		unsigned int factor;
 
-		factor = skb->len + (tp->mss_cache_std - 1);
-		factor /= tp->mss_cache_std;
+		factor = skb->len + (tp->mss_cache - 1);
+		factor /= tp->mss_cache;
 		skb_shinfo(skb)->tso_segs = factor;
-		skb_shinfo(skb)->tso_size = tp->mss_cache_std;
+		skb_shinfo(skb)->tso_size = tp->mss_cache;
 	}
 }
 
@@ -537,6 +506,7 @@ static int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len)
 	}
 
 	/* Link BUFF into the send queue. */
+	skb_header_release(buff);
 	__skb_append(skb, buff);
 
 	return 0;
@@ -657,7 +627,7 @@ unsigned int tcp_sync_mss(struct sock *sk, u32 pmtu)
 
 	/* And store cached results */
 	tp->pmtu_cookie = pmtu;
-	tp->mss_cache = tp->mss_cache_std = mss_now;
+	tp->mss_cache = mss_now;
 
 	return mss_now;
 }
@@ -669,57 +639,316 @@ unsigned int tcp_sync_mss(struct sock *sk, u32 pmtu)
  * cannot be large. However, taking into account rare use of URG, this
  * is not a big flaw.
  */
-
-unsigned int tcp_current_mss(struct sock *sk, int large)
+unsigned int tcp_current_mss(struct sock *sk, int large_allowed)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct dst_entry *dst = __sk_dst_get(sk);
-	unsigned int do_large, mss_now;
+	u32 mss_now;
+	u16 xmit_size_goal;
+	int doing_tso = 0;
+
+	mss_now = tp->mss_cache;
+
+	if (large_allowed &&
+	    (sk->sk_route_caps & NETIF_F_TSO) &&
+	    !tp->urg_mode)
+		doing_tso = 1;
 
-	mss_now = tp->mss_cache_std;
 	if (dst) {
 		u32 mtu = dst_mtu(dst);
 		if (mtu != tp->pmtu_cookie)
 			mss_now = tcp_sync_mss(sk, mtu);
 	}
 
-	do_large = (large &&
-		    (sk->sk_route_caps & NETIF_F_TSO) &&
-		    !tp->urg_mode);
+	if (tp->rx_opt.eff_sacks)
+		mss_now -= (TCPOLEN_SACK_BASE_ALIGNED +
+			    (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK));
 
-	if (do_large) {
-		unsigned int large_mss, factor, limit;
+	xmit_size_goal = mss_now;
 
-		large_mss = 65535 - tp->af_specific->net_header_len -
+	if (doing_tso) {
+		xmit_size_goal = 65535 -
+			tp->af_specific->net_header_len -
 			tp->ext_header_len - tp->tcp_header_len;
 
-		if (tp->max_window && large_mss > (tp->max_window>>1))
-			large_mss = max((tp->max_window>>1),
-					68U - tp->tcp_header_len);
+		if (tp->max_window &&
+		    (xmit_size_goal > (tp->max_window >> 1)))
+			xmit_size_goal = max((tp->max_window >> 1),
+					     68U - tp->tcp_header_len);
+
+		xmit_size_goal -= (xmit_size_goal % mss_now);
+	}
+	tp->xmit_size_goal = xmit_size_goal;
 
-		factor = large_mss / mss_now;
+	return mss_now;
+}
 
-		/* Always keep large mss multiple of real mss, but
-		 * do not exceed 1/tso_win_divisor of the congestion window
-		 * so we can keep the ACK clock ticking and minimize
-		 * bursting.
-		 */
-		limit = tp->snd_cwnd;
-		if (sysctl_tcp_tso_win_divisor)
-			limit /= sysctl_tcp_tso_win_divisor;
-		limit = max(1U, limit);
-		if (factor > limit)
-			factor = limit;
+/* Congestion window validation. (RFC2861) */
 
-		tp->mss_cache = mss_now * factor;
+static inline void tcp_cwnd_validate(struct sock *sk, struct tcp_sock *tp)
+{
+	__u32 packets_out = tp->packets_out;
+
+	if (packets_out >= tp->snd_cwnd) {
+		/* Network is feed fully. */
+		tp->snd_cwnd_used = 0;
+		tp->snd_cwnd_stamp = tcp_time_stamp;
+	} else {
+		/* Network starves. */
+		if (tp->packets_out > tp->snd_cwnd_used)
+			tp->snd_cwnd_used = tp->packets_out;
 
