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

* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: (109 commits)
  [ETHTOOL]: Fix UFO typo
  [SCTP]: Fix persistent slowdown in sctp when a gap ack consumes rx buffer.
  [SCTP]: Send only 1 window update SACK per message.
  [SCTP]: Don't do CRC32C checksum over loopback.
  [SCTP] Reset rtt_in_progress for the chunk when processing its sack.
  [SCTP]: Reject sctp packets with broadcast addresses.
  [SCTP]: Limit association max_retrans setting in setsockopt.
  [PFKEYV2]: Fix inconsistent typing in struct sadb_x_kmprivate.
  [IPV6]: Sum real space for RTAs.
  [IRDA]: Use put_unaligned() in irlmp_do_discovery().
  [BRIDGE]: Add support for NETIF_F_HW_CSUM devices
  [NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM
  [TG3]: Convert to non-LLTX
  [TG3]: Remove unnecessary tx_lock
  [TCP]: Add tcp_slow_start_after_idle sysctl.
  [BNX2]: Update version and reldate
  [BNX2]: Use CPU native page size
  [BNX2]: Use compressed firmware
  [BNX2]: Add firmware decompression
  [BNX2]: Allow WoL settings on new 5708 chips
  ...

Manual fixup for conflict in drivers/net/tulip/winbond-840.c

Documentation/kernel-parameters.txt

@@ -1402,6 +1402,15 @@ running once the system is up.
 			If enabled at boot time, /selinux/disable can be used
 			later to disable prior to initial policy load.
 
+	selinux_compat_net =
+			[SELINUX] Set initial selinux_compat_net flag value.
+                        Format: { "0" | "1" }
+                        0 -- use new secmark-based packet controls
+                        1 -- use legacy packet controls
+                        Default value is 0 (preferred).
+                        Value can be changed at runtime via
+                        /selinux/compat_net.
+
 	serialnumber	[BUGS=IA-32]
 
 	sg_def_reserved_size=	[SCSI]

Documentation/networking/ip-sysctl.txt

@@ -362,6 +362,13 @@ tcp_workaround_signed_windows - BOOLEAN
 	not receive a window scaling option from them.
 	Default: 0
 
+tcp_slow_start_after_idle - BOOLEAN
+	If set, provide RFC2861 behavior and time out the congestion
+	window after an idle period.  An idle period is defined at
+	the current RTO.  If unset, the congestion window will not
+	be timed out after an idle period.
+	Default: 1
+
 IP Variables:
 
 ip_local_port_range - 2 INTEGERS

Documentation/networking/netdevices.txt

@@ -42,9 +42,9 @@ dev->get_stats:
 	Context: nominally process, but don't sleep inside an rwlock
 
 dev->hard_start_xmit:
-	Synchronization: dev->xmit_lock spinlock.
+	Synchronization: netif_tx_lock spinlock.
 	When the driver sets NETIF_F_LLTX in dev->features this will be
-	called without holding xmit_lock. In this case the driver 
+	called without holding netif_tx_lock. In this case the driver
 	has to lock by itself when needed. It is recommended to use a try lock
 	for this and return -1 when the spin lock fails. 
 	The locking there should also properly protect against 
@@ -62,12 +62,12 @@ dev->hard_start_xmit:
 	  Only valid when NETIF_F_LLTX is set.
 
 dev->tx_timeout:
-	Synchronization: dev->xmit_lock spinlock.
+	Synchronization: netif_tx_lock spinlock.
 	Context: BHs disabled
 	Notes: netif_queue_stopped() is guaranteed true
 
 dev->set_multicast_list:
-	Synchronization: dev->xmit_lock spinlock.
+	Synchronization: netif_tx_lock spinlock.
 	Context: BHs disabled
 
 dev->poll:

drivers/Kconfig

@@ -72,4 +72,6 @@ source "drivers/edac/Kconfig"
 
 source "drivers/rtc/Kconfig"
 
+source "drivers/dma/Kconfig"
+
 endmenu

drivers/Makefile

@@ -74,3 +74,4 @@ obj-$(CONFIG_SGI_SN)		+= sn/
 obj-y				+= firmware/
 obj-$(CONFIG_CRYPTO)		+= crypto/
 obj-$(CONFIG_SUPERH)		+= sh/
+obj-$(CONFIG_DMA_ENGINE)	+= dma/

drivers/block/aoe/aoenet.c

@@ -116,8 +116,7 @@ aoenet_rcv(struct sk_buff *skb, struct net_device *ifp, struct packet_type *pt,
 	skb = skb_share_check(skb, GFP_ATOMIC);
 	if (skb == NULL)
 		return 0;
-	if (skb_is_nonlinear(skb))
+	if (skb_linearize(skb))
-	if (skb_linearize(skb, GFP_ATOMIC) < 0)
 		goto exit;
 	if (!is_aoe_netif(ifp))
 		goto exit;

drivers/connector/cn_queue.c

@@ -127,7 +127,7 @@ void cn_queue_del_callback(struct cn_queue_dev *dev, struct cb_id *id)
 
 	if (found) {
 		cn_queue_free_callback(cbq);
-		atomic_dec_and_test(&dev->refcnt);
+		atomic_dec(&dev->refcnt);
 	}
 }
 

drivers/dma/Kconfig

@@ -0,0 +1,34 @@
+#
+# DMA engine configuration
+#
+
+menu "DMA Engine support"
+
+config DMA_ENGINE
+	bool "Support for DMA engines"
+	---help---
+	  DMA engines offload copy operations from the CPU to dedicated
+	  hardware, allowing the copies to happen asynchronously.
+
+comment "DMA Clients"
+
+config NET_DMA
+	bool "Network: TCP receive copy offload"
+	depends on DMA_ENGINE && NET
+	default y
+	---help---
+	  This enables the use of DMA engines in the network stack to
+	  offload receive copy-to-user operations, freeing CPU cycles.
+	  Since this is the main user of the DMA engine, it should be enabled;
+	  say Y here.
+
+comment "DMA Devices"
+
+config INTEL_IOATDMA
+	tristate "Intel I/OAT DMA support"
+	depends on DMA_ENGINE && PCI
+	default m
+	---help---
+	  Enable support for the Intel(R) I/OAT DMA engine.
+
+endmenu

drivers/dma/Makefile

@@ -0,0 +1,3 @@
+obj-$(CONFIG_DMA_ENGINE) += dmaengine.o
+obj-$(CONFIG_NET_DMA) += iovlock.o
+obj-$(CONFIG_INTEL_IOATDMA) += ioatdma.o

drivers/dma/dmaengine.c

@@ -0,0 +1,408 @@
+/*
+ * Copyright(c) 2004 - 2006 Intel Corporation. All rights reserved.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called COPYING.
+ */
+
+/*
+ * This code implements the DMA subsystem. It provides a HW-neutral interface
+ * for other kernel code to use asynchronous memory copy capabilities,
+ * if present, and allows different HW DMA drivers to register as providing
+ * this capability.
+ *
+ * Due to the fact we are accelerating what is already a relatively fast
+ * operation, the code goes to great lengths to avoid additional overhead,
+ * such as locking.
+ *
+ * LOCKING:
+ *
+ * The subsystem keeps two global lists, dma_device_list and dma_client_list.
+ * Both of these are protected by a mutex, dma_list_mutex.
+ *
+ * Each device has a channels list, which runs unlocked but is never modified
+ * once the device is registered, it's just setup by the driver.
+ *
+ * Each client has a channels list, it's only modified under the client->lock
+ * and in an RCU callback, so it's safe to read under rcu_read_lock().
+ *
+ * Each device has a kref, which is initialized to 1 when the device is
+ * registered. A kref_put is done for each class_device registered.  When the
+ * class_device is released, the coresponding kref_put is done in the release
+ * method. Every time one of the device's channels is allocated to a client,
+ * a kref_get occurs.  When the channel is freed, the coresponding kref_put
+ * happens. The device's release function does a completion, so
+ * unregister_device does a remove event, class_device_unregister, a kref_put
+ * for the first reference, then waits on the completion for all other
+ * references to finish.
+ *
+ * Each channel has an open-coded implementation of Rusty Russell's "bigref,"
+ * with a kref and a per_cpu local_t.  A single reference is set when on an
+ * ADDED event, and removed with a REMOVE event.  Net DMA client takes an
+ * extra reference per outstanding transaction.  The relase function does a
+ * kref_put on the device. -ChrisL
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/dmaengine.h>
+#include <linux/hardirq.h>
+#include <linux/spinlock.h>
+#include <linux/percpu.h>
+#include <linux/rcupdate.h>
+#include <linux/mutex.h>
+
+static DEFINE_MUTEX(dma_list_mutex);
+static LIST_HEAD(dma_device_list);
+static LIST_HEAD(dma_client_list);
+
+/* --- sysfs implementation --- */
+
+static ssize_t show_memcpy_count(struct class_device *cd, char *buf)
+{
+	struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev);
+	unsigned long count = 0;
+	int i;
+
+	for_each_possible_cpu(i)
+		count += per_cpu_ptr(chan->local, i)->memcpy_count;
+
+	return sprintf(buf, "%lu\n", count);
+}
+
+static ssize_t show_bytes_transferred(struct class_device *cd, char *buf)
+{
+	struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev);
+	unsigned long count = 0;
+	int i;
+
+	for_each_possible_cpu(i)
+		count += per_cpu_ptr(chan->local, i)->bytes_transferred;
+
+	return sprintf(buf, "%lu\n", count);
+}
+
+static ssize_t show_in_use(struct class_device *cd, char *buf)
+{
+	struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev);
+
+	return sprintf(buf, "%d\n", (chan->client ? 1 : 0));
+}
+
+static struct class_device_attribute dma_class_attrs[] = {
+	__ATTR(memcpy_count, S_IRUGO, show_memcpy_count, NULL),
+	__ATTR(bytes_transferred, S_IRUGO, show_bytes_transferred, NULL),
+	__ATTR(in_use, S_IRUGO, show_in_use, NULL),
+	__ATTR_NULL
+};
+
+static void dma_async_device_cleanup(struct kref *kref);
+
+static void dma_class_dev_release(struct class_device *cd)
+{
+	struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev);
+	kref_put(&chan->device->refcount, dma_async_device_cleanup);
+}
+
+static struct class dma_devclass = {
+	.name            = "dma",
+	.class_dev_attrs = dma_class_attrs,
+	.release = dma_class_dev_release,
+};
+
+/* --- client and device registration --- */
+
+/**
+ * dma_client_chan_alloc - try to allocate a channel to a client
+ * @client: &dma_client
+ *
+ * Called with dma_list_mutex held.
+ */
+static struct dma_chan *dma_client_chan_alloc(struct dma_client *client)
+{
+	struct dma_device *device;
+	struct dma_chan *chan;
+	unsigned long flags;
+	int desc;	/* allocated descriptor count */
+
+	/* Find a channel, any DMA engine will do */
+	list_for_each_entry(device, &dma_device_list, global_node) {
+		list_for_each_entry(chan, &device->channels, device_node) {
+			if (chan->client)
+				continue;
+
+			desc = chan->device->device_alloc_chan_resources(chan);
+			if (desc >= 0) {
+				kref_get(&device->refcount);
+				kref_init(&chan->refcount);
+				chan->slow_ref = 0;
+				INIT_RCU_HEAD(&chan->rcu);
+				chan->client = client;
+				spin_lock_irqsave(&client->lock, flags);
+				list_add_tail_rcu(&chan->client_node,
+				                  &client->channels);
+				spin_unlock_irqrestore(&client->lock, flags);
+				return chan;
+			}
+		}
+	}
+
+	return NULL;
+}
+
+/**
+ * dma_client_chan_free - release a DMA channel
+ * @chan: &dma_chan
+ */
+void dma_chan_cleanup(struct kref *kref)
+{
+	struct dma_chan *chan = container_of(kref, struct dma_chan, refcount);
+	chan->device->device_free_chan_resources(chan);
+	chan->client = NULL;
+	kref_put(&chan->device->refcount, dma_async_device_cleanup);
+}
+
+static void dma_chan_free_rcu(struct rcu_head *rcu)
+{
+	struct dma_chan *chan = container_of(rcu, struct dma_chan, rcu);
+	int bias = 0x7FFFFFFF;
+	int i;
+	for_each_possible_cpu(i)
+		bias -= local_read(&per_cpu_ptr(chan->local, i)->refcount);
+	atomic_sub(bias, &chan->refcount.refcount);
+	kref_put(&chan->refcount, dma_chan_cleanup);
+}
+
+static void dma_client_chan_free(struct dma_chan *chan)
+{
+	atomic_add(0x7FFFFFFF, &chan->refcount.refcount);
+	chan->slow_ref = 1;
+	call_rcu(&chan->rcu, dma_chan_free_rcu);
+}
+
+/**
+ * dma_chans_rebalance - reallocate channels to clients
+ *
+ * When the number of DMA channel in the system changes,
+ * channels need to be rebalanced among clients
+ */
+static void dma_chans_rebalance(void)
+{
+	struct dma_client *client;
+	struct dma_chan *chan;
+	unsigned long flags;
+
+	mutex_lock(&dma_list_mutex);
+
+	list_for_each_entry(client, &dma_client_list, global_node) {
+		while (client->chans_desired > client->chan_count) {
+			chan = dma_client_chan_alloc(client);
+			if (!chan)
+				break;
+			client->chan_count++;
+			client->event_callback(client,
+	                                       chan,
+	                                       DMA_RESOURCE_ADDED);
+		}
+		while (client->chans_desired < client->chan_count) {
+			spin_lock_irqsave(&client->lock, flags);
+			chan = list_entry(client->channels.next,
+			                  struct dma_chan,
+			                  client_node);
+			list_del_rcu(&chan->client_node);
+			spin_unlock_irqrestore(&client->lock, flags);
+			client->chan_count--;
+			client->event_callback(client,
+			                       chan,
+			                       DMA_RESOURCE_REMOVED);
+			dma_client_chan_free(chan);
+		}
+	}
+
+	mutex_unlock(&dma_list_mutex);
+}
+
+/**
+ * dma_async_client_register - allocate and register a &dma_client
+ * @event_callback: callback for notification of channel addition/removal
+ */
+struct dma_client *dma_async_client_register(dma_event_callback event_callback)
+{
+	struct dma_client *client;
+
+	client = kzalloc(sizeof(*client), GFP_KERNEL);
+	if (!client)
+		return NULL;
+
+	INIT_LIST_HEAD(&client->channels);
+	spin_lock_init(&client->lock);
+	client->chans_desired = 0;
+	client->chan_count = 0;
+	client->event_callback = event_callback;
+
+	mutex_lock(&dma_list_mutex);
+	list_add_tail(&client->global_node, &dma_client_list);
+	mutex_unlock(&dma_list_mutex);
+
+	return client;
+}
+
+/**
+ * dma_async_client_unregister - unregister a client and free the &dma_client
+ * @client:
+ *
+ * Force frees any allocated DMA channels, frees the &dma_client memory
+ */
+void dma_async_client_unregister(struct dma_client *client)
+{
+	struct dma_chan *chan;
+
+	if (!client)
+		return;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(chan, &client->channels, client_node)
+		dma_client_chan_free(chan);
+	rcu_read_unlock();
+
+	mutex_lock(&dma_list_mutex);
+	list_del(&client->global_node);
+	mutex_unlock(&dma_list_mutex);
+
+	kfree(client);
+	dma_chans_rebalance();
+}
+
+/**
+ * dma_async_client_chan_request - request DMA channels
+ * @client: &dma_client
+ * @number: count of DMA channels requested
+ *
+ * Clients call dma_async_client_chan_request() to specify how many
+ * DMA channels they need, 0 to free all currently allocated.
+ * The resulting allocations/frees are indicated to the client via the
+ * event callback.
+ */
+void dma_async_client_chan_request(struct dma_client *client,
+			unsigned int number)
+{
+	client->chans_desired = number;
+	dma_chans_rebalance();
+}
+
+/**
+ * dma_async_device_register -
+ * @device: &dma_device
+ */
+int dma_async_device_register(struct dma_device *device)
+{
+	static int id;
+	int chancnt = 0;
+	struct dma_chan* chan;
+
+	if (!device)
+		return -ENODEV;
+
+	init_completion(&device->done);
+	kref_init(&device->refcount);
+	device->dev_id = id++;
+
+	/* represent channels in sysfs. Probably want devs too */
+	list_for_each_entry(chan, &device->channels, device_node) {
+		chan->local = alloc_percpu(typeof(*chan->local));
+		if (chan->local == NULL)
+			continue;
+
+		chan->chan_id = chancnt++;
+		chan->class_dev.class = &dma_devclass;
+		chan->class_dev.dev = NULL;
+		snprintf(chan->class_dev.class_id, BUS_ID_SIZE, "dma%dchan%d",
+		         device->dev_id, chan->chan_id);
+
+		kref_get(&device->refcount);
+		class_device_register(&chan->class_dev);
+	}
+
+	mutex_lock(&dma_list_mutex);
+	list_add_tail(&device->global_node, &dma_device_list);
+	mutex_unlock(&dma_list_mutex);
+
+	dma_chans_rebalance();
+
+	return 0;
+}
+
+/**
+ * dma_async_device_unregister -
+ * @device: &dma_device
+ */
+static void dma_async_device_cleanup(struct kref *kref)
+{
+	struct dma_device *device;
+
+	device = container_of(kref, struct dma_device, refcount);
+	complete(&device->done);
+}
+
+void dma_async_device_unregister(struct dma_device* device)
+{
+	struct dma_chan *chan;
+	unsigned long flags;
+
+	mutex_lock(&dma_list_mutex);
+	list_del(&device->global_node);
+	mutex_unlock(&dma_list_mutex);
+
+	list_for_each_entry(chan, &device->channels, device_node) {
+		if (chan->client) {
+			spin_lock_irqsave(&chan->client->lock, flags);
+			list_del(&chan->client_node);
+			chan->client->chan_count--;
+			spin_unlock_irqrestore(&chan->client->lock, flags);
+			chan->client->event_callback(chan->client,
+			                             chan,
+			                             DMA_RESOURCE_REMOVED);
+			dma_client_chan_free(chan);
+		}
+		class_device_unregister(&chan->class_dev);
+	}
+	dma_chans_rebalance();
+
+	kref_put(&device->refcount, dma_async_device_cleanup);
+	wait_for_completion(&device->done);
+}
+
+static int __init dma_bus_init(void)
+{
+	mutex_init(&dma_list_mutex);
+	return class_register(&dma_devclass);
+}
+
+subsys_initcall(dma_bus_init);
+
+EXPORT_SYMBOL(dma_async_client_register);
+EXPORT_SYMBOL(dma_async_client_unregister);
+EXPORT_SYMBOL(dma_async_client_chan_request);
+EXPORT_SYMBOL(dma_async_memcpy_buf_to_buf);
+EXPORT_SYMBOL(dma_async_memcpy_buf_to_pg);
+EXPORT_SYMBOL(dma_async_memcpy_pg_to_pg);
+EXPORT_SYMBOL(dma_async_memcpy_complete);
+EXPORT_SYMBOL(dma_async_memcpy_issue_pending);
+EXPORT_SYMBOL(dma_async_device_register);
+EXPORT_SYMBOL(dma_async_device_unregister);
+EXPORT_SYMBOL(dma_chan_cleanup);