-		mss_now = tp->mss_cache;
+		if ((s32)(tcp_time_stamp - tp->snd_cwnd_stamp) >= tp->rto)
+			tcp_cwnd_application_limited(sk);
 	}
+}
 
-	if (tp->rx_opt.eff_sacks)
-		mss_now -= (TCPOLEN_SACK_BASE_ALIGNED +
-			    (tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK));
-	return mss_now;
+static unsigned int tcp_window_allows(struct tcp_sock *tp, struct sk_buff *skb, unsigned int mss_now, unsigned int cwnd)
+{
+	u32 window, cwnd_len;
+
+	window = (tp->snd_una + tp->snd_wnd - TCP_SKB_CB(skb)->seq);
+	cwnd_len = mss_now * cwnd;
+	return min(window, cwnd_len);
+}
+
+/* Can at least one segment of SKB be sent right now, according to the
+ * congestion window rules?  If so, return how many segments are allowed.
+ */
+static inline unsigned int tcp_cwnd_test(struct tcp_sock *tp, struct sk_buff *skb)
+{
+	u32 in_flight, cwnd;
+
+	/* Don't be strict about the congestion window for the final FIN.  */
+	if (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN)
+		return 1;
+
+	in_flight = tcp_packets_in_flight(tp);
+	cwnd = tp->snd_cwnd;
+	if (in_flight < cwnd)
+		return (cwnd - in_flight);
+
+	return 0;
+}
+
+/* This must be invoked the first time we consider transmitting
+ * SKB onto the wire.
+ */
+static inline int tcp_init_tso_segs(struct sock *sk, struct sk_buff *skb)
+{
+	int tso_segs = tcp_skb_pcount(skb);
+
+	if (!tso_segs) {
+		tcp_set_skb_tso_segs(sk, skb);
+		tso_segs = tcp_skb_pcount(skb);
+	}
+	return tso_segs;
+}
+
+static inline int tcp_minshall_check(const struct tcp_sock *tp)
+{
+	return after(tp->snd_sml,tp->snd_una) &&
+		!after(tp->snd_sml, tp->snd_nxt);
+}
+
+/* Return 0, if packet can be sent now without violation Nagle's rules:
+ * 1. It is full sized.
+ * 2. Or it contains FIN. (already checked by caller)
+ * 3. Or TCP_NODELAY was set.
+ * 4. Or TCP_CORK is not set, and all sent packets are ACKed.
+ *    With Minshall's modification: all sent small packets are ACKed.
+ */
+
+static inline int tcp_nagle_check(const struct tcp_sock *tp,
+				  const struct sk_buff *skb, 
+				  unsigned mss_now, int nonagle)
+{
+	return (skb->len < mss_now &&
+		((nonagle&TCP_NAGLE_CORK) ||
+		 (!nonagle &&
+		  tp->packets_out &&
+		  tcp_minshall_check(tp))));
+}
+
+/* Return non-zero if the Nagle test allows this packet to be
+ * sent now.
+ */
+static inline int tcp_nagle_test(struct tcp_sock *tp, struct sk_buff *skb,
+				 unsigned int cur_mss, int nonagle)
+{
+	/* Nagle rule does not apply to frames, which sit in the middle of the
+	 * write_queue (they have no chances to get new data).
+	 *
+	 * This is implemented in the callers, where they modify the 'nonagle'
+	 * argument based upon the location of SKB in the send queue.
+	 */
+	if (nonagle & TCP_NAGLE_PUSH)
+		return 1;
+
+	/* Don't use the nagle rule for urgent data (or for the final FIN).  */
+	if (tp->urg_mode ||
+	    (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN))
+		return 1;
+
+	if (!tcp_nagle_check(tp, skb, cur_mss, nonagle))
+		return 1;
+
+	return 0;
+}
+
+/* Does at least the first segment of SKB fit into the send window? */
+static inline int tcp_snd_wnd_test(struct tcp_sock *tp, struct sk_buff *skb, unsigned int cur_mss)
+{
+	u32 end_seq = TCP_SKB_CB(skb)->end_seq;
+
+	if (skb->len > cur_mss)
+		end_seq = TCP_SKB_CB(skb)->seq + cur_mss;
+
+	return !after(end_seq, tp->snd_una + tp->snd_wnd);
+}
+
+/* This checks if the data bearing packet SKB (usually sk->sk_send_head)
+ * should be put on the wire right now.  If so, it returns the number of
+ * packets allowed by the congestion window.
+ */
+static unsigned int tcp_snd_test(struct sock *sk, struct sk_buff *skb,
+				 unsigned int cur_mss, int nonagle)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+	unsigned int cwnd_quota;
+
+	tcp_init_tso_segs(sk, skb);
+