drivers/dma/ioatdma.c

@@ -0,0 +1,840 @@
+/*
+ * Copyright(c) 2004 - 2006 Intel Corporation. All rights reserved.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called COPYING.
+ */
+
+/*
+ * This driver supports an Intel I/OAT DMA engine, which does asynchronous
+ * copy operations.
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/interrupt.h>
+#include <linux/dmaengine.h>
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include "ioatdma.h"
+#include "ioatdma_io.h"
+#include "ioatdma_registers.h"
+#include "ioatdma_hw.h"
+
+#define to_ioat_chan(chan) container_of(chan, struct ioat_dma_chan, common)
+#define to_ioat_device(dev) container_of(dev, struct ioat_device, common)
+#define to_ioat_desc(lh) container_of(lh, struct ioat_desc_sw, node)
+
+/* internal functions */
+static int __devinit ioat_probe(struct pci_dev *pdev, const struct pci_device_id *ent);
+static void __devexit ioat_remove(struct pci_dev *pdev);
+
+static int enumerate_dma_channels(struct ioat_device *device)
+{
+	u8 xfercap_scale;
+	u32 xfercap;
+	int i;
+	struct ioat_dma_chan *ioat_chan;
+
+	device->common.chancnt = ioatdma_read8(device, IOAT_CHANCNT_OFFSET);
+	xfercap_scale = ioatdma_read8(device, IOAT_XFERCAP_OFFSET);
+	xfercap = (xfercap_scale == 0 ? -1 : (1UL << xfercap_scale));
+
+	for (i = 0; i < device->common.chancnt; i++) {
+		ioat_chan = kzalloc(sizeof(*ioat_chan), GFP_KERNEL);
+		if (!ioat_chan) {
+			device->common.chancnt = i;
+			break;
+		}
+
+		ioat_chan->device = device;
+		ioat_chan->reg_base = device->reg_base + (0x80 * (i + 1));
+		ioat_chan->xfercap = xfercap;
+		spin_lock_init(&ioat_chan->cleanup_lock);
+		spin_lock_init(&ioat_chan->desc_lock);
+		INIT_LIST_HEAD(&ioat_chan->free_desc);
+		INIT_LIST_HEAD(&ioat_chan->used_desc);
+		/* This should be made common somewhere in dmaengine.c */
+		ioat_chan->common.device = &device->common;
+		ioat_chan->common.client = NULL;
+		list_add_tail(&ioat_chan->common.device_node,
+		              &device->common.channels);
+	}
+	return device->common.chancnt;
+}
+
+static struct ioat_desc_sw *ioat_dma_alloc_descriptor(
+	struct ioat_dma_chan *ioat_chan,
+	int flags)
+{
+	struct ioat_dma_descriptor *desc;
+	struct ioat_desc_sw *desc_sw;
+	struct ioat_device *ioat_device;
+	dma_addr_t phys;
+
+	ioat_device = to_ioat_device(ioat_chan->common.device);
+	desc = pci_pool_alloc(ioat_device->dma_pool, flags, &phys);
+	if (unlikely(!desc))
+		return NULL;
+
+	desc_sw = kzalloc(sizeof(*desc_sw), flags);
+	if (unlikely(!desc_sw)) {
+		pci_pool_free(ioat_device->dma_pool, desc, phys);
+		return NULL;
+	}
+
+	memset(desc, 0, sizeof(*desc));
+	desc_sw->hw = desc;
+	desc_sw->phys = phys;
+
+	return desc_sw;
+}
+
+#define INITIAL_IOAT_DESC_COUNT 128
+
+static void ioat_start_null_desc(struct ioat_dma_chan *ioat_chan);
+
+/* returns the actual number of allocated descriptors */
+static int ioat_dma_alloc_chan_resources(struct dma_chan *chan)
+{
+	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
+	struct ioat_desc_sw *desc = NULL;
+	u16 chanctrl;
+	u32 chanerr;
+	int i;
+	LIST_HEAD(tmp_list);
+
+	/*
+	 * In-use bit automatically set by reading chanctrl
+	 * If 0, we got it, if 1, someone else did
+	 */
+	chanctrl = ioatdma_chan_read16(ioat_chan, IOAT_CHANCTRL_OFFSET);
+	if (chanctrl & IOAT_CHANCTRL_CHANNEL_IN_USE)
+		return -EBUSY;
+
+        /* Setup register to interrupt and write completion status on error */
+	chanctrl = IOAT_CHANCTRL_CHANNEL_IN_USE |
+		IOAT_CHANCTRL_ERR_INT_EN |
+		IOAT_CHANCTRL_ANY_ERR_ABORT_EN |
+		IOAT_CHANCTRL_ERR_COMPLETION_EN;
+        ioatdma_chan_write16(ioat_chan, IOAT_CHANCTRL_OFFSET, chanctrl);
+
+	chanerr = ioatdma_chan_read32(ioat_chan, IOAT_CHANERR_OFFSET);
+	if (chanerr) {
+		printk("IOAT: CHANERR = %x, clearing\n", chanerr);
+		ioatdma_chan_write32(ioat_chan, IOAT_CHANERR_OFFSET, chanerr);
+	}
+
+	/* Allocate descriptors */
+	for (i = 0; i < INITIAL_IOAT_DESC_COUNT; i++) {
+		desc = ioat_dma_alloc_descriptor(ioat_chan, GFP_KERNEL);
+		if (!desc) {
+			printk(KERN_ERR "IOAT: Only %d initial descriptors\n", i);
+			break;
+		}
+		list_add_tail(&desc->node, &tmp_list);
+	}
+	spin_lock_bh(&ioat_chan->desc_lock);
+	list_splice(&tmp_list, &ioat_chan->free_desc);
+	spin_unlock_bh(&ioat_chan->desc_lock);
+
+	/* allocate a completion writeback area */
+	/* doing 2 32bit writes to mmio since 1 64b write doesn't work */
+	ioat_chan->completion_virt =
+		pci_pool_alloc(ioat_chan->device->completion_pool,
+		               GFP_KERNEL,
+		               &ioat_chan->completion_addr);
+	memset(ioat_chan->completion_virt, 0,
+	       sizeof(*ioat_chan->completion_virt));
+	ioatdma_chan_write32(ioat_chan, IOAT_CHANCMP_OFFSET_LOW,
+	               ((u64) ioat_chan->completion_addr) & 0x00000000FFFFFFFF);
+	ioatdma_chan_write32(ioat_chan, IOAT_CHANCMP_OFFSET_HIGH,
+	               ((u64) ioat_chan->completion_addr) >> 32);
+
+	ioat_start_null_desc(ioat_chan);
+	return i;
+}
+
+static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *ioat_chan);
+
+static void ioat_dma_free_chan_resources(struct dma_chan *chan)
+{
+	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
+	struct ioat_device *ioat_device = to_ioat_device(chan->device);
+	struct ioat_desc_sw *desc, *_desc;
+	u16 chanctrl;
+	int in_use_descs = 0;
+
+	ioat_dma_memcpy_cleanup(ioat_chan);
+
+	ioatdma_chan_write8(ioat_chan, IOAT_CHANCMD_OFFSET, IOAT_CHANCMD_RESET);
+
+	spin_lock_bh(&ioat_chan->desc_lock);
+	list_for_each_entry_safe(desc, _desc, &ioat_chan->used_desc, node) {
+		in_use_descs++;
+		list_del(&desc->node);
+		pci_pool_free(ioat_device->dma_pool, desc->hw, desc->phys);
+		kfree(desc);
+	}
+	list_for_each_entry_safe(desc, _desc, &ioat_chan->free_desc, node) {
+		list_del(&desc->node);
+		pci_pool_free(ioat_device->dma_pool, desc->hw, desc->phys);
+		kfree(desc);
+	}
+	spin_unlock_bh(&ioat_chan->desc_lock);
+
+	pci_pool_free(ioat_device->completion_pool,
+	              ioat_chan->completion_virt,
+	              ioat_chan->completion_addr);
+
+	/* one is ok since we left it on there on purpose */
+	if (in_use_descs > 1)
+		printk(KERN_ERR "IOAT: Freeing %d in use descriptors!\n",
+			in_use_descs - 1);
+
+	ioat_chan->last_completion = ioat_chan->completion_addr = 0;
+
+	/* Tell hw the chan is free */
+	chanctrl = ioatdma_chan_read16(ioat_chan, IOAT_CHANCTRL_OFFSET);
+	chanctrl &= ~IOAT_CHANCTRL_CHANNEL_IN_USE;
+	ioatdma_chan_write16(ioat_chan, IOAT_CHANCTRL_OFFSET, chanctrl);
+}
+
+/**
+ * do_ioat_dma_memcpy - actual function that initiates a IOAT DMA transaction
+ * @chan: IOAT DMA channel handle
+ * @dest: DMA destination address
+ * @src: DMA source address
+ * @len: transaction length in bytes
+ */
+
+static dma_cookie_t do_ioat_dma_memcpy(struct ioat_dma_chan *ioat_chan,
+                                       dma_addr_t dest,
+                                       dma_addr_t src,
+                                       size_t len)
+{
+	struct ioat_desc_sw *first;
+	struct ioat_desc_sw *prev;
+	struct ioat_desc_sw *new;
+	dma_cookie_t cookie;
+	LIST_HEAD(new_chain);
+	u32 copy;
+	size_t orig_len;
+	dma_addr_t orig_src, orig_dst;
+	unsigned int desc_count = 0;
+	unsigned int append = 0;
+
+	if (!ioat_chan || !dest || !src)
+		return -EFAULT;
+
+	if (!len)
+		return ioat_chan->common.cookie;
+
+	orig_len = len;
+	orig_src = src;
+	orig_dst = dest;
+
+	first = NULL;
+	prev = NULL;
+
+	spin_lock_bh(&ioat_chan->desc_lock);
+
+	while (len) {
+		if (!list_empty(&ioat_chan->free_desc)) {
+			new = to_ioat_desc(ioat_chan->free_desc.next);
+			list_del(&new->node);
+		} else {
+			/* try to get another desc */
+			new = ioat_dma_alloc_descriptor(ioat_chan, GFP_ATOMIC);
+			/* will this ever happen? */
+			/* TODO add upper limit on these */
+			BUG_ON(!new);
+		}
+
+		copy = min((u32) len, ioat_chan->xfercap);
+
+		new->hw->size = copy;
+		new->hw->ctl = 0;
+		new->hw->src_addr = src;
+		new->hw->dst_addr = dest;
+		new->cookie = 0;
+
+		/* chain together the physical address list for the HW */
+		if (!first)
+			first = new;
+		else
+			prev->hw->next = (u64) new->phys;
+
+		prev = new;
+
+		len  -= copy;
+		dest += copy;
+		src  += copy;
+
+		list_add_tail(&new->node, &new_chain);
+		desc_count++;
+	}
+	new->hw->ctl = IOAT_DMA_DESCRIPTOR_CTL_CP_STS;
+	new->hw->next = 0;
+
+	/* cookie incr and addition to used_list must be atomic */
+
+	cookie = ioat_chan->common.cookie;
+	cookie++;
+	if (cookie < 0)
+		cookie = 1;
+	ioat_chan->common.cookie = new->cookie = cookie;
+
+	pci_unmap_addr_set(new, src, orig_src);
+	pci_unmap_addr_set(new, dst, orig_dst);
+	pci_unmap_len_set(new, src_len, orig_len);
+	pci_unmap_len_set(new, dst_len, orig_len);
+
+	/* write address into NextDescriptor field of last desc in chain */
+	to_ioat_desc(ioat_chan->used_desc.prev)->hw->next = first->phys;
+	list_splice_init(&new_chain, ioat_chan->used_desc.prev);
+
+	ioat_chan->pending += desc_count;
+	if (ioat_chan->pending >= 20) {
+		append = 1;
+		ioat_chan->pending = 0;
+	}
+
+	spin_unlock_bh(&ioat_chan->desc_lock);
+
+	if (append)
+		ioatdma_chan_write8(ioat_chan,
+		                    IOAT_CHANCMD_OFFSET,
+		                    IOAT_CHANCMD_APPEND);
+	return cookie;
+}
+
+/**
+ * ioat_dma_memcpy_buf_to_buf - wrapper that takes src & dest bufs
+ * @chan: IOAT DMA channel handle
+ * @dest: DMA destination address
+ * @src: DMA source address
+ * @len: transaction length in bytes
+ */
+
+static dma_cookie_t ioat_dma_memcpy_buf_to_buf(struct dma_chan *chan,
+                                               void *dest,
+                                               void *src,
+                                               size_t len)
+{
+	dma_addr_t dest_addr;
+	dma_addr_t src_addr;
+	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
+
+	dest_addr = pci_map_single(ioat_chan->device->pdev,
+		dest, len, PCI_DMA_FROMDEVICE);
+	src_addr = pci_map_single(ioat_chan->device->pdev,
+		src, len, PCI_DMA_TODEVICE);
+
+	return do_ioat_dma_memcpy(ioat_chan, dest_addr, src_addr, len);
+}
+
+/**
+ * ioat_dma_memcpy_buf_to_pg - wrapper, copying from a buf to a page
+ * @chan: IOAT DMA channel handle
+ * @page: pointer to the page to copy to
+ * @offset: offset into that page
+ * @src: DMA source address
+ * @len: transaction length in bytes
+ */
+
+static dma_cookie_t ioat_dma_memcpy_buf_to_pg(struct dma_chan *chan,
+                                              struct page *page,
+                                              unsigned int offset,
+                                              void *src,
+                                              size_t len)
+{
+	dma_addr_t dest_addr;
+	dma_addr_t src_addr;
+	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
+
+	dest_addr = pci_map_page(ioat_chan->device->pdev,
+		page, offset, len, PCI_DMA_FROMDEVICE);
+	src_addr = pci_map_single(ioat_chan->device->pdev,
+		src, len, PCI_DMA_TODEVICE);
+
+	return do_ioat_dma_memcpy(ioat_chan, dest_addr, src_addr, len);
+}
+
+/**
+ * ioat_dma_memcpy_pg_to_pg - wrapper, copying between two pages
+ * @chan: IOAT DMA channel handle
+ * @dest_pg: pointer to the page to copy to
+ * @dest_off: offset into that page
+ * @src_pg: pointer to the page to copy from
+ * @src_off: offset into that page
+ * @len: transaction length in bytes. This is guaranteed to not make a copy
+ *	 across a page boundary.
+ */
+
+static dma_cookie_t ioat_dma_memcpy_pg_to_pg(struct dma_chan *chan,
+                                             struct page *dest_pg,
+                                             unsigned int dest_off,
+                                             struct page *src_pg,
+                                             unsigned int src_off,
+                                             size_t len)
+{
+	dma_addr_t dest_addr;
+	dma_addr_t src_addr;
+	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
+
+	dest_addr = pci_map_page(ioat_chan->device->pdev,
+		dest_pg, dest_off, len, PCI_DMA_FROMDEVICE);
+	src_addr = pci_map_page(ioat_chan->device->pdev,
+		src_pg, src_off, len, PCI_DMA_TODEVICE);
+
+	return do_ioat_dma_memcpy(ioat_chan, dest_addr, src_addr, len);
+}
+
+/**
+ * ioat_dma_memcpy_issue_pending - push potentially unrecognoized appended descriptors to hw
+ * @chan: DMA channel handle
+ */
+
+static void ioat_dma_memcpy_issue_pending(struct dma_chan *chan)
+{
+	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
+
+	if (ioat_chan->pending != 0) {
+		ioat_chan->pending = 0;
+		ioatdma_chan_write8(ioat_chan,
+		                    IOAT_CHANCMD_OFFSET,
+		                    IOAT_CHANCMD_APPEND);
+	}
+}
+
+static void ioat_dma_memcpy_cleanup(struct ioat_dma_chan *chan)
+{
+	unsigned long phys_complete;
+	struct ioat_desc_sw *desc, *_desc;
+	dma_cookie_t cookie = 0;
+
+	prefetch(chan->completion_virt);
+
+	if (!spin_trylock(&chan->cleanup_lock))
+		return;
+
+	/* The completion writeback can happen at any time,
+	   so reads by the driver need to be atomic operations
+	   The descriptor physical addresses are limited to 32-bits
+	   when the CPU can only do a 32-bit mov */
+
+#if (BITS_PER_LONG == 64)
+	phys_complete =
+	chan->completion_virt->full & IOAT_CHANSTS_COMPLETED_DESCRIPTOR_ADDR;
+#else
+	phys_complete = chan->completion_virt->low & IOAT_LOW_COMPLETION_MASK;
+#endif
+
+	if ((chan->completion_virt->full & IOAT_CHANSTS_DMA_TRANSFER_STATUS) ==
+		IOAT_CHANSTS_DMA_TRANSFER_STATUS_HALTED) {
+		printk("IOAT: Channel halted, chanerr = %x\n",
+			ioatdma_chan_read32(chan, IOAT_CHANERR_OFFSET));
+
+		/* TODO do something to salvage the situation */
+	}
+
+	if (phys_complete == chan->last_completion) {
+		spin_unlock(&chan->cleanup_lock);
+		return;
+	}
+
+	spin_lock_bh(&chan->desc_lock);
+	list_for_each_entry_safe(desc, _desc, &chan->used_desc, node) {
+
+		/*
+		 * Incoming DMA requests may use multiple descriptors, due to
+		 * exceeding xfercap, perhaps. If so, only the last one will
+		 * have a cookie, and require unmapping.
+		 */
+		if (desc->cookie) {
+			cookie = desc->cookie;
+
+			/* yes we are unmapping both _page and _single alloc'd
+			   regions with unmap_page. Is this *really* that bad?
+			*/
+			pci_unmap_page(chan->device->pdev,
+					pci_unmap_addr(desc, dst),
+					pci_unmap_len(desc, dst_len),
+					PCI_DMA_FROMDEVICE);
+			pci_unmap_page(chan->device->pdev,
+					pci_unmap_addr(desc, src),
+					pci_unmap_len(desc, src_len),
+					PCI_DMA_TODEVICE);
+		}
+
+		if (desc->phys != phys_complete) {
+			/* a completed entry, but not the last, so cleanup */
+			list_del(&desc->node);
+			list_add_tail(&desc->node, &chan->free_desc);
+		} else {
+			/* last used desc. Do not remove, so we can append from
+			   it, but don't look at it next time, either */
+			desc->cookie = 0;
+
+			/* TODO check status bits? */
+			break;
+		}
+	}
+
+	spin_unlock_bh(&chan->desc_lock);
+
+	chan->last_completion = phys_complete;
+	if (cookie != 0)
+		chan->completed_cookie = cookie;
+
+	spin_unlock(&chan->cleanup_lock);
+}
+
+/**
+ * ioat_dma_is_complete - poll the status of a IOAT DMA transaction
+ * @chan: IOAT DMA channel handle
+ * @cookie: DMA transaction identifier
+ */
+
+static enum dma_status ioat_dma_is_complete(struct dma_chan *chan,
+                                            dma_cookie_t cookie,
+                                            dma_cookie_t *done,
+                                            dma_cookie_t *used)
+{
+	struct ioat_dma_chan *ioat_chan = to_ioat_chan(chan);
+	dma_cookie_t last_used;
+	dma_cookie_t last_complete;
+	enum dma_status ret;
+
+	last_used = chan->cookie;
+	last_complete = ioat_chan->completed_cookie;
+
+	if (done)
+		*done= last_complete;
+	if (used)
+		*used = last_used;
+
+	ret = dma_async_is_complete(cookie, last_complete, last_used);
+	if (ret == DMA_SUCCESS)
+		return ret;
+
+	ioat_dma_memcpy_cleanup(ioat_chan);
+
+	last_used = chan->cookie;
+	last_complete = ioat_chan->completed_cookie;
+
+	if (done)
+		*done= last_complete;
+	if (used)
+		*used = last_used;
+
+	return dma_async_is_complete(cookie, last_complete, last_used);
+}
+
+/* PCI API */
+
+static struct pci_device_id ioat_pci_tbl[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT) },
+	{ 0, }
+};
+
+static struct pci_driver ioat_pci_drv = {
+	.name 	= "ioatdma",
+	.id_table = ioat_pci_tbl,
+	.probe	= ioat_probe,
+	.remove	= __devexit_p(ioat_remove),
+};
+
+static irqreturn_t ioat_do_interrupt(int irq, void *data, struct pt_regs *regs)
+{
+	struct ioat_device *instance = data;
+	unsigned long attnstatus;
+	u8 intrctrl;
+
+	intrctrl = ioatdma_read8(instance, IOAT_INTRCTRL_OFFSET);
+
+	if (!(intrctrl & IOAT_INTRCTRL_MASTER_INT_EN))
+		return IRQ_NONE;
+
+	if (!(intrctrl & IOAT_INTRCTRL_INT_STATUS)) {
+		ioatdma_write8(instance, IOAT_INTRCTRL_OFFSET, intrctrl);
+		return IRQ_NONE;
+	}
+
+	attnstatus = ioatdma_read32(instance, IOAT_ATTNSTATUS_OFFSET);
+
+	printk(KERN_ERR "ioatdma error: interrupt! status %lx\n", attnstatus);
+
+	ioatdma_write8(instance, IOAT_INTRCTRL_OFFSET, intrctrl);
+	return IRQ_HANDLED;
+}
+
+static void ioat_start_null_desc(struct ioat_dma_chan *ioat_chan)
+{
+	struct ioat_desc_sw *desc;
+
+	spin_lock_bh(&ioat_chan->desc_lock);
+
+	if (!list_empty(&ioat_chan->free_desc)) {
+		desc = to_ioat_desc(ioat_chan->free_desc.next);
+		list_del(&desc->node);
+	} else {
+		/* try to get another desc */
+		spin_unlock_bh(&ioat_chan->desc_lock);
+		desc = ioat_dma_alloc_descriptor(ioat_chan, GFP_KERNEL);
+		spin_lock_bh(&ioat_chan->desc_lock);
+		/* will this ever happen? */
+		BUG_ON(!desc);
+	}
+
+	desc->hw->ctl = IOAT_DMA_DESCRIPTOR_NUL;
+	desc->hw->next = 0;
+
+	list_add_tail(&desc->node, &ioat_chan->used_desc);
+	spin_unlock_bh(&ioat_chan->desc_lock);
+
+#if (BITS_PER_LONG == 64)
+	ioatdma_chan_write64(ioat_chan, IOAT_CHAINADDR_OFFSET, desc->phys);
+#else
+	ioatdma_chan_write32(ioat_chan,
+	                     IOAT_CHAINADDR_OFFSET_LOW,
+	                     (u32) desc->phys);
+	ioatdma_chan_write32(ioat_chan, IOAT_CHAINADDR_OFFSET_HIGH, 0);
+#endif
+	ioatdma_chan_write8(ioat_chan, IOAT_CHANCMD_OFFSET, IOAT_CHANCMD_START);
+}
+
+/*
+ * Perform a IOAT transaction to verify the HW works.
+ */
+#define IOAT_TEST_SIZE 2000
+
+static int ioat_self_test(struct ioat_device *device)
+{
+	int i;
+	u8 *src;
+	u8 *dest;
+	struct dma_chan *dma_chan;
+	dma_cookie_t cookie;
+	int err = 0;
+
+	src = kzalloc(sizeof(u8) * IOAT_TEST_SIZE, SLAB_KERNEL);
+	if (!src)
+		return -ENOMEM;
+	dest = kzalloc(sizeof(u8) * IOAT_TEST_SIZE, SLAB_KERNEL);
+	if (!dest) {
+		kfree(src);
+		return -ENOMEM;
+	}
+
+	/* Fill in src buffer */
+	for (i = 0; i < IOAT_TEST_SIZE; i++)
+		src[i] = (u8)i;
+
+	/* Start copy, using first DMA channel */
+	dma_chan = container_of(device->common.channels.next,
+	                        struct dma_chan,
+	                        device_node);
+	if (ioat_dma_alloc_chan_resources(dma_chan) < 1) {
+		err = -ENODEV;
+		goto out;
+	}
+
+	cookie = ioat_dma_memcpy_buf_to_buf(dma_chan, dest, src, IOAT_TEST_SIZE);
+	ioat_dma_memcpy_issue_pending(dma_chan);
+	msleep(1);
+
+	if (ioat_dma_is_complete(dma_chan, cookie, NULL, NULL) != DMA_SUCCESS) {
+		printk(KERN_ERR "ioatdma: Self-test copy timed out, disabling\n");
+		err = -ENODEV;
+		goto free_resources;
+	}
+	if (memcmp(src, dest, IOAT_TEST_SIZE)) {
+		printk(KERN_ERR "ioatdma: Self-test copy failed compare, disabling\n");
+		err = -ENODEV;
+		goto free_resources;
+	}
+
+free_resources:
+	ioat_dma_free_chan_resources(dma_chan);
+out:
+	kfree(src);
+	kfree(dest);
+	return err;
+}
+
+static int __devinit ioat_probe(struct pci_dev *pdev,
+                                const struct pci_device_id *ent)
+{
+	int err;
+	unsigned long mmio_start, mmio_len;
+	void *reg_base;
+	struct ioat_device *device;
+
+	err = pci_enable_device(pdev);
+	if (err)
+		goto err_enable_device;
+
+	err = pci_set_dma_mask(pdev, DMA_64BIT_MASK);
+	if (err)
+		err = pci_set_dma_mask(pdev, DMA_32BIT_MASK);
+	if (err)
+		goto err_set_dma_mask;
+
+	err = pci_request_regions(pdev, ioat_pci_drv.name);
+	if (err)
+		goto err_request_regions;
+
+	mmio_start = pci_resource_start(pdev, 0);
+	mmio_len = pci_resource_len(pdev, 0);
+
+	reg_base = ioremap(mmio_start, mmio_len);
+	if (!reg_base) {
+		err = -ENOMEM;
+		goto err_ioremap;
+	}
+
+	device = kzalloc(sizeof(*device), GFP_KERNEL);
+	if (!device) {
+		err = -ENOMEM;
+		goto err_kzalloc;
+	}
+
+	/* DMA coherent memory pool for DMA descriptor allocations */
+	device->dma_pool = pci_pool_create("dma_desc_pool", pdev,
+		sizeof(struct ioat_dma_descriptor), 64, 0);
+	if (!device->dma_pool) {
+		err = -ENOMEM;
+		goto err_dma_pool;
+	}
+
+	device->completion_pool = pci_pool_create("completion_pool", pdev, sizeof(u64), SMP_CACHE_BYTES, SMP_CACHE_BYTES);
+	if (!device->completion_pool) {
+		err = -ENOMEM;
+		goto err_completion_pool;
+	}
+
+	device->pdev = pdev;
+	pci_set_drvdata(pdev, device);
+#ifdef CONFIG_PCI_MSI
+	if (pci_enable_msi(pdev) == 0) {
+		device->msi = 1;
+	} else {
+		device->msi = 0;
+	}
+#endif
+	err = request_irq(pdev->irq, &ioat_do_interrupt, SA_SHIRQ, "ioat",
+		device);
+	if (err)
+		goto err_irq;
+
+	device->reg_base = reg_base;
+
+	ioatdma_write8(device, IOAT_INTRCTRL_OFFSET, IOAT_INTRCTRL_MASTER_INT_EN);
+	pci_set_master(pdev);
+
+	INIT_LIST_HEAD(&device->common.channels);
+	enumerate_dma_channels(device);
+
+	device->common.device_alloc_chan_resources = ioat_dma_alloc_chan_resources;
+	device->common.device_free_chan_resources = ioat_dma_free_chan_resources;
+	device->common.device_memcpy_buf_to_buf = ioat_dma_memcpy_buf_to_buf;
+	device->common.device_memcpy_buf_to_pg = ioat_dma_memcpy_buf_to_pg;
+	device->common.device_memcpy_pg_to_pg = ioat_dma_memcpy_pg_to_pg;
+	device->common.device_memcpy_complete = ioat_dma_is_complete;
+	device->common.device_memcpy_issue_pending = ioat_dma_memcpy_issue_pending;
+	printk(KERN_INFO "Intel(R) I/OAT DMA Engine found, %d channels\n",
+		device->common.chancnt);
+
+	err = ioat_self_test(device);
+	if (err)
+		goto err_self_test;
+
+	dma_async_device_register(&device->common);
+
+	return 0;
+
+err_self_test:
+err_irq:
+	pci_pool_destroy(device->completion_pool);
+err_completion_pool:
+	pci_pool_destroy(device->dma_pool);
+err_dma_pool:
+	kfree(device);
+err_kzalloc:
+	iounmap(reg_base);
+err_ioremap:
+	pci_release_regions(pdev);
+err_request_regions:
+err_set_dma_mask:
+	pci_disable_device(pdev);
+err_enable_device:
+	return err;
+}
+
+static void __devexit ioat_remove(struct pci_dev *pdev)
+{
+	struct ioat_device *device;
+	struct dma_chan *chan, *_chan;
+	struct ioat_dma_chan *ioat_chan;
+
+	device = pci_get_drvdata(pdev);
+	dma_async_device_unregister(&device->common);
+
+	free_irq(device->pdev->irq, device);
+#ifdef CONFIG_PCI_MSI
+	if (device->msi)
+		pci_disable_msi(device->pdev);
+#endif
+	pci_pool_destroy(device->dma_pool);
+	pci_pool_destroy(device->completion_pool);
+	iounmap(device->reg_base);
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+	list_for_each_entry_safe(chan, _chan, &device->common.channels, device_node) {
+		ioat_chan = to_ioat_chan(chan);
+		list_del(&chan->device_node);
+		kfree(ioat_chan);
+	}
+	kfree(device);
+}
+
+/* MODULE API */
+MODULE_VERSION("1.7");
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Intel Corporation");
+
+static int __init ioat_init_module(void)
+{
+	/* it's currently unsafe to unload this module */
+	/* if forced, worst case is that rmmod hangs */
+	if (THIS_MODULE != NULL)
+		THIS_MODULE->unsafe = 1;
+
+	return pci_module_init(&ioat_pci_drv);
+}
+
+module_init(ioat_init_module);
+
+static void __exit ioat_exit_module(void)
+{
+	pci_unregister_driver(&ioat_pci_drv);
+}
+
+module_exit(ioat_exit_module);

drivers/dma/ioatdma.h

@@ -0,0 +1,125 @@
+/*
+ * Copyright(c) 2004 - 2006 Intel Corporation. All rights reserved.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called COPYING.
+ */
+#ifndef IOATDMA_H
+#define IOATDMA_H
+
+#include <linux/dmaengine.h>
+#include "ioatdma_hw.h"
+#include <linux/init.h>
+#include <linux/dmapool.h>
+#include <linux/cache.h>
+#include <linux/pci_ids.h>
+
+#define IOAT_LOW_COMPLETION_MASK	0xffffffc0
+
+extern struct list_head dma_device_list;
+extern struct list_head dma_client_list;
+
+/**
+ * struct ioat_device - internal representation of a IOAT device
+ * @pdev: PCI-Express device
+ * @reg_base: MMIO register space base address
+ * @dma_pool: for allocating DMA descriptors
+ * @common: embedded struct dma_device
+ * @msi: Message Signaled Interrupt number
+ */
+
+struct ioat_device {
+	struct pci_dev *pdev;
+	void *reg_base;
+	struct pci_pool *dma_pool;
+	struct pci_pool *completion_pool;
+
+	struct dma_device common;
+	u8 msi;
+};
+
+/**
+ * struct ioat_dma_chan - internal representation of a DMA channel
+ * @device:
+ * @reg_base:
+ * @sw_in_use:
+ * @completion:
+ * @completion_low:
+ * @completion_high:
+ * @completed_cookie: last cookie seen completed on cleanup
+ * @cookie: value of last cookie given to client
+ * @last_completion:
+ * @xfercap:
+ * @desc_lock:
+ * @free_desc:
+ * @used_desc:
+ * @resource:
+ * @device_node:
+ */
+
+struct ioat_dma_chan {
+
+	void *reg_base;
+
+	dma_cookie_t completed_cookie;
+	unsigned long last_completion;
+
+	u32 xfercap;	/* XFERCAP register value expanded out */
+
+	spinlock_t cleanup_lock;
+	spinlock_t desc_lock;
+	struct list_head free_desc;
+	struct list_head used_desc;
+
+	int pending;
+
+	struct ioat_device *device;
+	struct dma_chan common;
+
+	dma_addr_t completion_addr;
+	union {
+		u64 full; /* HW completion writeback */
+		struct {
+			u32 low;
+			u32 high;
+		};
+	} *completion_virt;
+};
+
+/* wrapper around hardware descriptor format + additional software fields */
+
+/**
+ * struct ioat_desc_sw - wrapper around hardware descriptor
+ * @hw: hardware DMA descriptor
+ * @node:
+ * @cookie:
+ * @phys:
+ */
+
+struct ioat_desc_sw {
+	struct ioat_dma_descriptor *hw;
+	struct list_head node;
+	dma_cookie_t cookie;
+	dma_addr_t phys;
+	DECLARE_PCI_UNMAP_ADDR(src)
+	DECLARE_PCI_UNMAP_LEN(src_len)
+	DECLARE_PCI_UNMAP_ADDR(dst)
+	DECLARE_PCI_UNMAP_LEN(dst_len)
+};
+
+#endif /* IOATDMA_H */
+

drivers/dma/ioatdma_hw.h

@@ -0,0 +1,52 @@
+/*
+ * Copyright(c) 2004 - 2006 Intel Corporation. All rights reserved.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called COPYING.
+ */
+#ifndef _IOAT_HW_H_
+#define _IOAT_HW_H_
+
+/* PCI Configuration Space Values */
+#define IOAT_PCI_VID			0x8086
+#define IOAT_PCI_DID			0x1A38
+#define IOAT_PCI_RID			0x00
+#define IOAT_PCI_SVID			0x8086
+#define IOAT_PCI_SID			0x8086
+#define IOAT_VER			0x12	/* Version 1.2 */
+
+struct ioat_dma_descriptor {
+	uint32_t	size;
+	uint32_t	ctl;
+	uint64_t	src_addr;
+	uint64_t	dst_addr;
+	uint64_t	next;
+	uint64_t	rsv1;
+	uint64_t	rsv2;
+	uint64_t	user1;
+	uint64_t	user2;
+};
+
+#define IOAT_DMA_DESCRIPTOR_CTL_INT_GN	0x00000001
+#define IOAT_DMA_DESCRIPTOR_CTL_SRC_SN	0x00000002
+#define IOAT_DMA_DESCRIPTOR_CTL_DST_SN	0x00000004
+#define IOAT_DMA_DESCRIPTOR_CTL_CP_STS	0x00000008
+#define IOAT_DMA_DESCRIPTOR_CTL_FRAME	0x00000010
+#define IOAT_DMA_DESCRIPTOR_NUL		0x00000020
+#define IOAT_DMA_DESCRIPTOR_OPCODE	0xFF000000
+
+#endif

drivers/dma/ioatdma_io.h

@@ -0,0 +1,118 @@
+/*
+ * Copyright(c) 2004 - 2006 Intel Corporation. All rights reserved.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called COPYING.
+ */
+#ifndef IOATDMA_IO_H
+#define IOATDMA_IO_H
+
+#include <asm/io.h>
+
+/*
+ * device and per-channel MMIO register read and write functions
+ * this is a lot of anoying inline functions, but it's typesafe
+ */
+
+static inline u8 ioatdma_read8(struct ioat_device *device,
+                               unsigned int offset)
+{
+	return readb(device->reg_base + offset);
+}
+
+static inline u16 ioatdma_read16(struct ioat_device *device,
+                                 unsigned int offset)
+{
+	return readw(device->reg_base + offset);
+}
+
+static inline u32 ioatdma_read32(struct ioat_device *device,
+                                 unsigned int offset)
+{
+	return readl(device->reg_base + offset);
+}
+
+static inline void ioatdma_write8(struct ioat_device *device,
+                                  unsigned int offset, u8 value)
+{
+	writeb(value, device->reg_base + offset);
+}
+
+static inline void ioatdma_write16(struct ioat_device *device,
+                                   unsigned int offset, u16 value)
+{
+	writew(value, device->reg_base + offset);
+}
+
+static inline void ioatdma_write32(struct ioat_device *device,
+                                   unsigned int offset, u32 value)
+{
+	writel(value, device->reg_base + offset);
+}
+
+static inline u8 ioatdma_chan_read8(struct ioat_dma_chan *chan,
+                                    unsigned int offset)
+{
+	return readb(chan->reg_base + offset);
+}
+
+static inline u16 ioatdma_chan_read16(struct ioat_dma_chan *chan,
+                                      unsigned int offset)
+{
+	return readw(chan->reg_base + offset);
+}
+
+static inline u32 ioatdma_chan_read32(struct ioat_dma_chan *chan,
+                                      unsigned int offset)
+{
+	return readl(chan->reg_base + offset);
+}
+
+static inline void ioatdma_chan_write8(struct ioat_dma_chan *chan,
+                                       unsigned int offset, u8 value)
+{
+	writeb(value, chan->reg_base + offset);
+}
+
+static inline void ioatdma_chan_write16(struct ioat_dma_chan *chan,
+                                        unsigned int offset, u16 value)
+{
+	writew(value, chan->reg_base + offset);
+}
+
+static inline void ioatdma_chan_write32(struct ioat_dma_chan *chan,
+                                        unsigned int offset, u32 value)
+{
+	writel(value, chan->reg_base + offset);
+}
+
+#if (BITS_PER_LONG == 64)
+static inline u64 ioatdma_chan_read64(struct ioat_dma_chan *chan,
+                                      unsigned int offset)
+{
+	return readq(chan->reg_base + offset);
+}
+
+static inline void ioatdma_chan_write64(struct ioat_dma_chan *chan,
+                                        unsigned int offset, u64 value)
+{
+	writeq(value, chan->reg_base + offset);
+}
+#endif
+
+#endif /* IOATDMA_IO_H */
+

drivers/dma/ioatdma_registers.h

@@ -0,0 +1,126 @@
+/*
+ * Copyright(c) 2004 - 2006 Intel Corporation. All rights reserved.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called COPYING.
+ */
+#ifndef _IOAT_REGISTERS_H_
+#define _IOAT_REGISTERS_H_
+
+
+/* MMIO Device Registers */
+#define IOAT_CHANCNT_OFFSET			0x00	/*  8-bit */
+
+#define IOAT_XFERCAP_OFFSET			0x01	/*  8-bit */
+#define IOAT_XFERCAP_4KB			12
+#define IOAT_XFERCAP_8KB			13
+#define IOAT_XFERCAP_16KB			14
+#define IOAT_XFERCAP_32KB			15
+#define IOAT_XFERCAP_32GB			0
+
+#define IOAT_GENCTRL_OFFSET			0x02	/*  8-bit */
+#define IOAT_GENCTRL_DEBUG_EN			0x01
+
+#define IOAT_INTRCTRL_OFFSET			0x03	/*  8-bit */
+#define IOAT_INTRCTRL_MASTER_INT_EN		0x01	/* Master Interrupt Enable */
+#define IOAT_INTRCTRL_INT_STATUS		0x02	/* ATTNSTATUS -or- Channel Int */
+#define IOAT_INTRCTRL_INT			0x04	/* INT_STATUS -and- MASTER_INT_EN */
+
+#define IOAT_ATTNSTATUS_OFFSET			0x04	/* Each bit is a channel */
+
+#define IOAT_VER_OFFSET				0x08	/*  8-bit */
+#define IOAT_VER_MAJOR_MASK			0xF0
+#define IOAT_VER_MINOR_MASK			0x0F
+#define GET_IOAT_VER_MAJOR(x)			((x) & IOAT_VER_MAJOR_MASK)
+#define GET_IOAT_VER_MINOR(x)			((x) & IOAT_VER_MINOR_MASK)
+
+#define IOAT_PERPORTOFFSET_OFFSET		0x0A	/* 16-bit */
+
+#define IOAT_INTRDELAY_OFFSET			0x0C	/* 16-bit */
+#define IOAT_INTRDELAY_INT_DELAY_MASK		0x3FFF	/* Interrupt Delay Time */
+#define IOAT_INTRDELAY_COALESE_SUPPORT		0x8000	/* Interrupt Coalesing Supported */
+
+#define IOAT_DEVICE_STATUS_OFFSET		0x0E	/* 16-bit */
+#define IOAT_DEVICE_STATUS_DEGRADED_MODE	0x0001
+
+
+#define IOAT_CHANNEL_MMIO_SIZE			0x80	/* Each Channel MMIO space is this size */
+
+/* DMA Channel Registers */
+#define IOAT_CHANCTRL_OFFSET			0x00	/* 16-bit Channel Control Register */
+#define IOAT_CHANCTRL_CHANNEL_PRIORITY_MASK	0xF000
+#define IOAT_CHANCTRL_CHANNEL_IN_USE		0x0100
+#define IOAT_CHANCTRL_DESCRIPTOR_ADDR_SNOOP_CONTROL	0x0020
+#define IOAT_CHANCTRL_ERR_INT_EN		0x0010
+#define IOAT_CHANCTRL_ANY_ERR_ABORT_EN		0x0008
+#define IOAT_CHANCTRL_ERR_COMPLETION_EN		0x0004
+#define IOAT_CHANCTRL_INT_DISABLE		0x0001
+
+#define IOAT_DMA_COMP_OFFSET			0x02	/* 16-bit DMA channel compatability */
+#define IOAT_DMA_COMP_V1			0x0001	/* Compatability with DMA version 1 */
+
+#define IOAT_CHANSTS_OFFSET			0x04	/* 64-bit Channel Status Register */
+#define IOAT_CHANSTS_OFFSET_LOW			0x04
+#define IOAT_CHANSTS_OFFSET_HIGH		0x08
+#define IOAT_CHANSTS_COMPLETED_DESCRIPTOR_ADDR	0xFFFFFFFFFFFFFFC0
+#define IOAT_CHANSTS_SOFT_ERR			0x0000000000000010
+#define IOAT_CHANSTS_DMA_TRANSFER_STATUS	0x0000000000000007
+#define IOAT_CHANSTS_DMA_TRANSFER_STATUS_ACTIVE	0x0
+#define IOAT_CHANSTS_DMA_TRANSFER_STATUS_DONE	0x1
+#define IOAT_CHANSTS_DMA_TRANSFER_STATUS_SUSPENDED	0x2
+#define IOAT_CHANSTS_DMA_TRANSFER_STATUS_HALTED	0x3
+
+#define IOAT_CHAINADDR_OFFSET			0x0C	/* 64-bit Descriptor Chain Address Register */
+#define IOAT_CHAINADDR_OFFSET_LOW		0x0C
+#define IOAT_CHAINADDR_OFFSET_HIGH		0x10
+
+#define IOAT_CHANCMD_OFFSET			0x14	/*  8-bit DMA Channel Command Register */
+#define IOAT_CHANCMD_RESET			0x20
+#define IOAT_CHANCMD_RESUME			0x10
+#define IOAT_CHANCMD_ABORT			0x08
+#define IOAT_CHANCMD_SUSPEND			0x04
+#define IOAT_CHANCMD_APPEND			0x02
+#define IOAT_CHANCMD_START			0x01
+
+#define IOAT_CHANCMP_OFFSET			0x18	/* 64-bit Channel Completion Address Register */
+#define IOAT_CHANCMP_OFFSET_LOW			0x18
+#define IOAT_CHANCMP_OFFSET_HIGH		0x1C
+
+#define IOAT_CDAR_OFFSET			0x20	/* 64-bit Current Descriptor Address Register */
+#define IOAT_CDAR_OFFSET_LOW			0x20
+#define IOAT_CDAR_OFFSET_HIGH			0x24
+
+#define IOAT_CHANERR_OFFSET			0x28	/* 32-bit Channel Error Register */
+#define IOAT_CHANERR_DMA_TRANSFER_SRC_ADDR_ERR	0x0001
+#define IOAT_CHANERR_DMA_TRANSFER_DEST_ADDR_ERR	0x0002
+#define IOAT_CHANERR_NEXT_DESCRIPTOR_ADDR_ERR	0x0004
+#define IOAT_CHANERR_NEXT_DESCRIPTOR_ALIGNMENT_ERR	0x0008
+#define IOAT_CHANERR_CHAIN_ADDR_VALUE_ERR	0x0010
+#define IOAT_CHANERR_CHANCMD_ERR		0x0020
+#define IOAT_CHANERR_CHIPSET_UNCORRECTABLE_DATA_INTEGRITY_ERR	0x0040
+#define IOAT_CHANERR_DMA_UNCORRECTABLE_DATA_INTEGRITY_ERR	0x0080
+#define IOAT_CHANERR_READ_DATA_ERR		0x0100
+#define IOAT_CHANERR_WRITE_DATA_ERR		0x0200
+#define IOAT_CHANERR_DESCRIPTOR_CONTROL_ERR	0x0400
+#define IOAT_CHANERR_DESCRIPTOR_LENGTH_ERR	0x0800
+#define IOAT_CHANERR_COMPLETION_ADDR_ERR	0x1000
+#define IOAT_CHANERR_INT_CONFIGURATION_ERR	0x2000
+#define IOAT_CHANERR_SOFT_ERR			0x4000
+
+#define IOAT_CHANERR_MASK_OFFSET		0x2C	/* 32-bit Channel Error Register */
+
+#endif /* _IOAT_REGISTERS_H_ */

drivers/dma/iovlock.c

@@ -0,0 +1,301 @@
+/*
+ * Copyright(c) 2004 - 2006 Intel Corporation. All rights reserved.
+ * Portions based on net/core/datagram.c and copyrighted by their authors.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called COPYING.
+ */
+
+/*
+ * This code allows the net stack to make use of a DMA engine for
+ * skb to iovec copies.
+ */
+
+#include <linux/dmaengine.h>
+#include <linux/pagemap.h>
+#include <net/tcp.h> /* for memcpy_toiovec */
+#include <asm/io.h>
+#include <asm/uaccess.h>
+
+int num_pages_spanned(struct iovec *iov)
+{
+	return
+	((PAGE_ALIGN((unsigned long)iov->iov_base + iov->iov_len) -
+	((unsigned long)iov->iov_base & PAGE_MASK)) >> PAGE_SHIFT);
+}
+
+/*
+ * Pin down all the iovec pages needed for len bytes.
+ * Return a struct dma_pinned_list to keep track of pages pinned down.
+ *
+ * We are allocating a single chunk of memory, and then carving it up into
+ * 3 sections, the latter 2 whose size depends on the number of iovecs and the
+ * total number of pages, respectively.
+ */
+struct dma_pinned_list *dma_pin_iovec_pages(struct iovec *iov, size_t len)
+{
+	struct dma_pinned_list *local_list;
+	struct page **pages;
+	int i;
+	int ret;
+	int nr_iovecs = 0;
+	int iovec_len_used = 0;
+	int iovec_pages_used = 0;
+	long err;
+
+	/* don't pin down non-user-based iovecs */
+	if (segment_eq(get_fs(), KERNEL_DS))
+		return NULL;
+
+	/* determine how many iovecs/pages there are, up front */
+	do {
+		iovec_len_used += iov[nr_iovecs].iov_len;
+		iovec_pages_used += num_pages_spanned(&iov[nr_iovecs]);
+		nr_iovecs++;
+	} while (iovec_len_used < len);
+
+	/* single kmalloc for pinned list, page_list[], and the page arrays */
+	local_list = kmalloc(sizeof(*local_list)
+		+ (nr_iovecs * sizeof (struct dma_page_list))
+		+ (iovec_pages_used * sizeof (struct page*)), GFP_KERNEL);
+	if (!local_list) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	/* list of pages starts right after the page list array */
+	pages = (struct page **) &local_list->page_list[nr_iovecs];
+
+	for (i = 0; i < nr_iovecs; i++) {
+		struct dma_page_list *page_list = &local_list->page_list[i];
+
+		len -= iov[i].iov_len;
+
+		if (!access_ok(VERIFY_WRITE, iov[i].iov_base, iov[i].iov_len)) {
+			err = -EFAULT;
+			goto unpin;
+		}
+
+		page_list->nr_pages = num_pages_spanned(&iov[i]);
+		page_list->base_address = iov[i].iov_base;
+
+		page_list->pages = pages;
+		pages += page_list->nr_pages;
+
+		/* pin pages down */
+		down_read(&current->mm->mmap_sem);
+		ret = get_user_pages(
+			current,
+			current->mm,
+			(unsigned long) iov[i].iov_base,
+			page_list->nr_pages,
+			1,	/* write */
+			0,	/* force */
+			page_list->pages,
+			NULL);
+		up_read(&current->mm->mmap_sem);
+
+		if (ret != page_list->nr_pages) {
+			err = -ENOMEM;
+			goto unpin;
+		}
+
+		local_list->nr_iovecs = i + 1;
+	}
+
+	return local_list;
+
+unpin:
+	dma_unpin_iovec_pages(local_list);
+out:
+	return ERR_PTR(err);
+}
+
+void dma_unpin_iovec_pages(struct dma_pinned_list *pinned_list)
+{
+	int i, j;
+
+	if (!pinned_list)
+		return;
+
+	for (i = 0; i < pinned_list->nr_iovecs; i++) {
+		struct dma_page_list *page_list = &pinned_list->page_list[i];
+		for (j = 0; j < page_list->nr_pages; j++) {
+			set_page_dirty_lock(page_list->pages[j]);
+			page_cache_release(page_list->pages[j]);
+		}
+	}
+
+	kfree(pinned_list);
+}
+
+static dma_cookie_t dma_memcpy_to_kernel_iovec(struct dma_chan *chan, struct
+	iovec *iov, unsigned char *kdata, size_t len)
+{
+	dma_cookie_t dma_cookie = 0;
+
+	while (len > 0) {
+		if (iov->iov_len) {
+			int copy = min_t(unsigned int, iov->iov_len, len);
+			dma_cookie = dma_async_memcpy_buf_to_buf(
+					chan,
+					iov->iov_base,
+					kdata,
+					copy);
+			kdata += copy;
+			len -= copy;
+			iov->iov_len -= copy;
+			iov->iov_base += copy;
+		}
+		iov++;
+	}
+
+	return dma_cookie;
+}
+
+/*
+ * We have already pinned down the pages we will be using in the iovecs.
+ * Each entry in iov array has corresponding entry in pinned_list->page_list.
+ * Using array indexing to keep iov[] and page_list[] in sync.
+ * Initial elements in iov array's iov->iov_len will be 0 if already copied into
+ *   by another call.
+ * iov array length remaining guaranteed to be bigger than len.
+ */
+dma_cookie_t dma_memcpy_to_iovec(struct dma_chan *chan, struct iovec *iov,
+	struct dma_pinned_list *pinned_list, unsigned char *kdata, size_t len)
+{
+	int iov_byte_offset;
+	int copy;
+	dma_cookie_t dma_cookie = 0;
+	int iovec_idx;
+	int page_idx;
+
+	if (!chan)
+		return memcpy_toiovec(iov, kdata, len);
+
+	/* -> kernel copies (e.g. smbfs) */
+	if (!pinned_list)
+		return dma_memcpy_to_kernel_iovec(chan, iov, kdata, len);
+
+	iovec_idx = 0;
+	while (iovec_idx < pinned_list->nr_iovecs) {
+		struct dma_page_list *page_list;
+
+		/* skip already used-up iovecs */
+		while (!iov[iovec_idx].iov_len)
+			iovec_idx++;
+
+		page_list = &pinned_list->page_list[iovec_idx];
+
+		iov_byte_offset = ((unsigned long)iov[iovec_idx].iov_base & ~PAGE_MASK);
+		page_idx = (((unsigned long)iov[iovec_idx].iov_base & PAGE_MASK)
+			 - ((unsigned long)page_list->base_address & PAGE_MASK)) >> PAGE_SHIFT;
+
+		/* break up copies to not cross page boundary */
+		while (iov[iovec_idx].iov_len) {
+			copy = min_t(int, PAGE_SIZE - iov_byte_offset, len);
+			copy = min_t(int, copy, iov[iovec_idx].iov_len);
+
+			dma_cookie = dma_async_memcpy_buf_to_pg(chan,
+					page_list->pages[page_idx],
+					iov_byte_offset,
+					kdata,
+					copy);
+
+			len -= copy;
+			iov[iovec_idx].iov_len -= copy;
+			iov[iovec_idx].iov_base += copy;
+
+			if (!len)
+				return dma_cookie;
+
+			kdata += copy;
+			iov_byte_offset = 0;
+			page_idx++;
+		}
+		iovec_idx++;
+	}
+
+	/* really bad if we ever run out of iovecs */
+	BUG();
+	return -EFAULT;
+}
+
+dma_cookie_t dma_memcpy_pg_to_iovec(struct dma_chan *chan, struct iovec *iov,
+	struct dma_pinned_list *pinned_list, struct page *page,
+	unsigned int offset, size_t len)
+{
+	int iov_byte_offset;
+	int copy;
+	dma_cookie_t dma_cookie = 0;
+	int iovec_idx;
+	int page_idx;
+	int err;
+
+	/* this needs as-yet-unimplemented buf-to-buff, so punt. */
+	/* TODO: use dma for this */
+	if (!chan || !pinned_list) {
+		u8 *vaddr = kmap(page);
+		err = memcpy_toiovec(iov, vaddr + offset, len);
+		kunmap(page);
+		return err;
+	}
+
+	iovec_idx = 0;
+	while (iovec_idx < pinned_list->nr_iovecs) {
+		struct dma_page_list *page_list;
+
+		/* skip already used-up iovecs */
+		while (!iov[iovec_idx].iov_len)
+			iovec_idx++;
+
+		page_list = &pinned_list->page_list[iovec_idx];
+
+		iov_byte_offset = ((unsigned long)iov[iovec_idx].iov_base & ~PAGE_MASK);
+		page_idx = (((unsigned long)iov[iovec_idx].iov_base & PAGE_MASK)
+			 - ((unsigned long)page_list->base_address & PAGE_MASK)) >> PAGE_SHIFT;
+
+		/* break up copies to not cross page boundary */
+		while (iov[iovec_idx].iov_len) {
+			copy = min_t(int, PAGE_SIZE - iov_byte_offset, len);
+			copy = min_t(int, copy, iov[iovec_idx].iov_len);
+
+			dma_cookie = dma_async_memcpy_pg_to_pg(chan,
+					page_list->pages[page_idx],
+					iov_byte_offset,
+					page,
+					offset,
+					copy);
+
+			len -= copy;
+			iov[iovec_idx].iov_len -= copy;
+			iov[iovec_idx].iov_base += copy;
+
+			if (!len)
+				return dma_cookie;
+
+			offset += copy;
+			iov_byte_offset = 0;
+			page_idx++;
+		}
+		iovec_idx++;
+	}
+
+	/* really bad if we ever run out of iovecs */
+	BUG();
+	return -EFAULT;
+}

drivers/infiniband/ulp/ipoib/ipoib_multicast.c

@@ -821,7 +821,8 @@ void ipoib_mcast_restart_task(void *dev_ptr)
 
 	ipoib_mcast_stop_thread(dev, 0);
 
-	spin_lock_irqsave(&dev->xmit_lock, flags);
+	local_irq_save(flags);
+	netif_tx_lock(dev);
 	spin_lock(&priv->lock);
 
 	/*
@@ -896,7 +897,8 @@ void ipoib_mcast_restart_task(void *dev_ptr)
 	}
 
 	spin_unlock(&priv->lock);
-	spin_unlock_irqrestore(&dev->xmit_lock, flags);
+	netif_tx_unlock(dev);
+	local_irq_restore(flags);
 
 	/* We have to cancel outside of the spinlock */
 	list_for_each_entry_safe(mcast, tmcast, &remove_list, list) {

drivers/media/dvb/dvb-core/dvb_net.c

@@ -1052,7 +1052,7 @@ static void wq_set_multicast_list (void *data)
 
 	dvb_net_feed_stop(dev);
 	priv->rx_mode = RX_MODE_UNI;
-	spin_lock_bh(&dev->xmit_lock);
+	netif_tx_lock_bh(dev);
 
 	if (dev->flags & IFF_PROMISC) {
 		dprintk("%s: promiscuous mode\n", dev->name);
@@ -1077,7 +1077,7 @@ static void wq_set_multicast_list (void *data)
 		}
 	}
 
-	spin_unlock_bh(&dev->xmit_lock);
+	netif_tx_unlock_bh(dev);
 	dvb_net_feed_start(dev);
 }
 

drivers/net/Kconfig

@@ -2180,6 +2180,8 @@ config TIGON3
 config BNX2
 	tristate "Broadcom NetXtremeII support"
 	depends on PCI
+	select CRC32
+	select ZLIB_INFLATE
 	help
 	  This driver supports Broadcom NetXtremeII gigabit Ethernet cards.
 

drivers/net/bnx2.c

@@ -32,6 +32,7 @@
 #include <asm/irq.h>
 #include <linux/delay.h>
 #include <asm/byteorder.h>
+#include <asm/page.h>
 #include <linux/time.h>
 #include <linux/ethtool.h>
 #include <linux/mii.h>
@@ -49,14 +50,15 @@
 #include <linux/crc32.h>
 #include <linux/prefetch.h>
 #include <linux/cache.h>
+#include <linux/zlib.h>
 
 #include "bnx2.h"
 #include "bnx2_fw.h"
 
 #define DRV_MODULE_NAME		"bnx2"
 #define PFX DRV_MODULE_NAME	": "
-#define DRV_MODULE_VERSION	"1.4.40"
+#define DRV_MODULE_VERSION	"1.4.42"
-#define DRV_MODULE_RELDATE	"May 22, 2006"
+#define DRV_MODULE_RELDATE	"June 12, 2006"
 
 #define RUN_AT(x) (jiffies + (x))
 
@@ -1820,7 +1822,7 @@ reuse_rx:
 		skb->protocol = eth_type_trans(skb, bp->dev);
 
 		if ((len > (bp->dev->mtu + ETH_HLEN)) &&
-			(htons(skb->protocol) != 0x8100)) {
+			(ntohs(skb->protocol) != 0x8100)) {
 
 			dev_kfree_skb_irq(skb);
 			goto next_rx;
@@ -2009,7 +2011,7 @@ bnx2_poll(struct net_device *dev, int *budget)
 	return 1;
 }
 
-/* Called with rtnl_lock from vlan functions and also dev->xmit_lock
+/* Called with rtnl_lock from vlan functions and also netif_tx_lock
  * from set_multicast.
  */
 static void
@@ -2083,6 +2085,92 @@ bnx2_set_rx_mode(struct net_device *dev)
 	spin_unlock_bh(&bp->phy_lock);
 }
 
+#define FW_BUF_SIZE	0x8000
+
+static int
+bnx2_gunzip_init(struct bnx2 *bp)
+{
+	if ((bp->gunzip_buf = vmalloc(FW_BUF_SIZE)) == NULL)
+		goto gunzip_nomem1;
+
+	if ((bp->strm = kmalloc(sizeof(*bp->strm), GFP_KERNEL)) == NULL)
+		goto gunzip_nomem2;
+
+	bp->strm->workspace = kmalloc(zlib_inflate_workspacesize(), GFP_KERNEL);
+	if (bp->strm->workspace == NULL)
+		goto gunzip_nomem3;
+
+	return 0;
+
+gunzip_nomem3:
+	kfree(bp->strm);
+	bp->strm = NULL;
+
+gunzip_nomem2:
+	vfree(bp->gunzip_buf);
+	bp->gunzip_buf = NULL;
+
+gunzip_nomem1:
+	printk(KERN_ERR PFX "%s: Cannot allocate firmware buffer for "
+			    "uncompression.\n", bp->dev->name);
+	return -ENOMEM;
+}
+
+static void
+bnx2_gunzip_end(struct bnx2 *bp)
+{
+	kfree(bp->strm->workspace);
+
+	kfree(bp->strm);
+	bp->strm = NULL;
+
+	if (bp->gunzip_buf) {
+		vfree(bp->gunzip_buf);
+		bp->gunzip_buf = NULL;
+	}
+}
+
+static int
+bnx2_gunzip(struct bnx2 *bp, u8 *zbuf, int len, void **outbuf, int *outlen)
+{
+	int n, rc;
+
+	/* check gzip header */
+	if ((zbuf[0] != 0x1f) || (zbuf[1] != 0x8b) || (zbuf[2] != Z_DEFLATED))
+		return -EINVAL;
+
+	n = 10;
+
+#define FNAME	0x8
+	if (zbuf[3] & FNAME)
+		while ((zbuf[n++] != 0) && (n < len));
+
+	bp->strm->next_in = zbuf + n;
+	bp->strm->avail_in = len - n;
+	bp->strm->next_out = bp->gunzip_buf;
+	bp->strm->avail_out = FW_BUF_SIZE;
+
+	rc = zlib_inflateInit2(bp->strm, -MAX_WBITS);
+	if (rc != Z_OK)
+		return rc;
+
+	rc = zlib_inflate(bp->strm, Z_FINISH);
+
+	*outlen = FW_BUF_SIZE - bp->strm->avail_out;
+	*outbuf = bp->gunzip_buf;
+
+	if ((rc != Z_OK) && (rc != Z_STREAM_END))
+		printk(KERN_ERR PFX "%s: Firmware decompression error: %s\n",
+		       bp->dev->name, bp->strm->msg);
+
+	zlib_inflateEnd(bp->strm);
+
+	if (rc == Z_STREAM_END)
+		return 0;
+
+	return rc;
+}
+
 static void
 load_rv2p_fw(struct bnx2 *bp, u32 *rv2p_code, u32 rv2p_code_len,
 	u32 rv2p_proc)
@@ -2092,9 +2180,9 @@ load_rv2p_fw(struct bnx2 *bp, u32 *rv2p_code, u32 rv2p_code_len,
 
 
 	for (i = 0; i < rv2p_code_len; i += 8) {
-		REG_WR(bp, BNX2_RV2P_INSTR_HIGH, *rv2p_code);
+		REG_WR(bp, BNX2_RV2P_INSTR_HIGH, cpu_to_le32(*rv2p_code));
 		rv2p_code++;
-		REG_WR(bp, BNX2_RV2P_INSTR_LOW, *rv2p_code);
+		REG_WR(bp, BNX2_RV2P_INSTR_LOW, cpu_to_le32(*rv2p_code));
 		rv2p_code++;
 
 		if (rv2p_proc == RV2P_PROC1) {
@@ -2134,7 +2222,7 @@ load_cpu_fw(struct bnx2 *bp, struct cpu_reg *cpu_reg, struct fw_info *fw)
 		int j;
 
 		for (j = 0; j < (fw->text_len / 4); j++, offset += 4) {
-			REG_WR_IND(bp, offset, fw->text[j]);
+			REG_WR_IND(bp, offset, cpu_to_le32(fw->text[j]));
 	        }
 	}
 
@@ -2190,15 +2278,32 @@ load_cpu_fw(struct bnx2 *bp, struct cpu_reg *cpu_reg, struct fw_info *fw)
 	REG_WR_IND(bp, cpu_reg->mode, val);
 }
 
-static void
+static int
 bnx2_init_cpus(struct bnx2 *bp)
 {
 	struct cpu_reg cpu_reg;
 	struct fw_info fw;
+	int rc = 0;
+	void *text;
+	u32 text_len;
+
+	if ((rc = bnx2_gunzip_init(bp)) != 0)
+		return rc;
 
 	/* Initialize the RV2P processor. */
-	load_rv2p_fw(bp, bnx2_rv2p_proc1, sizeof(bnx2_rv2p_proc1), RV2P_PROC1);
+	rc = bnx2_gunzip(bp, bnx2_rv2p_proc1, sizeof(bnx2_rv2p_proc1), &text,
-	load_rv2p_fw(bp, bnx2_rv2p_proc2, sizeof(bnx2_rv2p_proc2), RV2P_PROC2);
+			 &text_len);
+	if (rc)
+		goto init_cpu_err;
+
+	load_rv2p_fw(bp, text, text_len, RV2P_PROC1);
+
+	rc = bnx2_gunzip(bp, bnx2_rv2p_proc2, sizeof(bnx2_rv2p_proc2), &text,
+			 &text_len);
+	if (rc)
+		goto init_cpu_err;
+
+	load_rv2p_fw(bp, text, text_len, RV2P_PROC2);
 