+	if (!tcp_nagle_test(tp, skb, cur_mss, nonagle))
+		return 0;
+
+	cwnd_quota = tcp_cwnd_test(tp, skb);
+	if (cwnd_quota &&
+	    !tcp_snd_wnd_test(tp, skb, cur_mss))
+		cwnd_quota = 0;
+
+	return cwnd_quota;
+}
+
+static inline int tcp_skb_is_last(const struct sock *sk, 
+				  const struct sk_buff *skb)
+{
+	return skb->next == (struct sk_buff *)&sk->sk_write_queue;
+}
+
+int tcp_may_send_now(struct sock *sk, struct tcp_sock *tp)
+{
+	struct sk_buff *skb = sk->sk_send_head;
+
+	return (skb &&
+		tcp_snd_test(sk, skb, tcp_current_mss(sk, 1),
+			     (tcp_skb_is_last(sk, skb) ?
+			      TCP_NAGLE_PUSH :
+			      tp->nonagle)));
+}
+
+/* Trim TSO SKB to LEN bytes, put the remaining data into a new packet
+ * which is put after SKB on the list.  It is very much like
+ * tcp_fragment() except that it may make several kinds of assumptions
+ * in order to speed up the splitting operation.  In particular, we
+ * know that all the data is in scatter-gather pages, and that the
+ * packet has never been sent out before (and thus is not cloned).
+ */
+static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len)
+{
+	struct sk_buff *buff;
+	int nlen = skb->len - len;
+	u16 flags;
+
+	/* All of a TSO frame must be composed of paged data.  */
+	BUG_ON(skb->len != skb->data_len);
+
+	buff = sk_stream_alloc_pskb(sk, 0, 0, GFP_ATOMIC);
+	if (unlikely(buff == NULL))
+		return -ENOMEM;
+
+	buff->truesize = nlen;
+	skb->truesize -= nlen;
+
+	/* Correct the sequence numbers. */
+	TCP_SKB_CB(buff)->seq = TCP_SKB_CB(skb)->seq + len;
+	TCP_SKB_CB(buff)->end_seq = TCP_SKB_CB(skb)->end_seq;
+	TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(buff)->seq;
+
+	/* PSH and FIN should only be set in the second packet. */
+	flags = TCP_SKB_CB(skb)->flags;
+	TCP_SKB_CB(skb)->flags = flags & ~(TCPCB_FLAG_FIN|TCPCB_FLAG_PSH);
+	TCP_SKB_CB(buff)->flags = flags;
+
+	/* This packet was never sent out yet, so no SACK bits. */
+	TCP_SKB_CB(buff)->sacked = 0;
+
+	buff->ip_summed = skb->ip_summed = CHECKSUM_HW;
+	skb_split(skb, buff, len);
+
+	/* Fix up tso_factor for both original and new SKB.  */
+	tcp_set_skb_tso_segs(sk, skb);
+	tcp_set_skb_tso_segs(sk, buff);
+
+	/* Link BUFF into the send queue. */
+	skb_header_release(buff);
+	__skb_append(skb, buff);
+
+	return 0;
+}
+
+/* Try to defer sending, if possible, in order to minimize the amount
+ * of TSO splitting we do.  View it as a kind of TSO Nagle test.
+ *
+ * This algorithm is from John Heffner.
+ */
+static int tcp_tso_should_defer(struct sock *sk, struct tcp_sock *tp, struct sk_buff *skb)
+{
+	u32 send_win, cong_win, limit, in_flight;
+
+	if (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN)
+		return 0;
+
+	if (tp->ca_state != TCP_CA_Open)
+		return 0;
+
+	in_flight = tcp_packets_in_flight(tp);
+
+	BUG_ON(tcp_skb_pcount(skb) <= 1 ||
+	       (tp->snd_cwnd <= in_flight));
+
+	send_win = (tp->snd_una + tp->snd_wnd) - TCP_SKB_CB(skb)->seq;
+
+	/* From in_flight test above, we know that cwnd > in_flight.  */
+	cong_win = (tp->snd_cwnd - in_flight) * tp->mss_cache;
+
+	limit = min(send_win, cong_win);
+
+	/* If sk_send_head can be sent fully now, just do it.  */
+	if (skb->len <= limit)
+		return 0;
+
+	if (sysctl_tcp_tso_win_divisor) {
+		u32 chunk = min(tp->snd_wnd, tp->snd_cwnd * tp->mss_cache);
+
+		/* If at least some fraction of a window is available,
+		 * just use it.
+		 */
+		chunk /= sysctl_tcp_tso_win_divisor;
+		if (limit >= chunk)
+			return 0;
+	} else {
+		/* Different approach, try not to defer past a single
+		 * ACK.  Receiver should ACK every other full sized
+		 * frame, so if we have space for more than 3 frames
+		 * then send now.
+		 */
+		if (limit > tcp_max_burst(tp) * tp->mss_cache)
+			return 0;
+	}
+
+	/* Ok, it looks like it is advisable to defer.  */
+	return 1;
 }
 