 	/* Initialize the RX Processor. */
 	cpu_reg.mode = BNX2_RXP_CPU_MODE;
@@ -2222,7 +2327,13 @@ bnx2_init_cpus(struct bnx2 *bp)
 	fw.text_addr = bnx2_RXP_b06FwTextAddr;
 	fw.text_len = bnx2_RXP_b06FwTextLen;
 	fw.text_index = 0;
-	fw.text = bnx2_RXP_b06FwText;
+
+	rc = bnx2_gunzip(bp, bnx2_RXP_b06FwText, sizeof(bnx2_RXP_b06FwText),
+			 &text, &text_len);
+	if (rc)
+		goto init_cpu_err;
+
+	fw.text = text;
 
 	fw.data_addr = bnx2_RXP_b06FwDataAddr;
 	fw.data_len = bnx2_RXP_b06FwDataLen;
@@ -2268,7 +2379,13 @@ bnx2_init_cpus(struct bnx2 *bp)
 	fw.text_addr = bnx2_TXP_b06FwTextAddr;
 	fw.text_len = bnx2_TXP_b06FwTextLen;
 	fw.text_index = 0;
-	fw.text = bnx2_TXP_b06FwText;
+
+	rc = bnx2_gunzip(bp, bnx2_TXP_b06FwText, sizeof(bnx2_TXP_b06FwText),
+			 &text, &text_len);
+	if (rc)
+		goto init_cpu_err;
+
+	fw.text = text;
 
 	fw.data_addr = bnx2_TXP_b06FwDataAddr;
 	fw.data_len = bnx2_TXP_b06FwDataLen;
@@ -2314,7 +2431,13 @@ bnx2_init_cpus(struct bnx2 *bp)
 	fw.text_addr = bnx2_TPAT_b06FwTextAddr;
 	fw.text_len = bnx2_TPAT_b06FwTextLen;
 	fw.text_index = 0;
-	fw.text = bnx2_TPAT_b06FwText;
+
+	rc = bnx2_gunzip(bp, bnx2_TPAT_b06FwText, sizeof(bnx2_TPAT_b06FwText),
+			 &text, &text_len);
+	if (rc)
+		goto init_cpu_err;
+
+	fw.text = text;
 
 	fw.data_addr = bnx2_TPAT_b06FwDataAddr;
 	fw.data_len = bnx2_TPAT_b06FwDataLen;
@@ -2360,7 +2483,13 @@ bnx2_init_cpus(struct bnx2 *bp)
 	fw.text_addr = bnx2_COM_b06FwTextAddr;
 	fw.text_len = bnx2_COM_b06FwTextLen;
 	fw.text_index = 0;
-	fw.text = bnx2_COM_b06FwText;
+
+	rc = bnx2_gunzip(bp, bnx2_COM_b06FwText, sizeof(bnx2_COM_b06FwText),
+			 &text, &text_len);
+	if (rc)
+		goto init_cpu_err;
+
+	fw.text = text;
 
 	fw.data_addr = bnx2_COM_b06FwDataAddr;
 	fw.data_len = bnx2_COM_b06FwDataLen;
@@ -2384,6 +2513,9 @@ bnx2_init_cpus(struct bnx2 *bp)
 
 	load_cpu_fw(bp, &cpu_reg, &fw);
 
+init_cpu_err:
+	bnx2_gunzip_end(bp);
+	return rc;
 }
 
 static int
@@ -3256,7 +3388,9 @@ bnx2_init_chip(struct bnx2 *bp)
 	 * context block must have already been enabled. */
 	bnx2_init_context(bp);
 
-	bnx2_init_cpus(bp);
+	if ((rc = bnx2_init_cpus(bp)) != 0)
+		return rc;
+
 	bnx2_init_nvram(bp);
 
 	bnx2_set_mac_addr(bp);
@@ -3556,7 +3690,9 @@ bnx2_reset_nic(struct bnx2 *bp, u32 reset_code)
 	if (rc)
 		return rc;
 
-	bnx2_init_chip(bp);
+	if ((rc = bnx2_init_chip(bp)) != 0)
+		return rc;
+
 	bnx2_init_tx_ring(bp);
 	bnx2_init_rx_ring(bp);
 	return 0;
@@ -4034,6 +4170,8 @@ bnx2_timer(unsigned long data)
 	msg = (u32) ++bp->fw_drv_pulse_wr_seq;
 	REG_WR_IND(bp, bp->shmem_base + BNX2_DRV_PULSE_MB, msg);
 
+	bp->stats_blk->stat_FwRxDrop = REG_RD_IND(bp, BNX2_FW_RX_DROP_COUNT);
+
 	if ((bp->phy_flags & PHY_SERDES_FLAG) &&
 	    (CHIP_NUM(bp) == CHIP_NUM_5706)) {
 
@@ -4252,7 +4390,7 @@ bnx2_vlan_rx_kill_vid(struct net_device *dev, uint16_t vid)
 }
 #endif
 
-/* Called with dev->xmit_lock.
+/* Called with netif_tx_lock.
  * hard_start_xmit is pseudo-lockless - a lock is only required when
  * the tx queue is full. This way, we get the benefit of lockless
  * operations most of the time without the complexities to handle
@@ -4310,7 +4448,7 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		ip_tcp_len = (skb->nh.iph->ihl << 2) + sizeof(struct tcphdr);
 
 		skb->nh.iph->check = 0;
-		skb->nh.iph->tot_len = ntohs(mss + ip_tcp_len + tcp_opt_len);
+		skb->nh.iph->tot_len = htons(mss + ip_tcp_len + tcp_opt_len);
 		skb->h.th->check =
 			~csum_tcpudp_magic(skb->nh.iph->saddr,
 					    skb->nh.iph->daddr,
@@ -4504,6 +4642,10 @@ bnx2_get_stats(struct net_device *dev)
 		net_stats->tx_aborted_errors +
 		net_stats->tx_carrier_errors;
 
+	net_stats->rx_missed_errors =
+		(unsigned long) (stats_blk->stat_IfInMBUFDiscards +
+		stats_blk->stat_FwRxDrop);
+
 	return net_stats;
 }
 
@@ -4986,7 +5128,7 @@ bnx2_set_rx_csum(struct net_device *dev, u32 data)
 	return 0;
 }
 
-#define BNX2_NUM_STATS 45
+#define BNX2_NUM_STATS 46
 
 static struct {
 	char string[ETH_GSTRING_LEN];
@@ -5036,6 +5178,7 @@ static struct {
 	{ "rx_mac_ctrl_frames" },
 	{ "rx_filtered_packets" },
 	{ "rx_discards" },
+	{ "rx_fw_discards" },
 };
 
 #define STATS_OFFSET32(offset_name) (offsetof(struct statistics_block, offset_name) / 4)
@@ -5086,6 +5229,7 @@ static const unsigned long bnx2_stats_offset_arr[BNX2_NUM_STATS] = {
     STATS_OFFSET32(stat_MacControlFramesReceived),                    
     STATS_OFFSET32(stat_IfInFramesL2FilterDiscards),                  
     STATS_OFFSET32(stat_IfInMBUFDiscards),                            
+    STATS_OFFSET32(stat_FwRxDrop),
 };
 
 /* stat_IfHCInBadOctets and stat_Dot3StatsCarrierSenseErrors are
@@ -5096,7 +5240,7 @@ static u8 bnx2_5706_stats_len_arr[BNX2_NUM_STATS] = {
 	4,0,4,4,4,4,4,4,4,4,
 	4,4,4,4,4,4,4,4,4,4,
 	4,4,4,4,4,4,4,4,4,4,
-	4,4,4,4,4,
+	4,4,4,4,4,4,
 };
 
 static u8 bnx2_5708_stats_len_arr[BNX2_NUM_STATS] = {
@@ -5104,7 +5248,7 @@ static u8 bnx2_5708_stats_len_arr[BNX2_NUM_STATS] = {
 	4,4,4,4,4,4,4,4,4,4,
 	4,4,4,4,4,4,4,4,4,4,
 	4,4,4,4,4,4,4,4,4,4,
-	4,4,4,4,4,
+	4,4,4,4,4,4,
 };
 
 #define BNX2_NUM_TESTS 6
@@ -5634,7 +5778,9 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev)
 		}
 	}
 
-	if (CHIP_NUM(bp) == CHIP_NUM_5708)
+	if ((CHIP_ID(bp) == CHIP_ID_5708_A0) ||
+	    (CHIP_ID(bp) == CHIP_ID_5708_B0) ||
+	    (CHIP_ID(bp) == CHIP_ID_5708_B1))
 		bp->flags |= NO_WOL_FLAG;
 
 	if (CHIP_ID(bp) == CHIP_ID_5706_A0) {

drivers/net/bnx2.h

@@ -231,6 +231,7 @@ struct statistics_block {
 	u32 stat_GenStat13;
 	u32 stat_GenStat14;
 	u32 stat_GenStat15;
+	u32 stat_FwRxDrop;
 };
 
 
@@ -3481,6 +3482,8 @@ struct l2_fhdr {
 
 #define BNX2_COM_SCRATCH				0x00120000
 
+#define BNX2_FW_RX_DROP_COUNT				 0x00120084
+
 
 /*
  *  cp_reg definition
@@ -3747,7 +3750,12 @@ struct l2_fhdr {
 #define DMA_READ_CHANS	5
 #define DMA_WRITE_CHANS	3
 
-#define BCM_PAGE_BITS	12
+/* Use CPU native page size up to 16K for the ring sizes.  */
+#if (PAGE_SHIFT > 14)
+#define BCM_PAGE_BITS	14
+#else
+#define BCM_PAGE_BITS	PAGE_SHIFT
+#endif
 #define BCM_PAGE_SIZE	(1 << BCM_PAGE_BITS)
 
 #define TX_DESC_CNT  (BCM_PAGE_SIZE / sizeof(struct tx_bd))
@@ -3770,7 +3778,7 @@ struct l2_fhdr {
 
 #define RX_RING_IDX(x) ((x) & bp->rx_max_ring_idx)
 
-#define RX_RING(x) (((x) & ~MAX_RX_DESC_CNT) >> 8)
+#define RX_RING(x) (((x) & ~MAX_RX_DESC_CNT) >> (BCM_PAGE_BITS - 4))
 #define RX_IDX(x) ((x) & MAX_RX_DESC_CNT)
 
 /* Context size. */
@@ -4048,6 +4056,9 @@ struct bnx2 {
 	u32			flash_size;
 
 	int			status_stats_size;
+
+	struct z_stream_s	*strm;
+	void			*gunzip_buf;
 };
 
 static u32 bnx2_reg_rd_ind(struct bnx2 *bp, u32 offset);

drivers/net/bonding/bond_main.c

@@ -1199,8 +1199,7 @@ int bond_sethwaddr(struct net_device *bond_dev, struct net_device *slave_dev)
 }
 
 #define BOND_INTERSECT_FEATURES \
-	(NETIF_F_SG|NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM|\
+	(NETIF_F_SG | NETIF_F_ALL_CSUM | NETIF_F_TSO | NETIF_F_UFO)
-	NETIF_F_TSO|NETIF_F_UFO)
 
 /* 
  * Compute the common dev->feature set available to all slaves.  Some
@@ -1218,9 +1217,7 @@ static int bond_compute_features(struct bonding *bond)
 		features &= (slave->dev->features & BOND_INTERSECT_FEATURES);
 
 	if ((features & NETIF_F_SG) && 
-	    !(features & (NETIF_F_IP_CSUM |
+	    !(features & NETIF_F_ALL_CSUM))
-			  NETIF_F_NO_CSUM |
-			  NETIF_F_HW_CSUM)))
 		features &= ~NETIF_F_SG;
 
 	/* 
@@ -4191,7 +4188,7 @@ static int bond_init(struct net_device *bond_dev, struct bond_params *params)
 	 */
 	bond_dev->features |= NETIF_F_VLAN_CHALLENGED;
 
-	/* don't acquire bond device's xmit_lock when 
+	/* don't acquire bond device's netif_tx_lock when
 	 * transmitting */
 	bond_dev->features |= NETIF_F_LLTX;
 

drivers/net/forcedeth.c

@@ -669,9 +669,9 @@ static const struct register_test nv_registers_test[] = {
  * critical parts:
  * - rx is (pseudo-) lockless: it relies on the single-threading provided
  *	by the arch code for interrupts.
- * - tx setup is lockless: it relies on dev->xmit_lock. Actual submission
+ * - tx setup is lockless: it relies on netif_tx_lock. Actual submission
  *	needs dev->priv->lock :-(
- * - set_multicast_list: preparation lockless, relies on dev->xmit_lock.
+ * - set_multicast_list: preparation lockless, relies on netif_tx_lock.
  */
 
 /* in dev: base, irq */
@@ -1405,7 +1405,7 @@ static void drain_ring(struct net_device *dev)
 
 /*
  * nv_start_xmit: dev->hard_start_xmit function
- * Called with dev->xmit_lock held.
+ * Called with netif_tx_lock held.
  */
 static int nv_start_xmit(struct sk_buff *skb, struct net_device *dev)
 {
@@ -1599,7 +1599,7 @@ static void nv_tx_done(struct net_device *dev)
 
 /*
  * nv_tx_timeout: dev->tx_timeout function
- * Called with dev->xmit_lock held.
+ * Called with netif_tx_lock held.
  */
 static void nv_tx_timeout(struct net_device *dev)
 {
@@ -1930,7 +1930,7 @@ static int nv_change_mtu(struct net_device *dev, int new_mtu)
 		 * Changing the MTU is a rare event, it shouldn't matter.
 		 */
 		nv_disable_irq(dev);
-		spin_lock_bh(&dev->xmit_lock);
+		netif_tx_lock_bh(dev);
 		spin_lock(&np->lock);
 		/* stop engines */
 		nv_stop_rx(dev);
@@ -1958,7 +1958,7 @@ static int nv_change_mtu(struct net_device *dev, int new_mtu)
 		nv_start_rx(dev);
 		nv_start_tx(dev);
 		spin_unlock(&np->lock);
-		spin_unlock_bh(&dev->xmit_lock);
+		netif_tx_unlock_bh(dev);
 		nv_enable_irq(dev);
 	}
 	return 0;
@@ -1993,7 +1993,7 @@ static int nv_set_mac_address(struct net_device *dev, void *addr)
 	memcpy(dev->dev_addr, macaddr->sa_data, ETH_ALEN);
 
 	if (netif_running(dev)) {
-		spin_lock_bh(&dev->xmit_lock);
+		netif_tx_lock_bh(dev);
 		spin_lock_irq(&np->lock);
 
 		/* stop rx engine */
@@ -2005,7 +2005,7 @@ static int nv_set_mac_address(struct net_device *dev, void *addr)
 		/* restart rx engine */
 		nv_start_rx(dev);
 		spin_unlock_irq(&np->lock);
-		spin_unlock_bh(&dev->xmit_lock);
+		netif_tx_unlock_bh(dev);
 	} else {
 		nv_copy_mac_to_hw(dev);
 	}
@@ -2014,7 +2014,7 @@ static int nv_set_mac_address(struct net_device *dev, void *addr)
 
 /*
  * nv_set_multicast: dev->set_multicast function
- * Called with dev->xmit_lock held.
+ * Called with netif_tx_lock held.
  */
 static void nv_set_multicast(struct net_device *dev)
 {

drivers/net/hamradio/6pack.c

@@ -308,9 +308,9 @@ static int sp_set_mac_address(struct net_device *dev, void *addr)
 {
 	struct sockaddr_ax25 *sa = addr;
 
-	spin_lock_irq(&dev->xmit_lock);
+	netif_tx_lock_bh(dev);
 	memcpy(dev->dev_addr, &sa->sax25_call, AX25_ADDR_LEN);
-	spin_unlock_irq(&dev->xmit_lock);
+	netif_tx_unlock_bh(dev);
 
 	return 0;
 }
@@ -767,9 +767,9 @@ static int sixpack_ioctl(struct tty_struct *tty, struct file *file,
 			break;
 		}
 
-		spin_lock_irq(&dev->xmit_lock);
+		netif_tx_lock_bh(dev);
 		memcpy(dev->dev_addr, &addr, AX25_ADDR_LEN);
-		spin_unlock_irq(&dev->xmit_lock);
+		netif_tx_unlock_bh(dev);
 
 		err = 0;
 		break;

drivers/net/hamradio/mkiss.c

@@ -357,9 +357,9 @@ static int ax_set_mac_address(struct net_device *dev, void *addr)
 {
 	struct sockaddr_ax25 *sa = addr;
 
-	spin_lock_irq(&dev->xmit_lock);
+	netif_tx_lock_bh(dev);
 	memcpy(dev->dev_addr, &sa->sax25_call, AX25_ADDR_LEN);
-	spin_unlock_irq(&dev->xmit_lock);
+	netif_tx_unlock_bh(dev);
 
 	return 0;
 }
@@ -886,9 +886,9 @@ static int mkiss_ioctl(struct tty_struct *tty, struct file *file,
 			break;
 		}
 
-		spin_lock_irq(&dev->xmit_lock);
+		netif_tx_lock_bh(dev);
 		memcpy(dev->dev_addr, addr, AX25_ADDR_LEN);
-		spin_unlock_irq(&dev->xmit_lock);
+		netif_tx_unlock_bh(dev);
 
 		err = 0;
 		break;

drivers/net/ifb.c

@@ -76,13 +76,13 @@ static void ri_tasklet(unsigned long dev)
 	dp->st_task_enter++;
 	if ((skb = skb_peek(&dp->tq)) == NULL) {
 		dp->st_txq_refl_try++;
-		if (spin_trylock(&_dev->xmit_lock)) {
+		if (netif_tx_trylock(_dev)) {
 			dp->st_rxq_enter++;
 			while ((skb = skb_dequeue(&dp->rq)) != NULL) {
 				skb_queue_tail(&dp->tq, skb);
 				dp->st_rx2tx_tran++;
 			}
-			spin_unlock(&_dev->xmit_lock);
+			netif_tx_unlock(_dev);
 		} else {
 			/* reschedule */
 			dp->st_rxq_notenter++;
@@ -110,7 +110,7 @@ static void ri_tasklet(unsigned long dev)
 		}
 	}
 
-	if (spin_trylock(&_dev->xmit_lock)) {
+	if (netif_tx_trylock(_dev)) {
 		dp->st_rxq_check++;
 		if ((skb = skb_peek(&dp->rq)) == NULL) {
 			dp->tasklet_pending = 0;
@@ -118,10 +118,10 @@ static void ri_tasklet(unsigned long dev)
 				netif_wake_queue(_dev);
 		} else {
 			dp->st_rxq_rsch++;
-			spin_unlock(&_dev->xmit_lock);
+			netif_tx_unlock(_dev);
 			goto resched;
 		}
-		spin_unlock(&_dev->xmit_lock);
+		netif_tx_unlock(_dev);
 	} else {
 resched:
 		dp->tasklet_pending = 1;

drivers/net/irda/Kconfig

@@ -417,5 +417,20 @@ config PXA_FICP
 	  available capabilities may vary from one PXA2xx target to
 	  another.
 
+config MCS_FIR
+	tristate "MosChip MCS7780 IrDA-USB dongle"
+	depends on IRDA && USB && EXPERIMENTAL
+	help
+	  Say Y or M here if you want to build support for the MosChip
+	  MCS7780 IrDA-USB bridge device driver.
+
+	  USB bridge based on the MosChip MCS7780 don't conform to the
+	  IrDA-USB device class specification, and therefore need their
+	  own specific driver. Those dongles support SIR and FIR (4Mbps)
+	  speeds.
+
+	  To compile it as a module, choose M here: the module will be called
+	  mcs7780.
+
 endmenu
 

drivers/net/irda/Makefile

@@ -19,6 +19,7 @@ obj-$(CONFIG_ALI_FIR)		+= ali-ircc.o
 obj-$(CONFIG_VLSI_FIR)		+= vlsi_ir.o
 obj-$(CONFIG_VIA_FIR)		+= via-ircc.o
 obj-$(CONFIG_PXA_FICP)	        += pxaficp_ir.o
+obj-$(CONFIG_MCS_FIR)	        += mcs7780.o
 # Old dongle drivers for old SIR drivers
 obj-$(CONFIG_ESI_DONGLE_OLD)		+= esi.o
 obj-$(CONFIG_TEKRAM_DONGLE_OLD)	+= tekram.o

drivers/net/irda/ali-ircc.c

@@ -34,14 +34,12 @@
 #include <linux/rtnetlink.h>
 #include <linux/serial_reg.h>
 #include <linux/dma-mapping.h>
+#include <linux/platform_device.h>
 
 #include <asm/io.h>
 #include <asm/dma.h>
 #include <asm/byteorder.h>
 
-#include <linux/pm.h>
-#include <linux/pm_legacy.h>
-
 #include <net/irda/wrapper.h>
 #include <net/irda/irda.h>
 #include <net/irda/irda_device.h>
@@ -51,7 +49,19 @@
 #define CHIP_IO_EXTENT 8
 #define BROKEN_DONGLE_ID
 
-static char *driver_name = "ali-ircc";
+#define ALI_IRCC_DRIVER_NAME "ali-ircc"
+
+/* Power Management */
+static int ali_ircc_suspend(struct platform_device *dev, pm_message_t state);
+static int ali_ircc_resume(struct platform_device *dev);
+
+static struct platform_driver ali_ircc_driver = {
+	.suspend	= ali_ircc_suspend,
+	.resume		= ali_ircc_resume,
+	.driver		= {
+		.name	= ALI_IRCC_DRIVER_NAME,
+	},
+};
 
 /* Module parameters */
 static int qos_mtt_bits = 0x07;  /* 1 ms or more */
@@ -97,10 +107,7 @@ static int  ali_ircc_is_receiving(struct ali_ircc_cb *self);
 static int  ali_ircc_net_open(struct net_device *dev);
 static int  ali_ircc_net_close(struct net_device *dev);
 static int  ali_ircc_net_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
-static int  ali_ircc_pmproc(struct pm_dev *dev, pm_request_t rqst, void *data);
 static void ali_ircc_change_speed(struct ali_ircc_cb *self, __u32 baud);
-static void ali_ircc_suspend(struct ali_ircc_cb *self);
-static void ali_ircc_wakeup(struct ali_ircc_cb *self);
 static struct net_device_stats *ali_ircc_net_get_stats(struct net_device *dev);
 
 /* SIR function */
@@ -146,6 +153,14 @@ static int __init ali_ircc_init(void)
 	
 	IRDA_DEBUG(2, "%s(), ---------------- Start ----------------\n", __FUNCTION__);
 
+	ret = platform_driver_register(&ali_ircc_driver);
+        if (ret) {
+                IRDA_ERROR("%s, Can't register driver!\n",
+			   ALI_IRCC_DRIVER_NAME);
+                return ret;
+        }
+
+	
 	/* Probe for all the ALi chipsets we know about */
 	for (chip= chips; chip->name; chip++, i++) 
 	{
@@ -214,6 +229,10 @@ static int __init ali_ircc_init(void)
 	}		
 		
 	IRDA_DEBUG(2, "%s(), ----------------- End -----------------\n", __FUNCTION__);					   		
+
+	if (ret)
+		platform_driver_unregister(&ali_ircc_driver);
+
 	return ret;
 }
 
@@ -229,13 +248,13 @@ static void __exit ali_ircc_cleanup(void)
 
 	IRDA_DEBUG(2, "%s(), ---------------- Start ----------------\n", __FUNCTION__);	
 
-	pm_unregister_all(ali_ircc_pmproc);
-
 	for (i=0; i < 4; i++) {
 		if (dev_self[i])
 			ali_ircc_close(dev_self[i]);
 	}
 	
+	platform_driver_unregister(&ali_ircc_driver);
+
 	IRDA_DEBUG(2, "%s(), ----------------- End -----------------\n", __FUNCTION__);
 }
 
@@ -249,7 +268,6 @@ static int ali_ircc_open(int i, chipio_t *info)
 {
 	struct net_device *dev;
 	struct ali_ircc_cb *self;
-	struct pm_dev *pmdev;
 	int dongle_id;
 	int err;
 			
@@ -284,7 +302,8 @@ static int ali_ircc_open(int i, chipio_t *info)
         self->io.fifo_size = 16;		/* SIR: 16, FIR: 32 Benjamin 2000/11/1 */
 	
 	/* Reserve the ioports that we need */
-	if (!request_region(self->io.fir_base, self->io.fir_ext, driver_name)) {
+	if (!request_region(self->io.fir_base, self->io.fir_ext,
+			    ALI_IRCC_DRIVER_NAME)) {
 		IRDA_WARNING("%s(), can't get iobase of 0x%03x\n", __FUNCTION__,
 			self->io.fir_base);
 		err = -ENODEV;
@@ -354,14 +373,11 @@ static int ali_ircc_open(int i, chipio_t *info)
 
 	/* Check dongle id */
 	dongle_id = ali_ircc_read_dongle_id(i, info);
-	IRDA_MESSAGE("%s(), %s, Found dongle: %s\n", __FUNCTION__, driver_name, dongle_types[dongle_id]);
+	IRDA_MESSAGE("%s(), %s, Found dongle: %s\n", __FUNCTION__,
+		     ALI_IRCC_DRIVER_NAME, dongle_types[dongle_id]);
 		
 	self->io.dongle_id = dongle_id;
 
-        pmdev = pm_register(PM_SYS_DEV, PM_SYS_IRDA, ali_ircc_pmproc);
-        if (pmdev)
-                pmdev->data = self;
-
 	IRDA_DEBUG(2, "%s(), ----------------- End -----------------\n", __FUNCTION__);
 	
 	return 0;
@@ -548,12 +564,11 @@ static int ali_ircc_setup(chipio_t *info)
 	/* Should be 0x00 in the M1535/M1535D */
 	if(version != 0x00)
 	{
-		IRDA_ERROR("%s, Wrong chip version %02x\n", driver_name, version);
+		IRDA_ERROR("%s, Wrong chip version %02x\n",
+			   ALI_IRCC_DRIVER_NAME, version);
 		return -1;
 	}
 	
-	// IRDA_MESSAGE("%s, Found chip at base=0x%03x\n", driver_name, info->cfg_base);
-	
 	/* Set FIR FIFO Threshold Register */
 	switch_bank(iobase, BANK1);
 	outb(RX_FIFO_Threshold, iobase+FIR_FIFO_TR);
@@ -583,7 +598,8 @@ static int ali_ircc_setup(chipio_t *info)
 	/* Switch to SIR space */
 	FIR2SIR(iobase);
 	
-	IRDA_MESSAGE("%s, driver loaded (Benjamin Kong)\n", driver_name);
+	IRDA_MESSAGE("%s, driver loaded (Benjamin Kong)\n",
+		     ALI_IRCC_DRIVER_NAME);
 	
 	/* Enable receive interrupts */ 
 	// outb(UART_IER_RDI, iobase+UART_IER); //benjamin 2000/11/23 01:25PM
@@ -647,7 +663,8 @@ static irqreturn_t ali_ircc_interrupt(int irq, void *dev_id,
 	IRDA_DEBUG(2, "%s(), ---------------- Start ----------------\n", __FUNCTION__);
 		
  	if (!dev) {
-		IRDA_WARNING("%s: irq %d for unknown device.\n", driver_name, irq);
+		IRDA_WARNING("%s: irq %d for unknown device.\n",
+			     ALI_IRCC_DRIVER_NAME, irq);
 		return IRQ_NONE;
 	}	
 	
@@ -1328,7 +1345,8 @@ static int ali_ircc_net_open(struct net_device *dev)
 	/* Request IRQ and install Interrupt Handler */
 	if (request_irq(self->io.irq, ali_ircc_interrupt, 0, dev->name, dev)) 
 	{
-		IRDA_WARNING("%s, unable to allocate irq=%d\n", driver_name,
+		IRDA_WARNING("%s, unable to allocate irq=%d\n",
+			     ALI_IRCC_DRIVER_NAME,
 			     self->io.irq);
 		return -EAGAIN;
 	}
@@ -1338,7 +1356,8 @@ static int ali_ircc_net_open(struct net_device *dev)
 	 * failure.
 	 */
 	if (request_dma(self->io.dma, dev->name)) {
-		IRDA_WARNING("%s, unable to allocate dma=%d\n", driver_name,
+		IRDA_WARNING("%s, unable to allocate dma=%d\n",
+			     ALI_IRCC_DRIVER_NAME,
 			     self->io.dma);
 		free_irq(self->io.irq, self);
 		return -EAGAIN;
@@ -2108,60 +2127,37 @@ static struct net_device_stats *ali_ircc_net_get_stats(struct net_device *dev)
 	return &self->stats;
 }
 
-static void ali_ircc_suspend(struct ali_ircc_cb *self)
+static int ali_ircc_suspend(struct platform_device *dev, pm_message_t state)
 {
-	IRDA_DEBUG(2, "%s(), ---------------- Start ----------------\n", __FUNCTION__ );
+	struct ali_ircc_cb *self = platform_get_drvdata(dev);
 	
-	IRDA_MESSAGE("%s, Suspending\n", driver_name);
+	IRDA_MESSAGE("%s, Suspending\n", ALI_IRCC_DRIVER_NAME);
 
 	if (self->io.suspended)
-		return;
+		return 0;
 
 	ali_ircc_net_close(self->netdev);
 
 	self->io.suspended = 1;
 	
-	IRDA_DEBUG(2, "%s(), ----------------- End ------------------\n", __FUNCTION__ );	
-}
-
-static void ali_ircc_wakeup(struct ali_ircc_cb *self)
-{
-	IRDA_DEBUG(2, "%s(), ---------------- Start ----------------\n", __FUNCTION__ );
-	
-	if (!self->io.suspended)
-		return;
-	
-	ali_ircc_net_open(self->netdev);
-	
-	IRDA_MESSAGE("%s, Waking up\n", driver_name);
-
-	self->io.suspended = 0;
-	
-	IRDA_DEBUG(2, "%s(), ----------------- End ------------------\n", __FUNCTION__ );	
-}
-
-static int ali_ircc_pmproc(struct pm_dev *dev, pm_request_t rqst, void *data)
-{
-        struct ali_ircc_cb *self = (struct ali_ircc_cb*) dev->data;
-        
-        IRDA_DEBUG(2, "%s(), ---------------- Start ----------------\n", __FUNCTION__ );
-	
-        if (self) {
-                switch (rqst) {
-                case PM_SUSPEND:
-                        ali_ircc_suspend(self);
-                        break;
-                case PM_RESUME:
-                        ali_ircc_wakeup(self);
-                        break;
-                }
-        }
-        
-        IRDA_DEBUG(2, "%s(), ----------------- End ------------------\n", __FUNCTION__ );	
-        
 	return 0;
 }
 
+static int ali_ircc_resume(struct platform_device *dev)
+{
+	struct ali_ircc_cb *self = platform_get_drvdata(dev);
+	
+	if (!self->io.suspended)
+		return 0;
+	
+	ali_ircc_net_open(self->netdev);
+	
+	IRDA_MESSAGE("%s, Waking up\n", ALI_IRCC_DRIVER_NAME);
+
+	self->io.suspended = 0;
+
+	return 0;
+}
 
 /* ALi Chip Function */
 

drivers/net/irda/irda-usb.c

@@ -83,9 +83,9 @@ static struct usb_device_id dongles[] = {
 	/* Extended Systems, Inc.,  XTNDAccess IrDA USB (ESI-9685) */
 	{ USB_DEVICE(0x8e9, 0x100), .driver_info = IUC_SPEED_BUG | IUC_NO_WINDOW },
 	/* SigmaTel STIR4210/4220/4116 USB IrDA (VFIR) Bridge */
-	{ USB_DEVICE(0x66f, 0x4210), .driver_info = IUC_STIR_4210 | IUC_SPEED_BUG },
+	{ USB_DEVICE(0x66f, 0x4210), .driver_info = IUC_STIR421X | IUC_SPEED_BUG },
-	{ USB_DEVICE(0x66f, 0x4220), .driver_info = IUC_STIR_4210 | IUC_SPEED_BUG },
+	{ USB_DEVICE(0x66f, 0x4220), .driver_info = IUC_STIR421X | IUC_SPEED_BUG },
-	{ USB_DEVICE(0x66f, 0x4116), .driver_info = IUC_STIR_4210 | IUC_SPEED_BUG },
+	{ USB_DEVICE(0x66f, 0x4116), .driver_info = IUC_STIR421X | IUC_SPEED_BUG },
 	{ .match_flags = USB_DEVICE_ID_MATCH_INT_CLASS |
 	  USB_DEVICE_ID_MATCH_INT_SUBCLASS,
 	  .bInterfaceClass = USB_CLASS_APP_SPEC,
@@ -154,7 +154,7 @@ static void irda_usb_build_header(struct irda_usb_cb *self,
 	 * and if either speed or xbofs (or both) needs
 	 * to be changed.
 	 */
-	if (self->capability & IUC_STIR_4210 &&
+	if (self->capability & IUC_STIR421X &&
 	    ((self->new_speed != -1) || (self->new_xbofs != -1))) {
 
 		/* With STIR421x, speed and xBOFs must be set at the same
@@ -318,7 +318,7 @@ static void irda_usb_change_speed_xbofs(struct irda_usb_cb *self)
 	/* Set the new speed and xbofs in this fake frame */
 	irda_usb_build_header(self, frame, 1);
 
-	if ( self->capability & IUC_STIR_4210 ) {
+	if (self->capability & IUC_STIR421X) {
 		if (frame[0] == 0) return ; // do nothing if no change
 		frame[1] = 0; // other parameters don't change here
 		frame[2] = 0;
@@ -455,7 +455,7 @@ static int irda_usb_hard_xmit(struct sk_buff *skb, struct net_device *netdev)
 
 	/* Change setting for next frame */
 
-	if ( self->capability & IUC_STIR_4210 ) {
+	if (self->capability & IUC_STIR421X) {
 		__u8 turnaround_time;
 		__u8* frame;
 		turnaround_time = get_turnaround_time( skb );
@@ -897,10 +897,13 @@ static void irda_usb_receive(struct urb *urb, struct pt_regs *regs)
 	docopy = (urb->actual_length < IRDA_RX_COPY_THRESHOLD);
 
 	/* Allocate a new skb */
-	if ( self->capability & IUC_STIR_4210 )
+	if (self->capability & IUC_STIR421X)
-		newskb = dev_alloc_skb(docopy ? urb->actual_length : IRDA_SKB_MAX_MTU + USB_IRDA_SIGMATEL_HEADER);
+		newskb = dev_alloc_skb(docopy ? urb->actual_length :
+				       IRDA_SKB_MAX_MTU +
+				       USB_IRDA_STIR421X_HEADER);
 	else
-		newskb = dev_alloc_skb(docopy ? urb->actual_length : IRDA_SKB_MAX_MTU);
+		newskb = dev_alloc_skb(docopy ? urb->actual_length :
+				       IRDA_SKB_MAX_MTU);
 
 	if (!newskb)  {
 		self->stats.rx_dropped++;
@@ -1022,188 +1025,140 @@ static int irda_usb_is_receiving(struct irda_usb_cb *self)
 	return 0; /* For now */
 }
 
-
+#define STIR421X_PATCH_PRODUCT_VER     "Product Version: "
-#define STIR421X_PATCH_PRODUCT_VERSION_STR       "Product Version: "
+#define STIR421X_PATCH_STMP_TAG        "STMP"
-#define STIR421X_PATCH_COMPONENT_VERSION_STR     "Component Version: "
+#define STIR421X_PATCH_CODE_OFFSET     512 /* patch image starts before here */
-#define STIR421X_PATCH_DATA_TAG_STR              "STMP"
+/* marks end of patch file header (PC DOS text file EOF character) */
-#define STIR421X_PATCH_FILE_VERSION_MAX_OFFSET   512     /* version info is before here */
+#define STIR421X_PATCH_END_OF_HDR_TAG  0x1A
-#define STIR421X_PATCH_FILE_IMAGE_MAX_OFFSET     512     /* patch image starts before here */
+#define STIR421X_PATCH_BLOCK_SIZE      1023
-#define STIR421X_PATCH_FILE_END_OF_HEADER_TAG    0x1A    /* marks end of patch file header (PC DOS text file EOF character) */
 
 /*
- * Known firmware patches for STIR421x dongles
+ * Function stir421x_fwupload (struct irda_usb_cb *self,
+ *                             unsigned char *patch,
+ *                             const unsigned int patch_len)
+ *
+ *   Upload firmware code to SigmaTel 421X IRDA-USB dongle
  */
-static char * stir421x_patches[] = {
+static int stir421x_fw_upload(struct irda_usb_cb *self,
-	"42101001.sb",
+			     unsigned char *patch,
-	"42101002.sb",
-};
-
-static int stir421x_get_patch_version(unsigned char * patch, const unsigned long patch_len)
-{
-	unsigned int version_offset;
-	unsigned long version_major, version_minor, version_build;
-	unsigned char * version_start;
-	int version_found = 0;
-
-	for (version_offset = 0;
-	     version_offset < STIR421X_PATCH_FILE_END_OF_HEADER_TAG;
-	     version_offset++) {
-		if (!memcmp(patch + version_offset,
-			    STIR421X_PATCH_PRODUCT_VERSION_STR,
-			    sizeof(STIR421X_PATCH_PRODUCT_VERSION_STR) - 1)) {
-				    version_found = 1;
-				    version_start = patch +
-					    version_offset +
-					    sizeof(STIR421X_PATCH_PRODUCT_VERSION_STR) - 1;
-				    break;
-		}
-	}
-
-	/* We couldn't find a product version on this patch */
-	if (!version_found)
-		return -EINVAL;
-
-	/* Let's check if the product version is dotted */
-	if (version_start[3] != '.' ||
-	    version_start[7] != '.')
-		return -EINVAL;
-
-	version_major = simple_strtoul(version_start, NULL, 10);
-	version_minor = simple_strtoul(version_start + 4, NULL, 10);
-	version_build = simple_strtoul(version_start + 8, NULL, 10);
-
-	IRDA_DEBUG(2, "%s(), Major: %ld Minor: %ld Build: %ld\n",
-		   __FUNCTION__,
-		   version_major, version_minor, version_build);
-
-	return (((version_major) << 12) +
-		((version_minor) << 8) +
-		((version_build / 10) << 4) +
-		(version_build % 10));
-
-}
-
-
-static int stir421x_upload_patch (struct irda_usb_cb *self,
-				  unsigned char * patch,
 			     const unsigned int patch_len)
 {
-    int retval = 0;
+        int ret = -ENOMEM;
-    int actual_len;
+        int actual_len = 0;
-    unsigned int i = 0, download_amount = 0;
+        unsigned int i;
-    unsigned char * patch_chunk;
+        unsigned int block_size = 0;
+        unsigned char *patch_block;
 
-    IRDA_DEBUG (2, "%s(), Uploading STIR421x Patch\n", __FUNCTION__);
+        patch_block = kzalloc(STIR421X_PATCH_BLOCK_SIZE, GFP_KERNEL);
-
+	if (patch_block == NULL)
-    patch_chunk = kzalloc(STIR421X_MAX_PATCH_DOWNLOAD_SIZE, GFP_KERNEL);
-    if (patch_chunk == NULL)
 		return -ENOMEM;
 
 	/* break up patch into 1023-byte sections */
-    for (i = 0; retval >= 0 && i < patch_len; i += download_amount) {
+	for (i = 0; i < patch_len; i += block_size) {
-	    download_amount = patch_len - i;
+		block_size = patch_len - i;
-	    if (download_amount > STIR421X_MAX_PATCH_DOWNLOAD_SIZE)
-		    download_amount = STIR421X_MAX_PATCH_DOWNLOAD_SIZE;
 
-	    /* download the patch section */
+		if (block_size > STIR421X_PATCH_BLOCK_SIZE)
-	    memcpy(patch_chunk, patch + i, download_amount);
+			block_size = STIR421X_PATCH_BLOCK_SIZE;
 
-	    retval = usb_bulk_msg (self->usbdev,
+		/* upload the patch section */
-				   usb_sndbulkpipe (self->usbdev,
+		memcpy(patch_block, patch + i, block_size);
+
+		ret = usb_bulk_msg(self->usbdev,
+				   usb_sndbulkpipe(self->usbdev,
 						   self->bulk_out_ep),
-				   patch_chunk, download_amount,
+				   patch_block, block_size,
-				   &actual_len, msecs_to_jiffies (500));
+				   &actual_len, msecs_to_jiffies(500));
-	    IRDA_DEBUG (2, "%s(), Sent %u bytes\n", __FUNCTION__,
+		IRDA_DEBUG(3,"%s(): Bulk send %u bytes, ret=%d\n",
-			actual_len);
+			   __FUNCTION__, actual_len, ret);
-	    if (retval == 0)
+
-		    mdelay(10);
+		if (ret < 0)
+			break;
 	}
 
-    kfree(patch_chunk);
+	kfree(patch_block);
 
-    if (i != patch_len) {
+        return ret;
-	    IRDA_ERROR ("%s(), Pushed %d bytes (!= patch_len (%d))\n",
-		       __FUNCTION__, i, patch_len);
-	    retval = -EIO;
  }
 
-    if (retval < 0)
+/*
-	    /* todo - mark device as not ready */
+ * Function stir421x_patch_device(struct irda_usb_cb *self)
-	    IRDA_ERROR ("%s(), STIR421x patch upload failed (%d)\n",
+ *
-			__FUNCTION__, retval);
+ * Get a firmware code from userspase using hotplug request_firmware() call
-
+  */
-    return retval;
-}
-
-
 static int stir421x_patch_device(struct irda_usb_cb *self)
 {
-	unsigned int i, patch_found = 0, data_found = 0, data_offset;
+        unsigned int i;
-	int patch_version, ret = 0;
+        int ret;
-	const struct firmware *fw_entry;
+        char stir421x_fw_name[11];
+        const struct firmware *fw;
+        unsigned char *fw_version_ptr; /* pointer to version string */
+	unsigned long fw_version = 0;
 
-	for (i = 0; i < ARRAY_SIZE(stir421x_patches); i++) {
+        /*
-		if(request_firmware(&fw_entry, stir421x_patches[i], &self->usbdev->dev) != 0) {
+         * Known firmware patch file names for STIR421x dongles
-			IRDA_ERROR( "%s(), Patch %s is not available\n", __FUNCTION__, stir421x_patches[i]);
+         * are "42101001.sb" or "42101002.sb"
-			continue;
+         */
-		}
+        sprintf(stir421x_fw_name, "4210%4X.sb",
+                self->usbdev->descriptor.bcdDevice);
+        ret = request_firmware(&fw, stir421x_fw_name, &self->usbdev->dev);
+        if (ret < 0)
+                return ret;
 
-                /* We found a patch from userspace */
+        /* We get a patch from userspace */
-		patch_version = stir421x_get_patch_version (fw_entry->data, fw_entry->size);
+        IRDA_MESSAGE("%s(): Received firmware %s (%u bytes)\n",
+                     __FUNCTION__, stir421x_fw_name, fw->size);
 
-		if (patch_version < 0) {
-			/* Couldn't fetch a version, let's move on to the next file */
-			IRDA_ERROR("%s(), version parsing failed\n", __FUNCTION__);
-			ret = patch_version;
-			release_firmware(fw_entry);
-			continue;
-		}
-
-		if (patch_version != self->usbdev->descriptor.bcdDevice) {
-			/* Patch version and device don't match */
-			IRDA_ERROR ("%s(), wrong patch version (%d <-> %d)\n",
-				    __FUNCTION__,
-				    patch_version, self->usbdev->descriptor.bcdDevice);
         ret = -EINVAL;
-			release_firmware(fw_entry);
-			continue;
-		}
 
-		/* If we're here, we've found a correct patch */
+	/* Get the bcd product version */
-		patch_found = 1;
+        if (!memcmp(fw->data, STIR421X_PATCH_PRODUCT_VER,
-		break;
+                    sizeof(STIR421X_PATCH_PRODUCT_VER) - 1)) {
+                fw_version_ptr = fw->data +
+			sizeof(STIR421X_PATCH_PRODUCT_VER) - 1;
 
-	}
+                /* Let's check if the product version is dotted */
+                if (fw_version_ptr[3] == '.' &&
+		    fw_version_ptr[7] == '.') {
+			unsigned long major, minor, build;
+			major = simple_strtoul(fw_version_ptr, NULL, 10);
+			minor = simple_strtoul(fw_version_ptr + 4, NULL, 10);
+			build = simple_strtoul(fw_version_ptr + 8, NULL, 10);
 
-	/* We couldn't find a valid firmware, let's leave */
+			fw_version = (major << 12)
-	if (!patch_found)
+				+ (minor << 8)
-		return ret;
+				+ ((build / 10) << 4)
+				+ (build % 10);
 
-	/* The actual image starts after the "STMP" keyword */
+			IRDA_DEBUG(3, "%s(): Firmware Product version %ld\n",
-	for (data_offset = 0; data_offset < STIR421X_PATCH_FILE_IMAGE_MAX_OFFSET; data_offset++) {
+                                   __FUNCTION__, fw_version);
-		if (!memcmp(fw_entry->data + data_offset,
-			    STIR421X_PATCH_DATA_TAG_STR,
-			    sizeof(STIR421X_PATCH_FILE_IMAGE_MAX_OFFSET))) {
-			IRDA_DEBUG(2, "%s(), found patch data for STIR421x at offset %d\n",
-				   __FUNCTION__, data_offset);
-			data_found = 1;
-			break;
                 }
         }
 
-	/* We couldn't find "STMP" from the header */
+        if (self->usbdev->descriptor.bcdDevice == fw_version) {
-	if (!data_found)
+                /*
-		return -EINVAL;
+		 * If we're here, we've found a correct patch
+                 * The actual image starts after the "STMP" keyword
+                 * so forward to the firmware header tag
+                 */
+                for (i = 0; (fw->data[i] != STIR421X_PATCH_END_OF_HDR_TAG)
+			     && (i < fw->size); i++) ;
+                /* here we check for the out of buffer case */
+                if ((STIR421X_PATCH_END_OF_HDR_TAG == fw->data[i])
+                    && (i < STIR421X_PATCH_CODE_OFFSET)) {
+                        if (!memcmp(fw->data + i + 1, STIR421X_PATCH_STMP_TAG,
+                                    sizeof(STIR421X_PATCH_STMP_TAG) - 1)) {
 
-	/* Let's upload the patch to the target */
+				/* We can upload the patch to the target */
-	ret = stir421x_upload_patch(self,
+				i += sizeof(STIR421X_PATCH_STMP_TAG);
-				    &fw_entry->data[data_offset + sizeof(STIR421X_PATCH_FILE_IMAGE_MAX_OFFSET)],
+                                ret = stir421x_fw_upload(self, &fw->data[i],
-				    fw_entry->size - (data_offset + sizeof(STIR421X_PATCH_FILE_IMAGE_MAX_OFFSET)));
+							 fw->size - i);
+                        }
+                }
+        }
 
-	release_firmware(fw_entry);
+        release_firmware(fw);
 
         return ret;
-
 }
 
 
@@ -1702,12 +1657,12 @@ static int irda_usb_probe(struct usb_interface *intf,
 	init_timer(&self->rx_defer_timer);
 
 	self->capability = id->driver_info;
-	self->needspatch = ((self->capability & IUC_STIR_4210) != 0) ;
+	self->needspatch = ((self->capability & IUC_STIR421X) != 0);
 
 	/* Create all of the needed urbs */
-	if (self->capability & IUC_STIR_4210) {
+	if (self->capability & IUC_STIR421X) {
 		self->max_rx_urb = IU_SIGMATEL_MAX_RX_URBS;
-		self->header_length = USB_IRDA_SIGMATEL_HEADER;
+		self->header_length = USB_IRDA_STIR421X_HEADER;
 	} else {
 		self->max_rx_urb = IU_MAX_RX_URBS;
 		self->header_length = USB_IRDA_HEADER;
@@ -1813,8 +1768,8 @@ static int irda_usb_probe(struct usb_interface *intf,
 		/* Now we fetch and upload the firmware patch */
 		ret = stir421x_patch_device(self);
 		self->needspatch = (ret < 0);
-		if (ret < 0) {
+		if (self->needspatch) {
-			printk("patch_device failed\n");
+			IRDA_ERROR("STIR421X: Couldn't upload patch\n");
 			goto err_out_5;
 		}
 

drivers/net/irda/irda-usb.h

@@ -34,9 +34,6 @@
 #include <net/irda/irda.h>
 #include <net/irda/irda_device.h>      /* struct irlap_cb */
 
-#define PATCH_FILE_SIZE_MAX     65536
-#define PATCH_FILE_SIZE_MIN     80
-
 #define RX_COPY_THRESHOLD 200
 #define IRDA_USB_MAX_MTU 2051
 #define IRDA_USB_SPEED_MTU 64		/* Weird, but work like this */
@@ -107,14 +104,15 @@
 #define IUC_SMALL_PKT	0x10	/* Device doesn't behave with big Rx packets */
 #define IUC_MAX_WINDOW	0x20	/* Device underestimate the Rx window */
 #define IUC_MAX_XBOFS	0x40	/* Device need more xbofs than advertised */
-#define IUC_STIR_4210	0x80	/* SigmaTel 4210/4220/4116 VFIR */
+#define IUC_STIR421X	0x80	/* SigmaTel 4210/4220/4116 VFIR */
 
 /* USB class definitions */
 #define USB_IRDA_HEADER            0x01
 #define USB_CLASS_IRDA             0x02 /* USB_CLASS_APP_SPEC subclass */
 #define USB_DT_IRDA                0x21
-#define USB_IRDA_SIGMATEL_HEADER   0x03
+#define USB_IRDA_STIR421X_HEADER   0x03
-#define IU_SIGMATEL_MAX_RX_URBS    (IU_MAX_ACTIVE_RX_URBS + USB_IRDA_SIGMATEL_HEADER)
+#define IU_SIGMATEL_MAX_RX_URBS    (IU_MAX_ACTIVE_RX_URBS + \
+                                    USB_IRDA_STIR421X_HEADER)
 
 struct irda_class_desc {
 	__u8  bLength;

drivers/net/irda/mcs7780.h

@@ -0,0 +1,167 @@
+/*****************************************************************************
+*
+* Filename:      mcs7780.h
+* Version:       0.2-alpha
+* Description:   Irda MosChip USB Dongle
+* Status:        Experimental
+* Authors:       Lukasz Stelmach <stlman@poczta.fm>
+*		 Brian Pugh <bpugh@cs.pdx.edu>
+*
+*       Copyright (C) 2005, Lukasz Stelmach <stlman@poczta.fm>
+*       Copyright (C) 2005, Brian Pugh <bpugh@cs.pdx.edu>
+*
+*       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.
+*
+*       This program is distributed in the hope that it will be useful,
+*       but WITHOUT ANY WARRANTY; without even the implied warranty of
+*       MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+*       GNU General Public License for more details.
+*
+*       You should have received a copy of the GNU General Public License
+*       along with this program; if not, write to the Free Software
+*       Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+*
+*****************************************************************************/
+#ifndef _MCS7780_H
+#define _MCS7780_H
+
+#define MCS_MODE_SIR		0
+#define MCS_MODE_MIR		1
+#define MCS_MODE_FIR		2
+
+#define MCS_CTRL_TIMEOUT	500
+#define MCS_XMIT_TIMEOUT	500
+/* Possible transceiver types */
+#define MCS_TSC_VISHAY		0	/* Vishay TFD, default choice */
+#define MCS_TSC_AGILENT		1	/* Agilent 3602/3600 */
+#define MCS_TSC_SHARP		2	/* Sharp GP2W1000YP */
+
+/* Requests */
+#define MCS_RD_RTYPE 0xC0
+#define MCS_WR_RTYPE 0x40
+#define MCS_RDREQ    0x0F
+#define MCS_WRREQ    0x0E
+
+/* Register 0x00 */
+#define MCS_MODE_REG	0
+#define MCS_FIR		((__u16)0x0001)
+#define MCS_SIR16US	((__u16)0x0002)
+#define MCS_BBTG	((__u16)0x0004)
+#define MCS_ASK		((__u16)0x0008)
+#define MCS_PARITY	((__u16)0x0010)
+
+/* SIR/MIR speed constants */
+#define MCS_SPEED_SHIFT	    5
+#define MCS_SPEED_MASK	    ((__u16)0x00E0)
+#define MCS_SPEED(x)	    ((x & MCS_SPEED_MASK) >> MCS_SPEED_SHIFT)
+#define MCS_SPEED_2400	    ((0 << MCS_SPEED_SHIFT) & MCS_SPEED_MASK)
+#define MCS_SPEED_9600	    ((1 << MCS_SPEED_SHIFT) & MCS_SPEED_MASK)
+#define MCS_SPEED_19200	    ((2 << MCS_SPEED_SHIFT) & MCS_SPEED_MASK)
+#define MCS_SPEED_38400	    ((3 << MCS_SPEED_SHIFT) & MCS_SPEED_MASK)
+#define MCS_SPEED_57600	    ((4 << MCS_SPEED_SHIFT) & MCS_SPEED_MASK)
+#define MCS_SPEED_115200    ((5 << MCS_SPEED_SHIFT) & MCS_SPEED_MASK)
+#define MCS_SPEED_576000    ((6 << MCS_SPEED_SHIFT) & MCS_SPEED_MASK)
+#define MCS_SPEED_1152000   ((7 << MCS_SPEED_SHIFT) & MCS_SPEED_MASK)
+
+#define MCS_PLLPWDN	((__u16)0x0100)
+#define MCS_DRIVER	((__u16)0x0200)
+#define MCS_DTD		((__u16)0x0400)
+#define MCS_DIR		((__u16)0x0800)
+#define MCS_SIPEN	((__u16)0x1000)
+#define MCS_SENDSIP	((__u16)0x2000)
+#define MCS_CHGDIR	((__u16)0x4000)
+#define MCS_RESET	((__u16)0x8000)
+
+/* Register 0x02 */
+#define MCS_XCVR_REG	2
+#define MCS_MODE0	((__u16)0x0001)
+#define MCS_STFIR	((__u16)0x0002)
+#define MCS_XCVR_CONF	((__u16)0x0004)
+#define MCS_RXFAST	((__u16)0x0008)
+/* TXCUR [6:4] */
+#define MCS_TXCUR_SHIFT	4
+#define MCS_TXCUR_MASK	((__u16)0x0070)
+#define MCS_TXCUR(x)	((x & MCS_TXCUR_MASK) >> MCS_TXCUR_SHIFT)
+#define MCS_SETTXCUR(x,y) \
+	((x & ~MCS_TXCUR_MASK) | (y << MCS_TXCUR_SHIFT) & MCS_TXCUR_MASK)
+
+#define MCS_MODE1	((__u16)0x0080)
+#define MCS_SMODE0	((__u16)0x0100)
+#define MCS_SMODE1	((__u16)0x0200)
+#define MCS_INVTX	((__u16)0x0400)
+#define MCS_INVRX	((__u16)0x0800)
+
+#define MCS_MINRXPW_REG	4
+
+#define MCS_RESV_REG 7
+#define MCS_IRINTX	((__u16)0x0001)
+#define MCS_IRINRX	((__u16)0x0002)
+
+struct mcs_cb {
+	struct usb_device *usbdev;	/* init: probe_irda */
+	struct net_device *netdev;	/* network layer */
+	struct irlap_cb *irlap;	/* The link layer we are binded to */
+	struct net_device_stats stats;	/* network statistics */
+	struct qos_info qos;
+	unsigned int speed;	/* Current speed */
+	unsigned int new_speed;	/* new speed */
+
+	struct work_struct work; /* Change speed work */
+
+	struct sk_buff *tx_pending;
+	char in_buf[4096];	/* transmit/receive buffer */
+	char out_buf[4096];	/* transmit/receive buffer */
+	__u8 *fifo_status;
+
+	iobuff_t rx_buff;	/* receive unwrap state machine */
+	struct timeval rx_time;
+	spinlock_t lock;
+	int receiving;
+
+	__u8 ep_in;
+	__u8 ep_out;
+
+	struct urb *rx_urb;
+	struct urb *tx_urb;
+
+	int transceiver_type;
+	int sir_tweak;
+	int receive_mode;
+};
+
+static int mcs_set_reg(struct mcs_cb *mcs, __u16 reg, __u16 val);
+static int mcs_get_reg(struct mcs_cb *mcs, __u16 reg, __u16 * val);
+
+static inline int mcs_setup_transceiver_vishay(struct mcs_cb *mcs);
+static inline int mcs_setup_transceiver_agilent(struct mcs_cb *mcs);
+static inline int mcs_setup_transceiver_sharp(struct mcs_cb *mcs);
+static inline int mcs_setup_transceiver(struct mcs_cb *mcs);
+static inline int mcs_wrap_sir_skb(struct sk_buff *skb, __u8 * buf);
+static unsigned mcs_wrap_fir_skb(const struct sk_buff *skb, __u8 *buf);
+static unsigned mcs_wrap_mir_skb(const struct sk_buff *skb, __u8 *buf);
+static void mcs_unwrap_mir(struct mcs_cb *mcs, __u8 *buf, int len);
+static void mcs_unwrap_fir(struct mcs_cb *mcs, __u8 *buf, int len);
+static inline int mcs_setup_urbs(struct mcs_cb *mcs);
+static inline int mcs_receive_start(struct mcs_cb *mcs);
+static inline int mcs_find_endpoints(struct mcs_cb *mcs,
+				     struct usb_host_endpoint *ep, int epnum);
+
+static int mcs_speed_change(struct mcs_cb *mcs);
+
+static int mcs_net_ioctl(struct net_device *netdev, struct ifreq *rq, int cmd);
+static int mcs_net_close(struct net_device *netdev);
+static int mcs_net_open(struct net_device *netdev);
+static struct net_device_stats *mcs_net_get_stats(struct net_device *netdev);
+
+static void mcs_receive_irq(struct urb *urb, struct pt_regs *regs);
+static void mcs_send_irq(struct urb *urb, struct pt_regs *regs);
+static int mcs_hard_xmit(struct sk_buff *skb, struct net_device *netdev);
+
+static int mcs_probe(struct usb_interface *intf,
+		     const struct usb_device_id *id);
+static void mcs_disconnect(struct usb_interface *intf);
+
+#endif				/* _MCS7780_H */

drivers/net/irda/stir4200.c

@@ -50,6 +50,7 @@
 #include <linux/delay.h>
 #include <linux/usb.h>
 #include <linux/crc32.h>
+#include <linux/kthread.h>
 #include <net/irda/irda.h>
 #include <net/irda/irlap.h>
 #include <net/irda/irda_device.h>
@@ -173,9 +174,7 @@ struct stir_cb {
         struct qos_info   qos;
 	unsigned 	  speed;	/* Current speed */
 