 /* This routine writes packets to the network.  It advances the
@@ -729,57 +958,158 @@ unsigned int tcp_current_mss(struct sock *sk, int large)
  * Returns 1, if no segments are in flight and we have queued segments, but
  * cannot send anything now because of SWS or another problem.
  */
-int tcp_write_xmit(struct sock *sk, int nonagle)
+static int tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
-	unsigned int mss_now;
+	struct sk_buff *skb;
+	unsigned int tso_segs, sent_pkts;
+	int cwnd_quota;
 
 	/* If we are closed, the bytes will have to remain here.
 	 * In time closedown will finish, we empty the write queue and all
 	 * will be happy.
 	 */
-	if (sk->sk_state != TCP_CLOSE) {
-		struct sk_buff *skb;
-		int sent_pkts = 0;
+	if (unlikely(sk->sk_state == TCP_CLOSE))
+		return 0;
+
+	skb = sk->sk_send_head;
+	if (unlikely(!skb))
+		return 0;
+
+	tso_segs = tcp_init_tso_segs(sk, skb);
+	cwnd_quota = tcp_cwnd_test(tp, skb);
+	if (unlikely(!cwnd_quota))
+		goto out;
+
+	sent_pkts = 0;
+	while (likely(tcp_snd_wnd_test(tp, skb, mss_now))) {
+		BUG_ON(!tso_segs);
+
+		if (tso_segs == 1) {
+			if (unlikely(!tcp_nagle_test(tp, skb, mss_now,
+						     (tcp_skb_is_last(sk, skb) ?
+						      nonagle : TCP_NAGLE_PUSH))))
+				break;
+		} else {
+			if (tcp_tso_should_defer(sk, tp, skb))
+				break;
+		}
 
-		/* Account for SACKS, we may need to fragment due to this.
-		 * It is just like the real MSS changing on us midstream.
-		 * We also handle things correctly when the user adds some
-		 * IP options mid-stream.  Silly to do, but cover it.
-		 */
-		mss_now = tcp_current_mss(sk, 1);
-
-		while ((skb = sk->sk_send_head) &&
-		       tcp_snd_test(sk, skb, mss_now,
-			       	    tcp_skb_is_last(sk, skb) ? nonagle :
-				    			       TCP_NAGLE_PUSH)) {
-			if (skb->len > mss_now) {
-				if (tcp_fragment(sk, skb, mss_now))
+		if (tso_segs > 1) {
+			u32 limit = tcp_window_allows(tp, skb,
+						      mss_now, cwnd_quota);
+
+			if (skb->len < limit) {
+				unsigned int trim = skb->len % mss_now;
+
+				if (trim)
+					limit = skb->len - trim;
+			}
+			if (skb->len > limit) {
+				if (tso_fragment(sk, skb, limit))
 					break;
 			}
-
-			TCP_SKB_CB(skb)->when = tcp_time_stamp;
-			tcp_tso_set_push(skb);
-			if (tcp_transmit_skb(sk, skb_clone(skb, GFP_ATOMIC)))
+		} else if (unlikely(skb->len > mss_now)) {
+			if (unlikely(tcp_fragment(sk, skb,  mss_now)))
 				break;
+		}
 