-	wait_queue_head_t thr_wait;	/* transmit thread wakeup */
+        struct task_struct *thread;     /* transmit thread */
-	struct completion thr_exited;
-	pid_t		  thr_pid;
 
 	struct sk_buff	  *tx_pending;
 	void		  *io_buf;	/* transmit/receive buffer */
@@ -577,7 +576,7 @@ static int stir_hard_xmit(struct sk_buff *skb, struct net_device *netdev)
 	SKB_LINEAR_ASSERT(skb);
 
 	skb = xchg(&stir->tx_pending, skb);
-	wake_up(&stir->thr_wait);
+        wake_up_process(stir->thread);
 	
 	/* this should never happen unless stop/wakeup problem */
 	if (unlikely(skb)) {
@@ -753,13 +752,7 @@ static int stir_transmit_thread(void *arg)
 	struct net_device *dev = stir->netdev;
 	struct sk_buff *skb;
 
-	daemonize("%s", dev->name);
+        while (!kthread_should_stop()) {
-	allow_signal(SIGTERM);
-
-	while (netif_running(dev)
-	       && netif_device_present(dev)
-	       && !signal_pending(current))
-	{
 #ifdef CONFIG_PM
 		/* if suspending, then power off and wait */
 		if (unlikely(freezing(current))) {
@@ -813,10 +806,11 @@ static int stir_transmit_thread(void *arg)
 		}
 
 		/* sleep if nothing to send */
-		wait_event_interruptible(stir->thr_wait, stir->tx_pending);
+                set_current_state(TASK_INTERRUPTIBLE);
-	}
+                schedule();
 
-	complete_and_exit (&stir->thr_exited, 0);
+	}
+        return 0;
 }
 
 
@@ -859,7 +853,7 @@ static void stir_rcv_irq(struct urb *urb, struct pt_regs *regs)
 		warn("%s: usb receive submit error: %d",
 			stir->netdev->name, err);
 		stir->receiving = 0;
-		wake_up(&stir->thr_wait);
+		wake_up_process(stir->thread);
 	}
 }
 
@@ -928,10 +922,10 @@ static int stir_net_open(struct net_device *netdev)
 	}
 
 	/** Start kernel thread for transmit.  */
-	stir->thr_pid = kernel_thread(stir_transmit_thread, stir,
+	stir->thread = kthread_run(stir_transmit_thread, stir,
-				      CLONE_FS|CLONE_FILES);
+				   "%s", stir->netdev->name);
-	if (stir->thr_pid < 0) {
+        if (IS_ERR(stir->thread)) {
-		err = stir->thr_pid;
+                err = PTR_ERR(stir->thread);
 		err("stir4200: unable to start kernel thread");
 		goto err_out6;
 	}
@@ -968,8 +962,7 @@ static int stir_net_close(struct net_device *netdev)
 	netif_stop_queue(netdev);
 
 	/* Kill transmit thread */
-	kill_proc(stir->thr_pid, SIGTERM, 1);
+	kthread_stop(stir->thread);
-	wait_for_completion(&stir->thr_exited);
 	kfree(stir->fifo_status);
 
 	/* Mop up receive urb's */
@@ -1084,9 +1077,6 @@ static int stir_probe(struct usb_interface *intf,
 	stir->qos.min_turn_time.bits   &= qos_mtt_bits;
 	irda_qos_bits_to_value(&stir->qos);
 
-	init_completion (&stir->thr_exited);
-	init_waitqueue_head (&stir->thr_wait);
-
 	/* Override the network functions we need to use */
 	net->hard_start_xmit = stir_hard_xmit;
 	net->open            = stir_net_open;

drivers/net/irda/vlsi_ir.c

@@ -959,7 +959,7 @@ static int vlsi_hard_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 			    ||  (now.tv_sec==ready.tv_sec && now.tv_usec>=ready.tv_usec))
 			    	break;
 			udelay(100);
-			/* must not sleep here - we are called under xmit_lock! */
+			/* must not sleep here - called under netif_tx_lock! */
 		}
 	}
 

drivers/net/mv643xx_eth.c

@@ -1200,7 +1200,7 @@ static int mv643xx_eth_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	}
 
 	if (has_tiny_unaligned_frags(skb)) {
-		if ((skb_linearize(skb, GFP_ATOMIC) != 0)) {
+		if (__skb_linearize(skb)) {
 			stats->tx_dropped++;
 			printk(KERN_DEBUG "%s: failed to linearize tiny "
 					"unaligned fragment\n", dev->name);

drivers/net/natsemi.c

@@ -318,12 +318,12 @@ performance critical codepaths:
 The rx process only runs in the interrupt handler. Access from outside
 the interrupt handler is only permitted after disable_irq().
 
-The rx process usually runs under the dev->xmit_lock. If np->intr_tx_reap
+The rx process usually runs under the netif_tx_lock. If np->intr_tx_reap
 is set, then access is permitted under spin_lock_irq(&np->lock).
 
 Thus configuration functions that want to access everything must call
 	disable_irq(dev->irq);
-	spin_lock_bh(dev->xmit_lock);
+	netif_tx_lock_bh(dev);
 	spin_lock_irq(&np->lock);
 
 IV. Notes

drivers/net/ppp_generic.c

@@ -1609,8 +1609,6 @@ ppp_receive_nonmp_frame(struct ppp *ppp, struct sk_buff *skb)
 			kfree_skb(skb);
 			skb = ns;
 		}
-		else if (!pskb_may_pull(skb, skb->len))
-			goto err;
 		else
 			skb->ip_summed = CHECKSUM_NONE;
 

drivers/net/tg3.c

@@ -69,8 +69,8 @@
 
 #define DRV_MODULE_NAME		"tg3"
 #define PFX DRV_MODULE_NAME	": "
-#define DRV_MODULE_VERSION	"3.59"
+#define DRV_MODULE_VERSION	"3.60"
-#define DRV_MODULE_RELDATE	"June 8, 2006"
+#define DRV_MODULE_RELDATE	"June 17, 2006"
 
 #define TG3_DEF_MAC_MODE	0
 #define TG3_DEF_RX_MODE		0
@@ -229,6 +229,8 @@ static struct pci_device_id tg3_pci_tbl[] = {
 	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL },
 	{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5755M,
 	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL },
+	{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5786,
+	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL },
 	{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5787,
 	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL },
 	{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5787M,
@@ -2965,6 +2967,27 @@ static int tg3_setup_phy(struct tg3 *tp, int force_reset)
 	return err;
 }
 
+/* This is called whenever we suspect that the system chipset is re-
+ * ordering the sequence of MMIO to the tx send mailbox. The symptom
+ * is bogus tx completions. We try to recover by setting the
+ * TG3_FLAG_MBOX_WRITE_REORDER flag and resetting the chip later
+ * in the workqueue.
+ */
+static void tg3_tx_recover(struct tg3 *tp)
+{
+	BUG_ON((tp->tg3_flags & TG3_FLAG_MBOX_WRITE_REORDER) ||
+	       tp->write32_tx_mbox == tg3_write_indirect_mbox);
+
+	printk(KERN_WARNING PFX "%s: The system may be re-ordering memory-"
+	       "mapped I/O cycles to the network device, attempting to "
+	       "recover. Please report the problem to the driver maintainer "
+	       "and include system chipset information.\n", tp->dev->name);
+
+	spin_lock(&tp->lock);
+	tp->tg3_flags |= TG3_FLAG_TX_RECOVERY_PENDING;
+	spin_unlock(&tp->lock);
+}
+
 /* Tigon3 never reports partial packet sends.  So we do not
  * need special logic to handle SKBs that have not had all
  * of their frags sent yet, like SunGEM does.
@@ -2977,9 +3000,13 @@ static void tg3_tx(struct tg3 *tp)
 	while (sw_idx != hw_idx) {
 		struct tx_ring_info *ri = &tp->tx_buffers[sw_idx];
 		struct sk_buff *skb = ri->skb;
-		int i;
+		int i, tx_bug = 0;
+
+		if (unlikely(skb == NULL)) {
+			tg3_tx_recover(tp);
+			return;
+		}
 
-		BUG_ON(skb == NULL);
 		pci_unmap_single(tp->pdev,
 				 pci_unmap_addr(ri, mapping),
 				 skb_headlen(skb),
@@ -2990,10 +3017,9 @@ static void tg3_tx(struct tg3 *tp)
 		sw_idx = NEXT_TX(sw_idx);
 
 		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
-			BUG_ON(sw_idx == hw_idx);
-
 			ri = &tp->tx_buffers[sw_idx];
-			BUG_ON(ri->skb != NULL);
+			if (unlikely(ri->skb != NULL || sw_idx == hw_idx))
+				tx_bug = 1;
 
 			pci_unmap_page(tp->pdev,
 				       pci_unmap_addr(ri, mapping),
@@ -3004,6 +3030,11 @@ static void tg3_tx(struct tg3 *tp)
 		}
 
 		dev_kfree_skb(skb);
+
+		if (unlikely(tx_bug)) {
+			tg3_tx_recover(tp);
+			return;
+		}
 	}
 
 	tp->tx_cons = sw_idx;
@@ -3331,6 +3362,11 @@ static int tg3_poll(struct net_device *netdev, int *budget)
 	/* run TX completion thread */
 	if (sblk->idx[0].tx_consumer != tp->tx_cons) {
 		tg3_tx(tp);
+		if (unlikely(tp->tg3_flags & TG3_FLAG_TX_RECOVERY_PENDING)) {
+			netif_rx_complete(netdev);
+			schedule_work(&tp->reset_task);
+			return 0;
+		}
 	}
 
 	/* run RX thread, within the bounds set by NAPI.
@@ -3391,12 +3427,10 @@ static inline void tg3_full_lock(struct tg3 *tp, int irq_sync)
 	if (irq_sync)
 		tg3_irq_quiesce(tp);
 	spin_lock_bh(&tp->lock);
-	spin_lock(&tp->tx_lock);
 }
 
 static inline void tg3_full_unlock(struct tg3 *tp)
 {
-	spin_unlock(&tp->tx_lock);
 	spin_unlock_bh(&tp->lock);
 }
 
@@ -3579,6 +3613,13 @@ static void tg3_reset_task(void *_data)
 	restart_timer = tp->tg3_flags2 & TG3_FLG2_RESTART_TIMER;
 	tp->tg3_flags2 &= ~TG3_FLG2_RESTART_TIMER;
 
+	if (tp->tg3_flags & TG3_FLAG_TX_RECOVERY_PENDING) {
+		tp->write32_tx_mbox = tg3_write32_tx_mbox;
+		tp->write32_rx_mbox = tg3_write_flush_reg32;
+		tp->tg3_flags |= TG3_FLAG_MBOX_WRITE_REORDER;
+		tp->tg3_flags &= ~TG3_FLAG_TX_RECOVERY_PENDING;
+	}
+
 	tg3_halt(tp, RESET_KIND_SHUTDOWN, 0);
 	tg3_init_hw(tp, 1);
 
@@ -3718,14 +3759,11 @@ static int tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	len = skb_headlen(skb);
 
-	/* No BH disabling for tx_lock here.  We are running in BH disabled
+	/* We are running in BH disabled context with netif_tx_lock
-	 * context and TX reclaim runs via tp->poll inside of a software
+	 * and TX reclaim runs via tp->poll inside of a software
 	 * interrupt.  Furthermore, IRQ processing runs lockless so we have
 	 * no IRQ context deadlocks to worry about either.  Rejoice!
 	 */
-	if (!spin_trylock(&tp->tx_lock))
-		return NETDEV_TX_LOCKED;
-
 	if (unlikely(TX_BUFFS_AVAIL(tp) <= (skb_shinfo(skb)->nr_frags + 1))) {
 		if (!netif_queue_stopped(dev)) {
 			netif_stop_queue(dev);
@@ -3734,7 +3772,6 @@ static int tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			printk(KERN_ERR PFX "%s: BUG! Tx Ring full when "
 			       "queue awake!\n", dev->name);
 		}
-		spin_unlock(&tp->tx_lock);
 		return NETDEV_TX_BUSY;
 	}
 
@@ -3817,15 +3854,16 @@ static int tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	tw32_tx_mbox((MAILBOX_SNDHOST_PROD_IDX_0 + TG3_64BIT_REG_LOW), entry);
 
 	tp->tx_prod = entry;
-	if (TX_BUFFS_AVAIL(tp) <= (MAX_SKB_FRAGS + 1)) {
+	if (unlikely(TX_BUFFS_AVAIL(tp) <= (MAX_SKB_FRAGS + 1))) {
+		spin_lock(&tp->tx_lock);
 		netif_stop_queue(dev);
 		if (TX_BUFFS_AVAIL(tp) > TG3_TX_WAKEUP_THRESH)
 			netif_wake_queue(tp->dev);
+		spin_unlock(&tp->tx_lock);
 	}
 
 out_unlock:
     	mmiowb();
-	spin_unlock(&tp->tx_lock);
 
 	dev->trans_start = jiffies;
 
@@ -3844,14 +3882,11 @@ static int tg3_start_xmit_dma_bug(struct sk_buff *skb, struct net_device *dev)
 
 	len = skb_headlen(skb);
 
-	/* No BH disabling for tx_lock here.  We are running in BH disabled
+	/* We are running in BH disabled context with netif_tx_lock
-	 * context and TX reclaim runs via tp->poll inside of a software
+	 * and TX reclaim runs via tp->poll inside of a software
 	 * interrupt.  Furthermore, IRQ processing runs lockless so we have
 	 * no IRQ context deadlocks to worry about either.  Rejoice!
 	 */
-	if (!spin_trylock(&tp->tx_lock))
-		return NETDEV_TX_LOCKED; 
-
 	if (unlikely(TX_BUFFS_AVAIL(tp) <= (skb_shinfo(skb)->nr_frags + 1))) {
 		if (!netif_queue_stopped(dev)) {
 			netif_stop_queue(dev);
@@ -3860,7 +3895,6 @@ static int tg3_start_xmit_dma_bug(struct sk_buff *skb, struct net_device *dev)
 			printk(KERN_ERR PFX "%s: BUG! Tx Ring full when "
 			       "queue awake!\n", dev->name);
 		}
-		spin_unlock(&tp->tx_lock);
 		return NETDEV_TX_BUSY;
 	}
 
@@ -3998,15 +4032,16 @@ static int tg3_start_xmit_dma_bug(struct sk_buff *skb, struct net_device *dev)
 	tw32_tx_mbox((MAILBOX_SNDHOST_PROD_IDX_0 + TG3_64BIT_REG_LOW), entry);
 
 	tp->tx_prod = entry;
-	if (TX_BUFFS_AVAIL(tp) <= (MAX_SKB_FRAGS + 1)) {
+	if (unlikely(TX_BUFFS_AVAIL(tp) <= (MAX_SKB_FRAGS + 1))) {
+		spin_lock(&tp->tx_lock);
 		netif_stop_queue(dev);
 		if (TX_BUFFS_AVAIL(tp) > TG3_TX_WAKEUP_THRESH)
 			netif_wake_queue(tp->dev);
+		spin_unlock(&tp->tx_lock);
 	}
 
 out_unlock:
     	mmiowb();
-	spin_unlock(&tp->tx_lock);
 
 	dev->trans_start = jiffies;
 
@@ -11243,7 +11278,6 @@ static int __devinit tg3_init_one(struct pci_dev *pdev,
 	SET_MODULE_OWNER(dev);
 	SET_NETDEV_DEV(dev, &pdev->dev);
 
-	dev->features |= NETIF_F_LLTX;
 #if TG3_VLAN_TAG_USED
 	dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX;
 	dev->vlan_rx_register = tg3_vlan_rx_register;

drivers/net/tg3.h

@@ -2074,12 +2074,22 @@ struct tg3 {
 
 	/* SMP locking strategy:
 	 *
-	 * lock: Held during all operations except TX packet
+	 * lock: Held during reset, PHY access, timer, and when
-	 *       processing.
+	 *       updating tg3_flags and tg3_flags2.
 	 *
-	 * tx_lock: Held during tg3_start_xmit and tg3_tx
+	 * tx_lock: Held during tg3_start_xmit and tg3_tx only
+	 *          when calling netif_[start|stop]_queue.
+	 *          tg3_start_xmit is protected by netif_tx_lock.
 	 *
 	 * Both of these locks are to be held with BH safety.
+	 *
+	 * Because the IRQ handler, tg3_poll, and tg3_start_xmit
+	 * are running lockless, it is necessary to completely
+	 * quiesce the chip with tg3_netif_stop and tg3_full_lock
+	 * before reconfiguring the device.
+	 *
+	 * indirect_lock: Held when accessing registers indirectly
+	 *                with IRQ disabling.
 	 */
 	spinlock_t			lock;
 	spinlock_t			indirect_lock;
@@ -2155,11 +2165,7 @@ struct tg3 {
 #define TG3_FLAG_ENABLE_ASF		0x00000020
 #define TG3_FLAG_5701_REG_WRITE_BUG	0x00000040
 #define TG3_FLAG_POLL_SERDES		0x00000080
-#if defined(CONFIG_X86)
 #define TG3_FLAG_MBOX_WRITE_REORDER	0x00000100
-#else
-#define TG3_FLAG_MBOX_WRITE_REORDER	0	/* disables code too */
-#endif
 #define TG3_FLAG_PCIX_TARGET_HWBUG	0x00000200
 #define TG3_FLAG_WOL_SPEED_100MB	0x00000400
 #define TG3_FLAG_WOL_ENABLE		0x00000800
@@ -2172,6 +2178,7 @@ struct tg3 {
 #define TG3_FLAG_PCI_HIGH_SPEED		0x00040000
 #define TG3_FLAG_PCI_32BIT		0x00080000
 #define TG3_FLAG_SRAM_USE_CONFIG	0x00100000
+#define TG3_FLAG_TX_RECOVERY_PENDING	0x00200000
 #define TG3_FLAG_SERDES_WOL_CAP		0x00400000
 #define TG3_FLAG_JUMBO_RING_ENABLE	0x00800000
 #define TG3_FLAG_10_100_ONLY		0x01000000

drivers/net/tulip/winbond-840.c

@@ -1605,11 +1605,11 @@ static void __devexit w840_remove1 (struct pci_dev *pdev)
  * - get_stats:
  * 	spin_lock_irq(np->lock), doesn't touch hw if not present
  * - hard_start_xmit:
- * 	netif_stop_queue + spin_unlock_wait(&dev->xmit_lock);
+ * 	synchronize_irq + netif_tx_disable;
  * - tx_timeout:
- * 	netif_device_detach + spin_unlock_wait(&dev->xmit_lock);
+ * 	netif_device_detach + netif_tx_disable;
  * - set_multicast_list
- * 	netif_device_detach + spin_unlock_wait(&dev->xmit_lock);
+ * 	netif_device_detach + netif_tx_disable;
  * - interrupt handler
  * 	doesn't touch hw if not present, synchronize_irq waits for
  * 	running instances of the interrupt handler.
@@ -1635,11 +1635,10 @@ static int w840_suspend (struct pci_dev *pdev, pm_message_t state)
 		netif_device_detach(dev);
 		update_csr6(dev, 0);
 		iowrite32(0, ioaddr + IntrEnable);
-		netif_stop_queue(dev);
 		spin_unlock_irq(&np->lock);
 
-		spin_unlock_wait(&dev->xmit_lock);
 		synchronize_irq(dev->irq);
+		netif_tx_disable(dev);
 	
 		np->stats.rx_missed_errors += ioread32(ioaddr + RxMissed) & 0xffff;
 

drivers/net/via-velocity.c

@@ -1899,6 +1899,13 @@ static int velocity_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	int pktlen = skb->len;
 
+#ifdef VELOCITY_ZERO_COPY_SUPPORT
+	if (skb_shinfo(skb)->nr_frags > 6 && __skb_linearize(skb)) {
+		kfree_skb(skb);
+		return 0;
+	}
+#endif
+
 	spin_lock_irqsave(&vptr->lock, flags);
 
 	index = vptr->td_curr[qnum];
@@ -1914,8 +1921,6 @@ static int velocity_xmit(struct sk_buff *skb, struct net_device *dev)
 	 */
 	if (pktlen < ETH_ZLEN) {
 		/* Cannot occur until ZC support */
-		if(skb_linearize(skb, GFP_ATOMIC))
-			return 0; 
 		pktlen = ETH_ZLEN;
 		memcpy(tdinfo->buf, skb->data, skb->len);
 		memset(tdinfo->buf + skb->len, 0, ETH_ZLEN - skb->len);
@@ -1933,7 +1938,6 @@ static int velocity_xmit(struct sk_buff *skb, struct net_device *dev)
 		int nfrags = skb_shinfo(skb)->nr_frags;
 		tdinfo->skb = skb;
 		if (nfrags > 6) {
-			skb_linearize(skb, GFP_ATOMIC);
 			memcpy(tdinfo->buf, skb->data, skb->len);
 			tdinfo->skb_dma[0] = tdinfo->buf_dma;
 			td_ptr->tdesc0.pktsize = 

drivers/net/wireless/orinoco.c

@@ -1787,7 +1787,9 @@ static int __orinoco_program_rids(struct net_device *dev)
 	/* Set promiscuity / multicast*/
 	priv->promiscuous = 0;
 	priv->mc_count = 0;
-	__orinoco_set_multicast_list(dev); /* FIXME: what about the xmit_lock */
+
+	/* FIXME: what about netif_tx_lock */
+	__orinoco_set_multicast_list(dev);
 
 	return 0;
 }

include/linux/dmaengine.h

@@ -0,0 +1,359 @@
+/*
+ * Copyright(c) 2004 - 2006 Intel Corporation. All rights reserved.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called COPYING.
+ */
+#ifndef DMAENGINE_H
+#define DMAENGINE_H
+#include <linux/config.h>
+#ifdef CONFIG_DMA_ENGINE
+
+#include <linux/device.h>
+#include <linux/uio.h>
+#include <linux/kref.h>
+#include <linux/completion.h>
+#include <linux/rcupdate.h>
+
+/**
+ * enum dma_event - resource PNP/power managment events
+ * @DMA_RESOURCE_SUSPEND: DMA device going into low power state
+ * @DMA_RESOURCE_RESUME: DMA device returning to full power
+ * @DMA_RESOURCE_ADDED: DMA device added to the system
+ * @DMA_RESOURCE_REMOVED: DMA device removed from the system
+ */
+enum dma_event {
+	DMA_RESOURCE_SUSPEND,
+	DMA_RESOURCE_RESUME,
+	DMA_RESOURCE_ADDED,
+	DMA_RESOURCE_REMOVED,
+};
+
+/**
+ * typedef dma_cookie_t
+ *
+ * if dma_cookie_t is >0 it's a DMA request cookie, <0 it's an error code
+ */
+typedef s32 dma_cookie_t;
+
+#define dma_submit_error(cookie) ((cookie) < 0 ? 1 : 0)
+
+/**
+ * enum dma_status - DMA transaction status
+ * @DMA_SUCCESS: transaction completed successfully
+ * @DMA_IN_PROGRESS: transaction not yet processed
+ * @DMA_ERROR: transaction failed
+ */
+enum dma_status {
+	DMA_SUCCESS,
+	DMA_IN_PROGRESS,
+	DMA_ERROR,
+};
+
+/**
+ * struct dma_chan_percpu - the per-CPU part of struct dma_chan
+ * @refcount: local_t used for open-coded "bigref" counting
+ * @memcpy_count: transaction counter
+ * @bytes_transferred: byte counter
+ */
+
+struct dma_chan_percpu {
+	local_t refcount;
+	/* stats */
+	unsigned long memcpy_count;
+	unsigned long bytes_transferred;
+};
+
+/**
+ * struct dma_chan - devices supply DMA channels, clients use them
+ * @client: ptr to the client user of this chan, will be NULL when unused
+ * @device: ptr to the dma device who supplies this channel, always !NULL
+ * @cookie: last cookie value returned to client
+ * @chan_id:
+ * @class_dev:
+ * @refcount: kref, used in "bigref" slow-mode
+ * @slow_ref:
+ * @rcu:
+ * @client_node: used to add this to the client chan list
+ * @device_node: used to add this to the device chan list
+ * @local: per-cpu pointer to a struct dma_chan_percpu
+ */
+struct dma_chan {
+	struct dma_client *client;
+	struct dma_device *device;
+	dma_cookie_t cookie;
+
+	/* sysfs */
+	int chan_id;
+	struct class_device class_dev;
+
+	struct kref refcount;
+	int slow_ref;
+	struct rcu_head rcu;
+
+	struct list_head client_node;
+	struct list_head device_node;
+	struct dma_chan_percpu *local;
+};
+
+void dma_chan_cleanup(struct kref *kref);
+
+static inline void dma_chan_get(struct dma_chan *chan)
+{
+	if (unlikely(chan->slow_ref))
+		kref_get(&chan->refcount);
+	else {
+		local_inc(&(per_cpu_ptr(chan->local, get_cpu())->refcount));
+		put_cpu();
+	}
+}
+
+static inline void dma_chan_put(struct dma_chan *chan)
+{
+	if (unlikely(chan->slow_ref))
+		kref_put(&chan->refcount, dma_chan_cleanup);
+	else {
+		local_dec(&(per_cpu_ptr(chan->local, get_cpu())->refcount));
+		put_cpu();
+	}
+}
+
+/*
+ * typedef dma_event_callback - function pointer to a DMA event callback
+ */
+typedef void (*dma_event_callback) (struct dma_client *client,
+		struct dma_chan *chan, enum dma_event event);
+
+/**
+ * struct dma_client - info on the entity making use of DMA services
+ * @event_callback: func ptr to call when something happens
+ * @chan_count: number of chans allocated
+ * @chans_desired: number of chans requested. Can be +/- chan_count
+ * @lock: protects access to the channels list
+ * @channels: the list of DMA channels allocated
+ * @global_node: list_head for global dma_client_list
+ */
+struct dma_client {
+	dma_event_callback	event_callback;
+	unsigned int		chan_count;
+	unsigned int		chans_desired;
+
+	spinlock_t		lock;
+	struct list_head	channels;
+	struct list_head	global_node;
+};
+
+/**
+ * struct dma_device - info on the entity supplying DMA services
+ * @chancnt: how many DMA channels are supported
+ * @channels: the list of struct dma_chan
+ * @global_node: list_head for global dma_device_list
+ * @refcount:
+ * @done:
+ * @dev_id:
+ * Other func ptrs: used to make use of this device's capabilities
+ */
+struct dma_device {
+
+	unsigned int chancnt;
+	struct list_head channels;
+	struct list_head global_node;
+
+	struct kref refcount;
+	struct completion done;
+
+	int dev_id;
+
+	int (*device_alloc_chan_resources)(struct dma_chan *chan);
+	void (*device_free_chan_resources)(struct dma_chan *chan);
+	dma_cookie_t (*device_memcpy_buf_to_buf)(struct dma_chan *chan,
+			void *dest, void *src, size_t len);
+	dma_cookie_t (*device_memcpy_buf_to_pg)(struct dma_chan *chan,
+			struct page *page, unsigned int offset, void *kdata,
+			size_t len);
+	dma_cookie_t (*device_memcpy_pg_to_pg)(struct dma_chan *chan,
+			struct page *dest_pg, unsigned int dest_off,
+			struct page *src_pg, unsigned int src_off, size_t len);
+	enum dma_status (*device_memcpy_complete)(struct dma_chan *chan,
+			dma_cookie_t cookie, dma_cookie_t *last,
+			dma_cookie_t *used);
+	void (*device_memcpy_issue_pending)(struct dma_chan *chan);
+};
+
+/* --- public DMA engine API --- */
+
+struct dma_client *dma_async_client_register(dma_event_callback event_callback);
+void dma_async_client_unregister(struct dma_client *client);
+void dma_async_client_chan_request(struct dma_client *client,
+		unsigned int number);
+
+/**
+ * dma_async_memcpy_buf_to_buf - offloaded copy between virtual addresses
+ * @chan: DMA channel to offload copy to
+ * @dest: destination address (virtual)
+ * @src: source address (virtual)
+ * @len: length
+ *
+ * Both @dest and @src must be mappable to a bus address according to the
+ * DMA mapping API rules for streaming mappings.
+ * Both @dest and @src must stay memory resident (kernel memory or locked
+ * user space pages)
+ */
+static inline dma_cookie_t dma_async_memcpy_buf_to_buf(struct dma_chan *chan,
+	void *dest, void *src, size_t len)
+{
+	int cpu = get_cpu();
+	per_cpu_ptr(chan->local, cpu)->bytes_transferred += len;
+	per_cpu_ptr(chan->local, cpu)->memcpy_count++;
+	put_cpu();
+
+	return chan->device->device_memcpy_buf_to_buf(chan, dest, src, len);
+}
+
+/**
+ * dma_async_memcpy_buf_to_pg - offloaded copy
+ * @chan: DMA channel to offload copy to
+ * @page: destination page
+ * @offset: offset in page to copy to
+ * @kdata: source address (virtual)
+ * @len: length
+ *
+ * Both @page/@offset and @kdata must be mappable to a bus address according
+ * to the DMA mapping API rules for streaming mappings.
+ * Both @page/@offset and @kdata must stay memory resident (kernel memory or
+ * locked user space pages)
+ */
+static inline dma_cookie_t dma_async_memcpy_buf_to_pg(struct dma_chan *chan,
+	struct page *page, unsigned int offset, void *kdata, size_t len)
+{
+	int cpu = get_cpu();
+	per_cpu_ptr(chan->local, cpu)->bytes_transferred += len;
+	per_cpu_ptr(chan->local, cpu)->memcpy_count++;
+	put_cpu();
+
+	return chan->device->device_memcpy_buf_to_pg(chan, page, offset,
+	                                             kdata, len);
+}
+
+/**
+ * dma_async_memcpy_buf_to_pg - offloaded copy
+ * @chan: DMA channel to offload copy to
+ * @dest_page: destination page
+ * @dest_off: offset in page to copy to
+ * @src_page: source page
+ * @src_off: offset in page to copy from
+ * @len: length
+ *
+ * Both @dest_page/@dest_off and @src_page/@src_off must be mappable to a bus
+ * address according to the DMA mapping API rules for streaming mappings.
+ * Both @dest_page/@dest_off and @src_page/@src_off must stay memory resident
+ * (kernel memory or locked user space pages)
+ */
+static inline dma_cookie_t dma_async_memcpy_pg_to_pg(struct dma_chan *chan,
+	struct page *dest_pg, unsigned int dest_off, struct page *src_pg,
+	unsigned int src_off, size_t len)
+{
+	int cpu = get_cpu();
+	per_cpu_ptr(chan->local, cpu)->bytes_transferred += len;
+	per_cpu_ptr(chan->local, cpu)->memcpy_count++;
+	put_cpu();
+
+	return chan->device->device_memcpy_pg_to_pg(chan, dest_pg, dest_off,
+	                                            src_pg, src_off, len);
+}
+
+/**
+ * dma_async_memcpy_issue_pending - flush pending copies to HW
+ * @chan:
+ *
+ * This allows drivers to push copies to HW in batches,
+ * reducing MMIO writes where possible.
+ */
+static inline void dma_async_memcpy_issue_pending(struct dma_chan *chan)
+{
+	return chan->device->device_memcpy_issue_pending(chan);
+}
+
+/**
+ * dma_async_memcpy_complete - poll for transaction completion
+ * @chan: DMA channel
+ * @cookie: transaction identifier to check status of
+ * @last: returns last completed cookie, can be NULL
+ * @used: returns last issued cookie, can be NULL
+ *
+ * If @last and @used are passed in, upon return they reflect the driver
+ * internal state and can be used with dma_async_is_complete() to check
+ * the status of multiple cookies without re-checking hardware state.
+ */
+static inline enum dma_status dma_async_memcpy_complete(struct dma_chan *chan,
+	dma_cookie_t cookie, dma_cookie_t *last, dma_cookie_t *used)
+{
+	return chan->device->device_memcpy_complete(chan, cookie, last, used);
+}
+
+/**
+ * dma_async_is_complete - test a cookie against chan state
+ * @cookie: transaction identifier to test status of
+ * @last_complete: last know completed transaction
+ * @last_used: last cookie value handed out
+ *
+ * dma_async_is_complete() is used in dma_async_memcpy_complete()
+ * the test logic is seperated for lightweight testing of multiple cookies
+ */
+static inline enum dma_status dma_async_is_complete(dma_cookie_t cookie,
+			dma_cookie_t last_complete, dma_cookie_t last_used)
+{
+	if (last_complete <= last_used) {
+		if ((cookie <= last_complete) || (cookie > last_used))
+			return DMA_SUCCESS;
+	} else {
+		if ((cookie <= last_complete) && (cookie > last_used))
+			return DMA_SUCCESS;
+	}
+	return DMA_IN_PROGRESS;
+}
+
+
+/* --- DMA device --- */
+
+int dma_async_device_register(struct dma_device *device);
+void dma_async_device_unregister(struct dma_device *device);
+
+/* --- Helper iov-locking functions --- */
+
+struct dma_page_list {
+	char *base_address;
+	int nr_pages;
+	struct page **pages;
+};
+
+struct dma_pinned_list {
+	int nr_iovecs;
+	struct dma_page_list page_list[0];
+};
+
+struct dma_pinned_list *dma_pin_iovec_pages(struct iovec *iov, size_t len);
+void dma_unpin_iovec_pages(struct dma_pinned_list* pinned_list);
+
+dma_cookie_t dma_memcpy_to_iovec(struct dma_chan *chan, struct iovec *iov,
+	struct dma_pinned_list *pinned_list, unsigned char *kdata, size_t len);
+dma_cookie_t dma_memcpy_pg_to_iovec(struct dma_chan *chan, struct iovec *iov,
+	struct dma_pinned_list *pinned_list, struct page *page,
+	unsigned int offset, size_t len);
+
+#endif /* CONFIG_DMA_ENGINE */
+#endif /* DMAENGINE_H */

include/linux/igmp.h

@@ -169,7 +169,7 @@ struct ip_sf_list
 struct ip_mc_list
 {
 	struct in_device	*interface;
-	unsigned long		multiaddr;
+	__be32			multiaddr;
 	struct ip_sf_list	*sources;
 	struct ip_sf_list	*tomb;
 	unsigned int		sfmode;

include/linux/netdevice.h

@@ -37,6 +37,7 @@
 #include <linux/config.h>
 #include <linux/device.h>
 #include <linux/percpu.h>
+#include <linux/dmaengine.h>
 
 struct divert_blk;
 struct vlan_group;
@@ -311,6 +312,9 @@ struct net_device
 #define NETIF_F_LLTX		4096	/* LockLess TX */
 #define NETIF_F_UFO             8192    /* Can offload UDP Large Send*/
 
+#define NETIF_F_GEN_CSUM	(NETIF_F_NO_CSUM | NETIF_F_HW_CSUM)
+#define NETIF_F_ALL_CSUM	(NETIF_F_IP_CSUM | NETIF_F_GEN_CSUM)
+
 	struct net_device	*next_sched;
 
 	/* Interface index. Unique device identifier	*/
@@ -406,7 +410,7 @@ struct net_device
  * One part is mostly used on xmit path (device)
  */
 	/* hard_start_xmit synchronizer */
-	spinlock_t		xmit_lock ____cacheline_aligned_in_smp;
+	spinlock_t		_xmit_lock ____cacheline_aligned_in_smp;
 	/* cpu id of processor entered to hard_start_xmit or -1,
 	   if nobody entered there.
 	 */
@@ -593,6 +597,9 @@ struct softnet_data
 	struct sk_buff		*completion_queue;
 
 	struct net_device	backlog_dev;	/* Sorry. 8) */
+#ifdef CONFIG_NET_DMA
+	struct dma_chan		*net_dma;
+#endif
 };
 
 DECLARE_PER_CPU(struct softnet_data,softnet_data);
@@ -889,11 +896,43 @@ static inline void __netif_rx_complete(struct net_device *dev)
 	clear_bit(__LINK_STATE_RX_SCHED, &dev->state);
 }
 
+static inline void netif_tx_lock(struct net_device *dev)
+{
+	spin_lock(&dev->_xmit_lock);
+	dev->xmit_lock_owner = smp_processor_id();
+}
+
+static inline void netif_tx_lock_bh(struct net_device *dev)
+{
+	spin_lock_bh(&dev->_xmit_lock);
+	dev->xmit_lock_owner = smp_processor_id();
+}
+
+static inline int netif_tx_trylock(struct net_device *dev)
+{
+	int err = spin_trylock(&dev->_xmit_lock);
+	if (!err)
+		dev->xmit_lock_owner = smp_processor_id();
+	return err;
+}
+
+static inline void netif_tx_unlock(struct net_device *dev)
+{
+	dev->xmit_lock_owner = -1;
+	spin_unlock(&dev->_xmit_lock);
+}
+
+static inline void netif_tx_unlock_bh(struct net_device *dev)
+{
+	dev->xmit_lock_owner = -1;
+	spin_unlock_bh(&dev->_xmit_lock);
+}
+
 static inline void netif_tx_disable(struct net_device *dev)
 {
-	spin_lock_bh(&dev->xmit_lock);
+	netif_tx_lock_bh(dev);
 	netif_stop_queue(dev);
-	spin_unlock_bh(&dev->xmit_lock);
+	netif_tx_unlock_bh(dev);
 }
 
 /* These functions live elsewhere (drivers/net/net_init.c, but related) */

include/linux/netfilter/nf_conntrack_common.h

@@ -69,6 +69,10 @@ enum ip_conntrack_status {
 	/* Connection is dying (removed from lists), can not be unset. */
 	IPS_DYING_BIT = 9,
 	IPS_DYING = (1 << IPS_DYING_BIT),
+
+	/* Connection has fixed timeout. */
+	IPS_FIXED_TIMEOUT_BIT = 10,
+	IPS_FIXED_TIMEOUT = (1 << IPS_FIXED_TIMEOUT_BIT),
 };
 
 /* Connection tracking event bits */

include/linux/netfilter/nfnetlink_conntrack.h

@@ -27,13 +27,15 @@ enum ctattr_type {
 	CTA_STATUS,
 	CTA_PROTOINFO,
 	CTA_HELP,
-	CTA_NAT,
+	CTA_NAT_SRC,
+#define CTA_NAT	CTA_NAT_SRC	/* backwards compatibility */
 	CTA_TIMEOUT,
 	CTA_MARK,
 	CTA_COUNTERS_ORIG,
 	CTA_COUNTERS_REPLY,
 	CTA_USE,
 	CTA_ID,
+	CTA_NAT_DST,
 	__CTA_MAX
 };
 #define CTA_MAX (__CTA_MAX - 1)

include/linux/netfilter/xt_CONNSECMARK.h

@@ -0,0 +1,13 @@
+#ifndef _XT_CONNSECMARK_H_target
+#define _XT_CONNSECMARK_H_target
+
+enum {
+	CONNSECMARK_SAVE = 1,
+	CONNSECMARK_RESTORE,
+};
+
+struct xt_connsecmark_target_info {
+	u_int8_t mode;
+};
+
+#endif /*_XT_CONNSECMARK_H_target */

include/linux/netfilter/xt_SECMARK.h

@@ -0,0 +1,26 @@
+#ifndef _XT_SECMARK_H_target
+#define _XT_SECMARK_H_target
+
+/*
+ * This is intended for use by various security subsystems (but not
+ * at the same time).
+ *
+ * 'mode' refers to the specific security subsystem which the
+ * packets are being marked for.
+ */
+#define SECMARK_MODE_SEL	0x01		/* SELinux */
+#define SECMARK_SELCTX_MAX	256
+
+struct xt_secmark_target_selinux_info {
+	u_int32_t selsid;
+	char selctx[SECMARK_SELCTX_MAX];
+};
+
+struct xt_secmark_target_info {
+	u_int8_t mode;
+	union {
+		struct xt_secmark_target_selinux_info sel;
+	} u;
+};
+
+#endif /*_XT_SECMARK_H_target */

include/linux/netfilter/xt_quota.h

@@ -0,0 +1,16 @@
+#ifndef _XT_QUOTA_H
+#define _XT_QUOTA_H
+
+enum xt_quota_flags {
+	XT_QUOTA_INVERT		= 0x1,
+};
+#define XT_QUOTA_MASK		0x1
+
+struct xt_quota_info {
+	u_int32_t		flags;
+	u_int32_t		pad;
+	aligned_u64		quota;
+	struct xt_quota_info	*master;
+};
+
+#endif /* _XT_QUOTA_H */

include/linux/netfilter/xt_statistic.h

@@ -0,0 +1,32 @@
+#ifndef _XT_STATISTIC_H
+#define _XT_STATISTIC_H
+
+enum xt_statistic_mode {
+	XT_STATISTIC_MODE_RANDOM,
+	XT_STATISTIC_MODE_NTH,
+	__XT_STATISTIC_MODE_MAX
+};
+#define XT_STATISTIC_MODE_MAX (__XT_STATISTIC_MODE_MAX - 1)
+
+enum xt_statistic_flags {
+	XT_STATISTIC_INVERT		= 0x1,
+};
+#define XT_STATISTIC_MASK		0x1
+
+struct xt_statistic_info {
+	u_int16_t			mode;
+	u_int16_t			flags;
+	union {
+		struct {
+			u_int32_t	probability;
+		} random;
+		struct {
+			u_int32_t	every;
+			u_int32_t	packet;
+			u_int32_t	count;
+		} nth;
+	} u;
+	struct xt_statistic_info	*master __attribute__((aligned(8)));
+};
+
+#endif /* _XT_STATISTIC_H */

include/linux/netfilter_ipv4/ip_conntrack.h

@@ -121,6 +121,10 @@ struct ip_conntrack
 	u_int32_t mark;
 #endif
 
+#ifdef CONFIG_IP_NF_CONNTRACK_SECMARK
+	u_int32_t secmark;
+#endif
+
 	/* Traversed often, so hopefully in different cacheline to top */
 	/* These are my tuples; original and reply */
 	struct ip_conntrack_tuple_hash tuplehash[IP_CT_DIR_MAX];
@@ -154,6 +158,7 @@ struct ip_conntrack_expect
 	unsigned int flags;
 
 #ifdef CONFIG_IP_NF_NAT_NEEDED
+	u_int32_t saved_ip;
 	/* This is the original per-proto part, used to map the
 	 * expected connection the way the recipient expects. */
 	union ip_conntrack_manip_proto saved_proto;
@@ -293,6 +298,7 @@ static inline int is_dying(struct ip_conntrack *ct)
 }
 
 extern unsigned int ip_conntrack_htable_size;
+extern int ip_conntrack_checksum;
  
 #define CONNTRACK_STAT_INC(count) (__get_cpu_var(ip_conntrack_stat).count++)
 

include/linux/netfilter_ipv4/ip_conntrack_h323.h

@@ -71,6 +71,13 @@ extern int (*nat_h245_hook) (struct sk_buff ** pskb, struct ip_conntrack * ct,
 			     unsigned char **data, int dataoff,
 			     TransportAddress * addr, u_int16_t port,
 			     struct ip_conntrack_expect * exp);
+extern int (*nat_callforwarding_hook) (struct sk_buff ** pskb,
+				       struct ip_conntrack * ct,
+				       enum ip_conntrack_info ctinfo,
+				       unsigned char **data, int dataoff,
+				       TransportAddress * addr,
+				       u_int16_t port,
+				       struct ip_conntrack_expect * exp);
 extern int (*nat_q931_hook) (struct sk_buff ** pskb, struct ip_conntrack * ct,
 			     enum ip_conntrack_info ctinfo,
 			     unsigned char **data, TransportAddress * addr,

include/linux/netfilter_ipv4/ip_conntrack_helper_h323_types.h

@@ -1,4 +1,4 @@
-/* Generated by Jing Min Zhao's ASN.1 parser, Mar 15 2006
+/* Generated by Jing Min Zhao's ASN.1 parser, Apr 20 2006
  *
  * Copyright (c) 2006 Jing Min Zhao <zhaojingmin@users.sourceforge.net>
  *
@@ -412,6 +412,7 @@ typedef struct Facility_UUIE {	/* SEQUENCE */
 		eFacility_UUIE_destinationInfo = (1 << 14),
 		eFacility_UUIE_h245SecurityMode = (1 << 13),
 	} options;
+	TransportAddress alternativeAddress;
 	FacilityReason reason;
 	TransportAddress h245Address;
 	Facility_UUIE_fastStart fastStart;

include/linux/netfilter_ipv4/ip_conntrack_sip.h

@@ -0,0 +1,44 @@
+#ifndef __IP_CONNTRACK_SIP_H__
+#define __IP_CONNTRACK_SIP_H__
+#ifdef __KERNEL__
+
+#define SIP_PORT	5060
+#define SIP_TIMEOUT	3600
+
+#define POS_VIA		0
+#define POS_CONTACT	1
+#define POS_CONTENT	2
+#define POS_MEDIA	3
+#define POS_OWNER	4
+#define POS_CONNECTION	5
+#define POS_REQ_HEADER	6
+#define POS_SDP_HEADER	7
+
+struct sip_header_nfo {
+	const char	*lname;
+	const char	*sname;
+	const char	*ln_str;
+	size_t		lnlen;
+	size_t		snlen;
+	size_t		ln_strlen;
+	int		(*match_len)(const char *, const char *, int *);
+};
+
+extern unsigned int (*ip_nat_sip_hook)(struct sk_buff **pskb,
+				       enum ip_conntrack_info ctinfo,
+				       struct ip_conntrack *ct,
+				       const char **dptr);
+extern unsigned int (*ip_nat_sdp_hook)(struct sk_buff **pskb,
+				       enum ip_conntrack_info ctinfo,
+				       struct ip_conntrack_expect *exp,
+				       const char *dptr);
+
+extern int ct_sip_get_info(const char *dptr, size_t dlen,
+			   unsigned int *matchoff,
+			   unsigned int *matchlen,
+			   struct sip_header_nfo *hnfo);
+extern int ct_sip_lnlen(const char *line, const char *limit);
+extern const char *ct_sip_search(const char *needle, const char *haystack,
+                                 size_t needle_len, size_t haystack_len);
+#endif /* __KERNEL__ */
+#endif /* __IP_CONNTRACK_SIP_H__ */

include/linux/pci_ids.h

@@ -1897,6 +1897,7 @@
 #define PCI_DEVICE_ID_TIGON3_5751F	0x167e
 #define PCI_DEVICE_ID_TIGON3_5787M	0x1693
 #define PCI_DEVICE_ID_TIGON3_5782	0x1696
+#define PCI_DEVICE_ID_TIGON3_5786	0x169a
 #define PCI_DEVICE_ID_TIGON3_5787	0x169b
 #define PCI_DEVICE_ID_TIGON3_5788	0x169c
 #define PCI_DEVICE_ID_TIGON3_5789	0x169d
@@ -2053,6 +2054,7 @@
 #define PCI_DEVICE_ID_INTEL_80960_RP	0x1960
 #define PCI_DEVICE_ID_INTEL_82840_HB	0x1a21
 #define PCI_DEVICE_ID_INTEL_82845_HB	0x1a30
+#define PCI_DEVICE_ID_INTEL_IOAT	0x1a38
 #define PCI_DEVICE_ID_INTEL_82801AA_0	0x2410
 #define PCI_DEVICE_ID_INTEL_82801AA_1	0x2411
 #define PCI_DEVICE_ID_INTEL_82801AA_3	0x2413

include/linux/pfkeyv2.h

@@ -159,7 +159,7 @@ struct sadb_spirange {
 struct sadb_x_kmprivate {
 	uint16_t	sadb_x_kmprivate_len;
 	uint16_t	sadb_x_kmprivate_exttype;
-	u_int32_t	sadb_x_kmprivate_reserved;
+	uint32_t	sadb_x_kmprivate_reserved;
 } __attribute__((packed));
 /* sizeof(struct sadb_x_kmprivate) == 8 */
 

include/linux/security.h

@@ -805,31 +805,37 @@ struct swap_info_struct;
  *	used by the XFRM system.
  *	@sec_ctx contains the security context information being provided by
  *	the user-level policy update program (e.g., setkey).
- *	Allocate a security structure to the xp->selector.security field.
+ *	Allocate a security structure to the xp->security field.
  *	The security field is initialized to NULL when the xfrm_policy is
  *	allocated.
  *	Return 0 if operation was successful (memory to allocate, legal context)
  * @xfrm_policy_clone_security:
  *	@old contains an existing xfrm_policy in the SPD.
  *	@new contains a new xfrm_policy being cloned from old.
- *	Allocate a security structure to the new->selector.security field
+ *	Allocate a security structure to the new->security field
- *	that contains the information from the old->selector.security field.
+ *	that contains the information from the old->security field.
  *	Return 0 if operation was successful (memory to allocate).
  * @xfrm_policy_free_security:
  *	@xp contains the xfrm_policy
- *	Deallocate xp->selector.security.
+ *	Deallocate xp->security.
+ * @xfrm_policy_delete_security:
+ *	@xp contains the xfrm_policy.
+ *	Authorize deletion of xp->security.
  * @xfrm_state_alloc_security:
  *	@x contains the xfrm_state being added to the Security Association
  *	Database by the XFRM system.
  *	@sec_ctx contains the security context information being provided by
  *	the user-level SA generation program (e.g., setkey or racoon).
- *	Allocate a security structure to the x->sel.security field.  The
+ *	Allocate a security structure to the x->security field.  The
  *	security field is initialized to NULL when the xfrm_state is
  *	allocated.
  *	Return 0 if operation was successful (memory to allocate, legal context).
  * @xfrm_state_free_security:
  *	@x contains the xfrm_state.
- *	Deallocate x>sel.security.
+ *	Deallocate x->security.
+ * @xfrm_state_delete_security:
+ *	@x contains the xfrm_state.
+ *	Authorize deletion of x->security.
  * @xfrm_policy_lookup:
  *	@xp contains the xfrm_policy for which the access control is being
  *	checked.
@@ -1298,8 +1304,10 @@ struct security_operations {
 	int (*xfrm_policy_alloc_security) (struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx);
 	int (*xfrm_policy_clone_security) (struct xfrm_policy *old, struct xfrm_policy *new);
 	void (*xfrm_policy_free_security) (struct xfrm_policy *xp);
+	int (*xfrm_policy_delete_security) (struct xfrm_policy *xp);
 	int (*xfrm_state_alloc_security) (struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx);
 	void (*xfrm_state_free_security) (struct xfrm_state *x);
+	int (*xfrm_state_delete_security) (struct xfrm_state *x);
 	int (*xfrm_policy_lookup)(struct xfrm_policy *xp, u32 sk_sid, u8 dir);
 #endif	/* CONFIG_SECURITY_NETWORK_XFRM */
 
@@ -2934,11 +2942,21 @@ static inline void security_xfrm_policy_free(struct xfrm_policy *xp)
 	security_ops->xfrm_policy_free_security(xp);
 }
 
+static inline int security_xfrm_policy_delete(struct xfrm_policy *xp)
+{
+	return security_ops->xfrm_policy_delete_security(xp);
+}
+
 static inline int security_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx)
 {
 	return security_ops->xfrm_state_alloc_security(x, sec_ctx);
 }
 
+static inline int security_xfrm_state_delete(struct xfrm_state *x)
+{
+	return security_ops->xfrm_state_delete_security(x);
+}
+
 static inline void security_xfrm_state_free(struct xfrm_state *x)
 {
 	security_ops->xfrm_state_free_security(x);
@@ -2963,6 +2981,11 @@ static inline void security_xfrm_policy_free(struct xfrm_policy *xp)
 {
 }
 
+static inline int security_xfrm_policy_delete(struct xfrm_policy *xp)
+{
+	return 0;
+}
+
 static inline int security_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx)
 {
 	return 0;
@@ -2972,6 +2995,11 @@ static inline void security_xfrm_state_free(struct xfrm_state *x)
 {
 }
 
+static inline int security_xfrm_state_delete(struct xfrm_state *x)
+{
+	return 0;
+}
+
 static inline int security_xfrm_policy_lookup(struct xfrm_policy *xp, u32 sk_sid, u8 dir)
 {
 	return 0;

include/linux/selinux.h

@@ -118,6 +118,27 @@ void selinux_get_ipc_sid(const struct kern_ipc_perm *ipcp, u32 *sid);
  */
 void selinux_get_task_sid(struct task_struct *tsk, u32 *sid);
 
+/**
+ *     selinux_string_to_sid - map a security context string to a security ID
+ *     @str: the security context string to be mapped
+ *     @sid: ID value returned via this.
+ *
+ *     Returns 0 if successful, with the SID stored in sid.  A value
+ *     of zero for sid indicates no SID could be determined (but no error
+ *     occurred).
+ */
+int selinux_string_to_sid(char *str, u32 *sid);
+
+/**
+ *     selinux_relabel_packet_permission - check permission to relabel a packet
+ *     @sid: ID value to be applied to network packet (via SECMARK, most likely)
+ *
+ *     Returns 0 if the current task is allowed to label packets with the
+ *     supplied security ID.  Note that it is implicit that the packet is always
+ *     being relabeled from the default unlabled value, and that the access
+ *     control decision is made in the AVC.
+ */
+int selinux_relabel_packet_permission(u32 sid);
 
 #else
 
@@ -172,6 +193,17 @@ static inline void selinux_get_task_sid(struct task_struct *tsk, u32 *sid)
 	*sid = 0;
 }
 
+static inline int selinux_string_to_sid(const char *str, u32 *sid)
+{
+       *sid = 0;
+       return 0;
+}
+
+static inline int selinux_relabel_packet_permission(u32 sid)
+{
+	return 0;
+}
+
 #endif	/* CONFIG_SECURITY_SELINUX */
 
 #endif /* _LINUX_SELINUX_H */

include/linux/skbuff.h

@@ -29,6 +29,7 @@
 #include <linux/net.h>
 #include <linux/textsearch.h>
 #include <net/checksum.h>
+#include <linux/dmaengine.h>
 