-			/* Advance the send_head.  This one is sent out.
-			 * This call will increment packets_out.
-			 */
-			update_send_head(sk, tp, skb);
+		TCP_SKB_CB(skb)->when = tcp_time_stamp;
+
+		if (unlikely(tcp_transmit_skb(sk, skb_clone(skb, GFP_ATOMIC))))
+			break;
+
+		/* Advance the send_head.  This one is sent out.
+		 * This call will increment packets_out.
+		 */
+		update_send_head(sk, tp, skb);
+
+		tcp_minshall_update(tp, mss_now, skb);
+		sent_pkts++;
+
+		/* Do not optimize this to use tso_segs. If we chopped up
+		 * the packet above, tso_segs will no longer be valid.
+		 */
+		cwnd_quota -= tcp_skb_pcount(skb);
+
+		BUG_ON(cwnd_quota < 0);
+		if (!cwnd_quota)
+			break;
+
+		skb = sk->sk_send_head;
+		if (!skb)
+			break;
+		tso_segs = tcp_init_tso_segs(sk, skb);
+	}
+
+	if (likely(sent_pkts)) {
+		tcp_cwnd_validate(sk, tp);
+		return 0;
+	}
+out:
+	return !tp->packets_out && sk->sk_send_head;
+}
+
+/* Push out any pending frames which were held back due to
+ * TCP_CORK or attempt at coalescing tiny packets.
+ * The socket must be locked by the caller.
+ */
+void __tcp_push_pending_frames(struct sock *sk, struct tcp_sock *tp,
+			       unsigned int cur_mss, int nonagle)
+{
+	struct sk_buff *skb = sk->sk_send_head;
 
-			tcp_minshall_update(tp, mss_now, skb);
-			sent_pkts = 1;
+	if (skb) {
+		if (tcp_write_xmit(sk, cur_mss, nonagle))
+			tcp_check_probe_timer(sk, tp);
+	}
+}
+
+/* Send _single_ skb sitting at the send head. This function requires
+ * true push pending frames to setup probe timer etc.
+ */
+void tcp_push_one(struct sock *sk, unsigned int mss_now)
+{
+	struct tcp_sock *tp = tcp_sk(sk);
+	struct sk_buff *skb = sk->sk_send_head;
+	unsigned int tso_segs, cwnd_quota;
+
+	BUG_ON(!skb || skb->len < mss_now);
+
+	tso_segs = tcp_init_tso_segs(sk, skb);
+	cwnd_quota = tcp_snd_test(sk, skb, mss_now, TCP_NAGLE_PUSH);
+
+	if (likely(cwnd_quota)) {
+		BUG_ON(!tso_segs);
+
+		if (tso_segs > 1) {
+			u32 limit = tcp_window_allows(tp, skb,
+						      mss_now, cwnd_quota);
+
+			if (skb->len < limit) {
+				unsigned int trim = skb->len % mss_now;
+
+				if (trim)
+					limit = skb->len - trim;
+			}
+			if (skb->len > limit) {
+				if (unlikely(tso_fragment(sk, skb, limit)))
+					return;
+			}
+		} else if (unlikely(skb->len > mss_now)) {
+			if (unlikely(tcp_fragment(sk, skb, mss_now)))
+				return;
 		}
 
-		if (sent_pkts) {
+		/* Send it out now. */
+		TCP_SKB_CB(skb)->when = tcp_time_stamp;
+
+		if (likely(!tcp_transmit_skb(sk, skb_clone(skb, sk->sk_allocation)))) {
+			update_send_head(sk, tp, skb);
 			tcp_cwnd_validate(sk, tp);
-			return 0;
+			return;
 		}
-
-		return !tp->packets_out && sk->sk_send_head;
 	}
-	return 0;
 }
 
 /* This function returns the amount that we can raise the
@@ -1039,7 +1369,6 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb)
 		if (sk->sk_route_caps & NETIF_F_TSO) {
 			sk->sk_route_caps &= ~NETIF_F_TSO;
 			sock_set_flag(sk, SOCK_NO_LARGESEND);
-			tp->mss_cache = tp->mss_cache_std;
 		}
 
 		if (tcp_trim_head(sk, skb, tp->snd_una - TCP_SKB_CB(skb)->seq))
@@ -1101,7 +1430,6 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb)
 	 * is still in somebody's hands, else make a clone.
 	 */
 	TCP_SKB_CB(skb)->when = tcp_time_stamp;
-	tcp_tso_set_push(skb);
 
 	err = tcp_transmit_skb(sk, (skb_cloned(skb) ?
 				    pskb_copy(skb, GFP_ATOMIC):
@@ -1670,14 +1998,12 @@ int tcp_write_wakeup(struct sock *sk)
 				if (sk->sk_route_caps & NETIF_F_TSO) {
 					sock_set_flag(sk, SOCK_NO_LARGESEND);
 					sk->sk_route_caps &= ~NETIF_F_TSO;
-					tp->mss_cache = tp->mss_cache_std;
 				}
 			} else if (!tcp_skb_pcount(skb))
 				tcp_set_skb_tso_segs(sk, skb);
 
 			TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH;
 			TCP_SKB_CB(skb)->when = tcp_time_stamp;
-			tcp_tso_set_push(skb);
 			err = tcp_transmit_skb(sk, skb_clone(skb, GFP_ATOMIC));
 			if (!err) {
 				update_send_head(sk, tp, skb);

net/ipv6/af_inet6.c

@@ -774,7 +774,6 @@ static int __init inet6_init(void)
 	if (if6_proc_init())
 		goto proc_if6_fail;
 #endif
-	ipv6_packet_init();
 	ip6_route_init();
 	ip6_flowlabel_init();
 	err = addrconf_init();
@@ -791,6 +790,8 @@ static int __init inet6_init(void)
 	/* Init v6 transport protocols. */
 	udpv6_init();
 	tcpv6_init();
+
+	ipv6_packet_init();
 	err = 0;
 out:
 	return err;
@@ -798,7 +799,6 @@ static int __init inet6_init(void)
 addrconf_fail:
 	ip6_flowlabel_cleanup();
 	ip6_route_cleanup();
-	ipv6_packet_cleanup();
 #ifdef CONFIG_PROC_FS
 	if6_proc_exit();
 proc_if6_fail:

net/ipv6/ip6_output.c

@@ -465,7 +465,6 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from)
 	to->pkt_type = from->pkt_type;
 	to->priority = from->priority;
 	to->protocol = from->protocol;
-	to->security = from->security;
 	dst_release(to->dst);
 	to->dst = dst_clone(from->dst);
 	to->dev = from->dev;

net/ipv6/tcp_ipv6.c

@@ -2018,7 +2018,7 @@ static int tcp_v6_init_sock(struct sock *sk)
 	 */
 	tp->snd_ssthresh = 0x7fffffff;
 	tp->snd_cwnd_clamp = ~0;
-	tp->mss_cache_std = tp->mss_cache = 536;
+	tp->mss_cache = 536;
 
 	tp->reordering = sysctl_tcp_reordering;
 

net/sched/Makefile

@@ -4,7 +4,7 @@
 
 obj-y	:= sch_generic.o
 
-obj-$(CONFIG_NET_SCHED)		+= sch_api.o sch_fifo.o
+obj-$(CONFIG_NET_SCHED)		+= sch_api.o sch_fifo.o sch_blackhole.o
 obj-$(CONFIG_NET_CLS)		+= cls_api.o
 obj-$(CONFIG_NET_CLS_ACT)	+= act_api.o
 obj-$(CONFIG_NET_ACT_POLICE)	+= police.o

net/sched/em_meta.c

@@ -205,11 +205,6 @@ META_COLLECTOR(int_protocol)
 	dst->value = skb->protocol;
 }
 
-META_COLLECTOR(int_security)
-{
-	dst->value = skb->security;
-}
-
 META_COLLECTOR(int_pkttype)
 {
 	dst->value = skb->pkt_type;
@@ -524,7 +519,6 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = {
 		[META_ID(REALDEV)]		= META_FUNC(int_realdev),
 		[META_ID(PRIORITY)]		= META_FUNC(int_priority),
 		[META_ID(PROTOCOL)]		= META_FUNC(int_protocol),
-		[META_ID(SECURITY)]		= META_FUNC(int_security),
 		[META_ID(PKTTYPE)]		= META_FUNC(int_pkttype),
 		[META_ID(PKTLEN)]		= META_FUNC(int_pktlen),
 		[META_ID(DATALEN)]		= META_FUNC(int_datalen),

net/sched/sch_api.c

@@ -399,10 +399,8 @@ qdisc_create(struct net_device *dev, u32 handle, struct rtattr **tca, int *errp)
 {
 	int err;
 	struct rtattr *kind = tca[TCA_KIND-1];
-	void *p = NULL;
 	struct Qdisc *sch;
 	struct Qdisc_ops *ops;
-	int size;
 
 	ops = qdisc_lookup_ops(kind);
 #ifdef CONFIG_KMOD
@@ -437,64 +435,55 @@ qdisc_create(struct net_device *dev, u32 handle, struct rtattr **tca, int *errp)
 	if (ops == NULL)
 		goto err_out;
 
-	/* ensure that the Qdisc and the private data are 32-byte aligned */
-	size = ((sizeof(*sch) + QDISC_ALIGN_CONST) & ~QDISC_ALIGN_CONST);
-	size += ops->priv_size + QDISC_ALIGN_CONST;
-
-	p = kmalloc(size, GFP_KERNEL);
-	err = -ENOBUFS;
-	if (!p)
+	sch = qdisc_alloc(dev, ops);
+	if (IS_ERR(sch)) {
+		err = PTR_ERR(sch);
 		goto err_out2;
-	memset(p, 0, size);
-	sch = (struct Qdisc *)(((unsigned long)p + QDISC_ALIGN_CONST)
-	                       & ~QDISC_ALIGN_CONST);
-	sch->padded = (char *)sch - (char *)p;
-
-	INIT_LIST_HEAD(&sch->list);
-	skb_queue_head_init(&sch->q);
+	}
 
-	if (handle == TC_H_INGRESS)
+	if (handle == TC_H_INGRESS) {
 		sch->flags |= TCQ_F_INGRESS;
-
-	sch->ops = ops;
-	sch->enqueue = ops->enqueue;
-	sch->dequeue = ops->dequeue;
-	sch->dev = dev;
-	dev_hold(dev);
-	atomic_set(&sch->refcnt, 1);
-	sch->stats_lock = &dev->queue_lock;
-	if (handle == 0) {
+		handle = TC_H_MAKE(TC_H_INGRESS, 0);
+	} else if (handle == 0) {
 		handle = qdisc_alloc_handle(dev);
 		err = -ENOMEM;
 		if (handle == 0)
 			goto err_out3;
 	}
 
-	if (handle == TC_H_INGRESS)
-                sch->handle =TC_H_MAKE(TC_H_INGRESS, 0);
-        else
-                sch->handle = handle;
+	sch->handle = handle;
 
 	if (!ops->init || (err = ops->init(sch, tca[TCA_OPTIONS-1])) == 0) {
+#ifdef CONFIG_NET_ESTIMATOR
+		if (tca[TCA_RATE-1]) {
+			err = gen_new_estimator(&sch->bstats, &sch->rate_est,
+						sch->stats_lock,
+						tca[TCA_RATE-1]);
+			if (err) {
+				/*
+				 * Any broken qdiscs that would require
+				 * a ops->reset() here? The qdisc was never
+				 * in action so it shouldn't be necessary.
+				 */
+				if (ops->destroy)
+					ops->destroy(sch);
+				goto err_out3;
+			}
+		}
+#endif
 		qdisc_lock_tree(dev);
 		list_add_tail(&sch->list, &dev->qdisc_list);
 		qdisc_unlock_tree(dev);
 
-#ifdef CONFIG_NET_ESTIMATOR
-		if (tca[TCA_RATE-1])
-			gen_new_estimator(&sch->bstats, &sch->rate_est,
-				sch->stats_lock, tca[TCA_RATE-1]);
-#endif
 		return sch;
 	}
 err_out3:
 	dev_put(dev);
+	kfree((char *) sch - sch->padded);
 err_out2:
 	module_put(ops->owner);
 err_out:
 	*errp = err;
-	if (p)
-		kfree(p);
 	return NULL;
 }
 

net/sched/sch_blackhole.c

+/*
+ * net/sched/sch_blackhole.c	Black hole queue
+ *
+ *		This program is free software; you can redistribute it and/or
+ *		modify it under the terms of the GNU General Public License
+ *		as published by the Free Software Foundation; either version
+ *		2 of the License, or (at your option) any later version.
+ *
+ * Authors:	Thomas Graf <tgraf@suug.ch>
+ *
+ * Note: Quantum tunneling is not supported.
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/netdevice.h>
+#include <linux/skbuff.h>
+#include <net/pkt_sched.h>
+
+static int blackhole_enqueue(struct sk_buff *skb, struct Qdisc *sch)
+{
+	qdisc_drop(skb, sch);
+	return NET_XMIT_SUCCESS;
+}
+
+static struct sk_buff *blackhole_dequeue(struct Qdisc *sch)
+{
+	return NULL;
+}
+
+static struct Qdisc_ops blackhole_qdisc_ops = {
+	.id		= "blackhole",
+	.priv_size	= 0,
+	.enqueue	= blackhole_enqueue,
+	.dequeue	= blackhole_dequeue,
+	.owner		= THIS_MODULE,
+};
+
+static int __init blackhole_module_init(void)
+{
+	return register_qdisc(&blackhole_qdisc_ops);
+}
+
+static void __exit blackhole_module_exit(void)
+{
+	unregister_qdisc(&blackhole_qdisc_ops);
+}
+
+module_init(blackhole_module_init)
+module_exit(blackhole_module_exit)
+
+MODULE_LICENSE("GPL");

net/sched/sch_generic.c

@@ -395,24 +395,23 @@ static struct Qdisc_ops pfifo_fast_ops = {
 	.owner		=	THIS_MODULE,
 };
 
-struct Qdisc * qdisc_create_dflt(struct net_device *dev, struct Qdisc_ops *ops)
+struct Qdisc *qdisc_alloc(struct net_device *dev, struct Qdisc_ops *ops)
 {
 	void *p;
 	struct Qdisc *sch;
-	int size;
+	unsigned int size;
+	int err = -ENOBUFS;
 
 	/* ensure that the Qdisc and the private data are 32-byte aligned */
-	size = ((sizeof(*sch) + QDISC_ALIGN_CONST) & ~QDISC_ALIGN_CONST);
-	size += ops->priv_size + QDISC_ALIGN_CONST;
+	size = QDISC_ALIGN(sizeof(*sch));
+	size += ops->priv_size + (QDISC_ALIGNTO - 1);
 
 	p = kmalloc(size, GFP_KERNEL);
 	if (!p)
-		return NULL;
+		goto errout;
 	memset(p, 0, size);
-
-	sch = (struct Qdisc *)(((unsigned long)p + QDISC_ALIGN_CONST) 
-			       & ~QDISC_ALIGN_CONST);
-	sch->padded = (char *)sch - (char *)p;
+	sch = (struct Qdisc *) QDISC_ALIGN((unsigned long) p);
+	sch->padded = (char *) sch - (char *) p;
 
 	INIT_LIST_HEAD(&sch->list);
 	skb_queue_head_init(&sch->q);
@@ -423,11 +422,24 @@ struct Qdisc * qdisc_create_dflt(struct net_device *dev, struct Qdisc_ops *ops)
 	dev_hold(dev);
 	sch->stats_lock = &dev->queue_lock;
 	atomic_set(&sch->refcnt, 1);
+
+	return sch;
+errout:
+	return ERR_PTR(-err);
+}
+
+struct Qdisc * qdisc_create_dflt(struct net_device *dev, struct Qdisc_ops *ops)
+{
+	struct Qdisc *sch;
+	
+	sch = qdisc_alloc(dev, ops);
+	if (IS_ERR(sch))
+		goto errout;
+
 	if (!ops->init || ops->init(sch, NULL) == 0)
 		return sch;
 
-	dev_put(dev);
-	kfree(p);
+errout:
 	return NULL;
 }
 
@@ -591,6 +603,7 @@ EXPORT_SYMBOL(__netdev_watchdog_up);
 EXPORT_SYMBOL(noop_qdisc);
 EXPORT_SYMBOL(noop_qdisc_ops);
 EXPORT_SYMBOL(qdisc_create_dflt);
+EXPORT_SYMBOL(qdisc_alloc);
 EXPORT_SYMBOL(qdisc_destroy);
 EXPORT_SYMBOL(qdisc_reset);
 EXPORT_SYMBOL(qdisc_restart);