 #define HAVE_ALLOC_SKB		/* For the drivers to know */
 #define HAVE_ALIGNABLE_SKB	/* Ditto 8)		   */
@@ -209,6 +210,7 @@ enum {
  *	@nf_bridge: Saved data about a bridged frame - see br_netfilter.c
  *	@tc_index: Traffic control index
  *	@tc_verd: traffic control verdict
+ *	@secmark: security marking
  */
 
 struct sk_buff {
@@ -285,6 +287,12 @@ struct sk_buff {
 	__u16			tc_verd;	/* traffic control verdict */
 #endif
 #endif
+#ifdef CONFIG_NET_DMA
+	dma_cookie_t		dma_cookie;
+#endif
+#ifdef CONFIG_NETWORK_SECMARK
+	__u32			secmark;
+#endif
 
 
 	/* These elements must be at the end, see alloc_skb() for details.  */
@@ -967,15 +975,16 @@ static inline void skb_reserve(struct sk_buff *skb, int len)
 #define NET_SKB_PAD	16
 #endif
 
-extern int ___pskb_trim(struct sk_buff *skb, unsigned int len, int realloc);
+extern int ___pskb_trim(struct sk_buff *skb, unsigned int len);
 
 static inline void __skb_trim(struct sk_buff *skb, unsigned int len)
 {
-	if (!skb->data_len) {
+	if (unlikely(skb->data_len)) {
+		WARN_ON(1);
+		return;
+	}
 	skb->len  = len;
 	skb->tail = skb->data + len;
-	} else
-		___pskb_trim(skb, len, 0);
 }
 
 /**
@@ -985,6 +994,7 @@ static inline void __skb_trim(struct sk_buff *skb, unsigned int len)
  *
  *	Cut the length of a buffer down by removing data from the tail. If
  *	the buffer is already under the length specified it is not modified.
+ *	The skb must be linear.
  */
 static inline void skb_trim(struct sk_buff *skb, unsigned int len)
 {
@@ -995,12 +1005,10 @@ static inline void skb_trim(struct sk_buff *skb, unsigned int len)
 
 static inline int __pskb_trim(struct sk_buff *skb, unsigned int len)
 {
-	if (!skb->data_len) {
+	if (skb->data_len)
-		skb->len  = len;
+		return ___pskb_trim(skb, len);
-		skb->tail = skb->data+len;
+	__skb_trim(skb, len);
 	return 0;
-	}
-	return ___pskb_trim(skb, len, 1);
 }
 
 static inline int pskb_trim(struct sk_buff *skb, unsigned int len)
@@ -1161,18 +1169,34 @@ static inline int skb_can_coalesce(struct sk_buff *skb, int i,
 	return 0;
 }
 
+static inline int __skb_linearize(struct sk_buff *skb)
+{
+	return __pskb_pull_tail(skb, skb->data_len) ? 0 : -ENOMEM;
+}
+
 /**
  *	skb_linearize - convert paged skb to linear one
  *	@skb: buffer to linarize
- *	@gfp: allocation mode
  *
  *	If there is no free memory -ENOMEM is returned, otherwise zero
  *	is returned and the old skb data released.
  */
-extern int __skb_linearize(struct sk_buff *skb, gfp_t gfp);
+static inline int skb_linearize(struct sk_buff *skb)
-static inline int skb_linearize(struct sk_buff *skb, gfp_t gfp)
 {
-	return __skb_linearize(skb, gfp);
+	return skb_is_nonlinear(skb) ? __skb_linearize(skb) : 0;
+}
+
+/**
+ *	skb_linearize_cow - make sure skb is linear and writable
+ *	@skb: buffer to process
+ *
+ *	If there is no free memory -ENOMEM is returned, otherwise zero
+ *	is returned and the old skb data released.
+ */
+static inline int skb_linearize_cow(struct sk_buff *skb)
+{
+	return skb_is_nonlinear(skb) || skb_cloned(skb) ?
+	       __skb_linearize(skb) : 0;
 }
 
 /**
@@ -1396,5 +1420,23 @@ static inline void nf_reset(struct sk_buff *skb)
 static inline void nf_reset(struct sk_buff *skb) {}
 #endif /* CONFIG_NETFILTER */
 
+#ifdef CONFIG_NETWORK_SECMARK
+static inline void skb_copy_secmark(struct sk_buff *to, const struct sk_buff *from)
+{
+	to->secmark = from->secmark;
+}
+
+static inline void skb_init_secmark(struct sk_buff *skb)
+{
+	skb->secmark = 0;
+}
+#else
+static inline void skb_copy_secmark(struct sk_buff *to, const struct sk_buff *from)
+{ }
+
+static inline void skb_init_secmark(struct sk_buff *skb)
+{ }
+#endif
+
 #endif	/* __KERNEL__ */
 #endif	/* _LINUX_SKBUFF_H */

include/linux/sysctl.h

@@ -313,6 +313,7 @@ enum
 	NET_NF_CONNTRACK_FRAG6_TIMEOUT=29,
 	NET_NF_CONNTRACK_FRAG6_LOW_THRESH=30,
 	NET_NF_CONNTRACK_FRAG6_HIGH_THRESH=31,
+	NET_NF_CONNTRACK_CHECKSUM=32,
 };
 
 /* /proc/sys/net/ipv4 */
@@ -403,6 +404,8 @@ enum
  	NET_TCP_MTU_PROBING=113,
 	NET_TCP_BASE_MSS=114,
 	NET_IPV4_TCP_WORKAROUND_SIGNED_WINDOWS=115,
+	NET_TCP_DMA_COPYBREAK=116,
+	NET_TCP_SLOW_START_AFTER_IDLE=117,
 };
 
 enum {
@@ -491,6 +494,7 @@ enum
  	NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_RECD=25,
  	NET_IPV4_NF_CONNTRACK_SCTP_TIMEOUT_SHUTDOWN_ACK_SENT=26,
 	NET_IPV4_NF_CONNTRACK_COUNT=27,
+	NET_IPV4_NF_CONNTRACK_CHECKSUM=28,
 };
  
 /* /proc/sys/net/ipv6 */

include/linux/tcp.h

@@ -18,6 +18,7 @@
 #define _LINUX_TCP_H
 
 #include <linux/types.h>
+#include <linux/dmaengine.h>
 #include <asm/byteorder.h>
 
 struct tcphdr {
@@ -233,6 +234,13 @@ struct tcp_sock {
 		struct iovec		*iov;
 		int			memory;
 		int			len;
+#ifdef CONFIG_NET_DMA
+		/* members for async copy */
+		struct dma_chan		*dma_chan;
+		int			wakeup;
+		struct dma_pinned_list	*pinned_list;
+		dma_cookie_t		dma_cookie;
+#endif
 	} ucopy;
 
 	__u32	snd_wl1;	/* Sequence for window update		*/

include/linux/xfrm.h

@@ -118,6 +118,10 @@ enum
 	XFRM_SHARE_UNIQUE	/* Use once */
 };
 
+#define XFRM_MODE_TRANSPORT 0
+#define XFRM_MODE_TUNNEL 1
+#define XFRM_MODE_MAX 2
+
 /* Netlink configuration messages.  */
 enum {
 	XFRM_MSG_BASE = 0x10,

include/net/ip.h

@@ -147,7 +147,6 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar
 struct ipv4_config
 {
 	int	log_martians;
-	int	autoconfig;
 	int	no_pmtu_disc;
 };
 

include/net/llc_if.h

@@ -16,6 +16,7 @@
 #include <linux/if.h>
 #include <linux/if_arp.h>
 #include <linux/llc.h>
+#include <linux/etherdevice.h>
 #include <net/llc.h>
 
 #define LLC_DATAUNIT_PRIM	1
@@ -61,8 +62,6 @@
 #define LLC_STATUS_CONFLICT	7 /* disconnect conn */
 #define LLC_STATUS_RESET_DONE	8 /*  */
 
-extern u8 llc_mac_null_var[IFHWADDRLEN];
-
 /**
  *      llc_mac_null - determines if a address is a null mac address
  *      @mac: Mac address to test if null.
@@ -70,16 +69,20 @@ extern u8 llc_mac_null_var[IFHWADDRLEN];
  *      Determines if a given address is a null mac address.  Returns 0 if the
  *      address is not a null mac, 1 if the address is a null mac.
  */
-static __inline__ int llc_mac_null(u8 *mac)
+static inline int llc_mac_null(const u8 *mac)
 {
-	return !memcmp(mac, llc_mac_null_var, IFHWADDRLEN);
+	return is_zero_ether_addr(mac);
 }
 
-static __inline__ int llc_addrany(struct llc_addr *addr)
+static inline int llc_addrany(const struct llc_addr *addr)
 {
 	return llc_mac_null(addr->mac) && !addr->lsap;
 }
 
+static inline int llc_mac_multicast(const u8 *mac)
+{
+	return is_multicast_ether_addr(mac);
+}
 /**
  *	llc_mac_match - determines if two mac addresses are the same
  *	@mac1: First mac address to compare.
@@ -89,9 +92,9 @@ static __inline__ int llc_addrany(struct llc_addr *addr)
  *	is not a complete match up to len, 1 if a complete match up to len is
  *	found.
  */
-static __inline__ int llc_mac_match(u8 *mac1, u8 *mac2)
+static inline int llc_mac_match(const u8 *mac1, const u8 *mac2)
 {
-	return !memcmp(mac1, mac2, IFHWADDRLEN);
+	return !compare_ether_addr(mac1, mac2);
 }
 
 extern int llc_establish_connection(struct sock *sk, u8 *lmac,

include/net/netdma.h

@@ -0,0 +1,44 @@
+/*
+ * Copyright(c) 2004 - 2006 Intel Corporation. All rights reserved.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called COPYING.
+ */
+#ifndef NETDMA_H
+#define NETDMA_H
+#include <linux/config.h>
+#ifdef CONFIG_NET_DMA
+#include <linux/dmaengine.h>
+#include <linux/skbuff.h>
+
+static inline struct dma_chan *get_softnet_dma(void)
+{
+	struct dma_chan *chan;
+	rcu_read_lock();
+	chan = rcu_dereference(__get_cpu_var(softnet_data.net_dma));
+	if (chan)
+		dma_chan_get(chan);
+	rcu_read_unlock();
+	return chan;
+}
+
+int dma_skb_copy_datagram_iovec(struct dma_chan* chan,
+		const struct sk_buff *skb, int offset, struct iovec *to,
+		size_t len, struct dma_pinned_list *pinned_list);
+
+#endif /* CONFIG_NET_DMA */
+#endif /* NETDMA_H */

include/net/netfilter/nf_conntrack.h

@@ -114,6 +114,10 @@ struct nf_conn
 	u_int32_t mark;
 #endif
 
+#ifdef CONFIG_NF_CONNTRACK_SECMARK
+	u_int32_t secmark;
+#endif
+
 	/* Storage reserved for other modules: */
 	union nf_conntrack_proto proto;
 
@@ -285,6 +289,7 @@ static inline int nf_ct_is_dying(struct nf_conn *ct)
 }
 
 extern unsigned int nf_conntrack_htable_size;
+extern int nf_conntrack_checksum;
 
 #define NF_CT_STAT_INC(count) (__get_cpu_var(nf_conntrack_stat).count++)
 

include/net/netfilter/nf_conntrack_compat.h

@@ -20,6 +20,19 @@ static inline u_int32_t *nf_ct_get_mark(const struct sk_buff *skb,
 }
 #endif /* CONFIG_IP_NF_CONNTRACK_MARK */
 
+#ifdef CONFIG_IP_NF_CONNTRACK_SECMARK
+static inline u_int32_t *nf_ct_get_secmark(const struct sk_buff *skb,
+					   u_int32_t *ctinfo)
+{
+	struct ip_conntrack *ct = ip_conntrack_get(skb, ctinfo);
+
+	if (ct)
+		return &ct->secmark;
+	else
+		return NULL;
+}
+#endif /* CONFIG_IP_NF_CONNTRACK_SECMARK */
+
 #ifdef CONFIG_IP_NF_CT_ACCT
 static inline struct ip_conntrack_counter *
 nf_ct_get_counters(const struct sk_buff *skb)
@@ -70,6 +83,19 @@ static inline u_int32_t *nf_ct_get_mark(const struct sk_buff *skb,
 }
 #endif /* CONFIG_NF_CONNTRACK_MARK */
 
+#ifdef CONFIG_NF_CONNTRACK_SECMARK
+static inline u_int32_t *nf_ct_get_secmark(const struct sk_buff *skb,
+					   u_int32_t *ctinfo)
+{
+	struct nf_conn *ct = nf_ct_get(skb, ctinfo);
+
+	if (ct)
+		return &ct->secmark;
+	else
+		return NULL;
+}
+#endif /* CONFIG_NF_CONNTRACK_MARK */
+
 #ifdef CONFIG_NF_CT_ACCT
 static inline struct ip_conntrack_counter *
 nf_ct_get_counters(const struct sk_buff *skb)

include/net/raw.h

@@ -36,7 +36,7 @@ extern rwlock_t raw_v4_lock;
 
 
 extern struct sock *__raw_v4_lookup(struct sock *sk, unsigned short num,
-				    unsigned long raddr, unsigned long laddr,
+				    __be32 raddr, __be32 laddr,
 				    int dif);
 
 extern int raw_v4_input(struct sk_buff *skb, struct iphdr *iph, int hash);

include/net/sctp/sctp.h

@@ -255,7 +255,7 @@ extern int sctp_debug_flag;
 #define SCTP_DEBUG_PRINTK_IPADDR(whatever...)
 #define SCTP_ENABLE_DEBUG
 #define SCTP_DISABLE_DEBUG
-#define SCTP_ASSERT(expr, str, func)
+#define SCTP_ASSERT(expr, str, func) BUG_ON(!(expr))
 
 #endif /* SCTP_DEBUG */
 

include/net/sctp/structs.h

@@ -555,7 +555,8 @@ struct sctp_af {
 	int		(*to_addr_param) (const union sctp_addr *,
 					  union sctp_addr_param *); 
 	int		(*addr_valid)	(union sctp_addr *,
-					 struct sctp_sock *);
+					 struct sctp_sock *,
+					 const struct sk_buff *);
 	sctp_scope_t	(*scope) (union sctp_addr *);
 	void		(*inaddr_any)	(union sctp_addr *, unsigned short);
 	int		(*is_any)	(const union sctp_addr *);

include/net/sock.h

@@ -132,6 +132,7 @@ struct sock_common {
   *	@sk_receive_queue: incoming packets
   *	@sk_wmem_alloc: transmit queue bytes committed
   *	@sk_write_queue: Packet sending queue
+  *	@sk_async_wait_queue: DMA copied packets
   *	@sk_omem_alloc: "o" is "option" or "other"
   *	@sk_wmem_queued: persistent queue size
   *	@sk_forward_alloc: space allocated forward
@@ -205,6 +206,7 @@ struct sock {
 	atomic_t		sk_omem_alloc;
 	struct sk_buff_head	sk_receive_queue;
 	struct sk_buff_head	sk_write_queue;
+	struct sk_buff_head	sk_async_wait_queue;
 	int			sk_wmem_queued;
 	int			sk_forward_alloc;
 	gfp_t			sk_allocation;
@@ -871,10 +873,7 @@ static inline int sk_filter(struct sock *sk, struct sk_buff *skb, int needlock)
 		if (filter) {
 			unsigned int pkt_len = sk_run_filter(skb, filter->insns,
 							     filter->len);
-			if (!pkt_len)
+			err = pkt_len ? pskb_trim(skb, pkt_len) : -EPERM;
-				err = -EPERM;
-			else
-				skb_trim(skb, pkt_len);
 		}
 
 		if (needlock)
@@ -1271,11 +1270,22 @@ sock_recv_timestamp(struct msghdr *msg, struct sock *sk, struct sk_buff *skb)
  * This routine must be called with interrupts disabled or with the socket
  * locked so that the sk_buff queue operation is ok.
 */
-static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb)
+#ifdef CONFIG_NET_DMA
+static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb, int copied_early)
+{
+	__skb_unlink(skb, &sk->sk_receive_queue);
+	if (!copied_early)
+		__kfree_skb(skb);
+	else
+		__skb_queue_tail(&sk->sk_async_wait_queue, skb);
+}
+#else
+static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb, int copied_early)
 {
 	__skb_unlink(skb, &sk->sk_receive_queue);
 	__kfree_skb(skb);
 }
+#endif
 
 extern void sock_enable_timestamp(struct sock *sk);
 extern int sock_get_timestamp(struct sock *, struct timeval __user *);

include/net/tcp.h

@@ -28,6 +28,7 @@
 #include <linux/cache.h>
 #include <linux/percpu.h>
 #include <linux/skbuff.h>
+#include <linux/dmaengine.h>
 
 #include <net/inet_connection_sock.h>
 #include <net/inet_timewait_sock.h>
@@ -218,6 +219,7 @@ extern int sysctl_tcp_adv_win_scale;
 extern int sysctl_tcp_tw_reuse;
 extern int sysctl_tcp_frto;
 extern int sysctl_tcp_low_latency;
+extern int sysctl_tcp_dma_copybreak;
 extern int sysctl_tcp_nometrics_save;
 extern int sysctl_tcp_moderate_rcvbuf;
 extern int sysctl_tcp_tso_win_divisor;
@@ -225,6 +227,7 @@ extern int sysctl_tcp_abc;
 extern int sysctl_tcp_mtu_probing;
 extern int sysctl_tcp_base_mss;
 extern int sysctl_tcp_workaround_signed_windows;
+extern int sysctl_tcp_slow_start_after_idle;
 
 extern atomic_t tcp_memory_allocated;
 extern atomic_t tcp_sockets_allocated;
@@ -293,6 +296,8 @@ extern int			tcp_rcv_established(struct sock *sk,
 
 extern void			tcp_rcv_space_adjust(struct sock *sk);
 
+extern void			tcp_cleanup_rbuf(struct sock *sk, int copied);
+
 extern int			tcp_twsk_unique(struct sock *sk,
 						struct sock *sktw, void *twp);
 
@@ -628,7 +633,7 @@ struct tcp_congestion_ops {
 	/* return slow start threshold (required) */
 	u32 (*ssthresh)(struct sock *sk);
 	/* lower bound for congestion window (optional) */
-	u32 (*min_cwnd)(struct sock *sk);
+	u32 (*min_cwnd)(const struct sock *sk);
 	/* do new cwnd calculation (required) */
 	void (*cong_avoid)(struct sock *sk, u32 ack,
 			   u32 rtt, u32 in_flight, int good_ack);
@@ -663,7 +668,7 @@ extern struct tcp_congestion_ops tcp_init_congestion_ops;
 extern u32 tcp_reno_ssthresh(struct sock *sk);
 extern void tcp_reno_cong_avoid(struct sock *sk, u32 ack,
 				u32 rtt, u32 in_flight, int flag);
-extern u32 tcp_reno_min_cwnd(struct sock *sk);
+extern u32 tcp_reno_min_cwnd(const struct sock *sk);
 extern struct tcp_congestion_ops tcp_reno;
 
 static inline void tcp_set_ca_state(struct sock *sk, const u8 ca_state)
@@ -817,6 +822,12 @@ static inline void tcp_prequeue_init(struct tcp_sock *tp)
 	tp->ucopy.len = 0;
 	tp->ucopy.memory = 0;
 	skb_queue_head_init(&tp->ucopy.prequeue);
+#ifdef CONFIG_NET_DMA
+	tp->ucopy.dma_chan = NULL;
+	tp->ucopy.wakeup = 0;
+	tp->ucopy.pinned_list = NULL;
+	tp->ucopy.dma_cookie = 0;
+#endif
 }
 
 /* Packet is added to VJ-style prequeue for processing in process

include/net/xfrm.h

@@ -20,6 +20,8 @@
 #include <net/ip6_fib.h>
 
 #define XFRM_ALIGN8(len)	(((len) + 7) & ~7)
+#define MODULE_ALIAS_XFRM_MODE(family, encap) \
+	MODULE_ALIAS("xfrm-mode-" __stringify(family) "-" __stringify(encap))
 
 extern struct sock *xfrm_nl;
 extern u32 sysctl_xfrm_aevent_etime;
@@ -164,6 +166,7 @@ struct xfrm_state
 	/* Reference to data common to all the instances of this
 	 * transformer. */
 	struct xfrm_type	*type;
+	struct xfrm_mode	*mode;
 
 	/* Security context */
 	struct xfrm_sec_ctx	*security;
@@ -204,8 +207,8 @@ struct xfrm_type;
 struct xfrm_dst;
 struct xfrm_policy_afinfo {
 	unsigned short		family;
-	rwlock_t		lock;
+	struct xfrm_type	*type_map[IPPROTO_MAX];
-	struct xfrm_type_map	*type_map;
+	struct xfrm_mode	*mode_map[XFRM_MODE_MAX];
 	struct dst_ops		*dst_ops;
 	void			(*garbage_collect)(void);
 	int			(*dst_lookup)(struct xfrm_dst **dst, struct flowi *fl);
@@ -232,7 +235,6 @@ extern int __xfrm_state_delete(struct xfrm_state *x);
 
 struct xfrm_state_afinfo {
 	unsigned short		family;
-	rwlock_t		lock;
 	struct list_head	*state_bydst;
 	struct list_head	*state_byspi;
 	int			(*init_flags)(struct xfrm_state *x);
@@ -264,16 +266,24 @@ struct xfrm_type
 	u32			(*get_max_size)(struct xfrm_state *, int size);
 };
 
-struct xfrm_type_map {
-	rwlock_t		lock;
-	struct xfrm_type	*map[256];
-};
-
 extern int xfrm_register_type(struct xfrm_type *type, unsigned short family);
 extern int xfrm_unregister_type(struct xfrm_type *type, unsigned short family);
 extern struct xfrm_type *xfrm_get_type(u8 proto, unsigned short family);
 extern void xfrm_put_type(struct xfrm_type *type);
 
+struct xfrm_mode {
+	int (*input)(struct xfrm_state *x, struct sk_buff *skb);
+	int (*output)(struct sk_buff *skb);
+
+	struct module *owner;
+	unsigned int encap;
+};
+
+extern int xfrm_register_mode(struct xfrm_mode *mode, int family);
+extern int xfrm_unregister_mode(struct xfrm_mode *mode, int family);
+extern struct xfrm_mode *xfrm_get_mode(unsigned int encap, int family);
+extern void xfrm_put_mode(struct xfrm_mode *mode);
+
 struct xfrm_tmpl
 {
 /* id in template is interpreted as:

net/Kconfig

@@ -66,6 +66,13 @@ source "net/ipv6/Kconfig"
 
 endif # if INET
 
+config NETWORK_SECMARK
+	bool "Security Marking"
+	help
+	  This enables security marking of network packets, similar
+	  to nfmark, but designated for security purposes.
+	  If you are unsure how to answer this question, answer N.
+
 menuconfig NETFILTER
 	bool "Network packet filtering (replaces ipchains)"
 	---help---
@@ -215,6 +222,21 @@ config NET_PKTGEN
 	  To compile this code as a module, choose M here: the
 	  module will be called pktgen.
 
+config NET_TCPPROBE
+	tristate "TCP connection probing"
+	depends on INET && EXPERIMENTAL && PROC_FS && KPROBES
+	---help---
+	This module allows for capturing the changes to TCP connection
+	state in response to incoming packets. It is used for debugging
+	TCP congestion avoidance modules. If you don't understand
+	what was just said, you don't need it: say N.
+
+	Documentation on how to use the packet generator can be found
+	at http://linux-net.osdl.org/index.php/TcpProbe
+
+	To compile this code as a module, choose M here: the
+	module will be called tcp_probe.
+
 endmenu
 
 endmenu

net/atm/clip.c

@@ -98,7 +98,7 @@ static void unlink_clip_vcc(struct clip_vcc *clip_vcc)
 		printk(KERN_CRIT "!clip_vcc->entry (clip_vcc %p)\n", clip_vcc);
 		return;
 	}
-	spin_lock_bh(&entry->neigh->dev->xmit_lock);	/* block clip_start_xmit() */
+	netif_tx_lock_bh(entry->neigh->dev);	/* block clip_start_xmit() */
 	entry->neigh->used = jiffies;
 	for (walk = &entry->vccs; *walk; walk = &(*walk)->next)
 		if (*walk == clip_vcc) {
@@ -122,7 +122,7 @@ static void unlink_clip_vcc(struct clip_vcc *clip_vcc)
 	printk(KERN_CRIT "ATMARP: unlink_clip_vcc failed (entry %p, vcc "
 	       "0x%p)\n", entry, clip_vcc);
       out:
-	spin_unlock_bh(&entry->neigh->dev->xmit_lock);
+	netif_tx_unlock_bh(entry->neigh->dev);
 }
 
 /* The neighbour entry n->lock is held. */

net/bridge/Makefile

@@ -6,7 +6,7 @@ obj-$(CONFIG_BRIDGE) += bridge.o
 
 bridge-y	:= br.o br_device.o br_fdb.o br_forward.o br_if.o br_input.o \
 			br_ioctl.o br_notify.o br_stp.o br_stp_bpdu.o \
-			br_stp_if.o br_stp_timer.o
+			br_stp_if.o br_stp_timer.o br_netlink.o
 
 bridge-$(CONFIG_SYSFS) += br_sysfs_if.o br_sysfs_br.o
 

net/bridge/br.c

@@ -30,36 +30,46 @@ static struct llc_sap *br_stp_sap;
 
 static int __init br_init(void)
 {
+	int err;
+
 	br_stp_sap = llc_sap_open(LLC_SAP_BSPAN, br_stp_rcv);
 	if (!br_stp_sap) {
 		printk(KERN_ERR "bridge: can't register sap for STP\n");
-		return -EBUSY;
+		return -EADDRINUSE;
 	}
 
 	br_fdb_init();
 
-#ifdef CONFIG_BRIDGE_NETFILTER
+	err = br_netfilter_init();
-	if (br_netfilter_init())
+	if (err)
-		return 1;
+		goto err_out1;
-#endif
+
+	err = register_netdevice_notifier(&br_device_notifier);
+	if (err)
+		goto err_out2;
+
+	br_netlink_init();
 	brioctl_set(br_ioctl_deviceless_stub);
 	br_handle_frame_hook = br_handle_frame;
 
 	br_fdb_get_hook = br_fdb_get;
 	br_fdb_put_hook = br_fdb_put;
 
-	register_netdevice_notifier(&br_device_notifier);
-
 	return 0;
+
+err_out2:
+	br_netfilter_fini();
+err_out1:
+	llc_sap_put(br_stp_sap);
+	return err;
 }
 
 static void __exit br_deinit(void)
 {
 	rcu_assign_pointer(br_stp_sap->rcv_func, NULL);
 
-#ifdef CONFIG_BRIDGE_NETFILTER
+	br_netlink_fini();
 	br_netfilter_fini();
-#endif
 	unregister_netdevice_notifier(&br_device_notifier);
 	brioctl_set(NULL);
 

net/bridge/br_device.c

@@ -145,9 +145,9 @@ static int br_set_tx_csum(struct net_device *dev, u32 data)
 	struct net_bridge *br = netdev_priv(dev);
 
 	if (data)
-		br->feature_mask |= NETIF_F_IP_CSUM;
+		br->feature_mask |= NETIF_F_NO_CSUM;
 	else
-		br->feature_mask &= ~NETIF_F_IP_CSUM;
+		br->feature_mask &= ~NETIF_F_ALL_CSUM;
 
 	br_features_recompute(br);
 	return 0;
@@ -185,5 +185,5 @@ void br_dev_setup(struct net_device *dev)
 	dev->priv_flags = IFF_EBRIDGE;
 
  	dev->features = NETIF_F_SG | NETIF_F_FRAGLIST
- 		| NETIF_F_HIGHDMA | NETIF_F_TSO | NETIF_F_IP_CSUM;
+ 		| NETIF_F_HIGHDMA | NETIF_F_TSO | NETIF_F_NO_CSUM;
 }

net/bridge/br_forward.c

@@ -20,14 +20,11 @@
 #include <linux/netfilter_bridge.h>
 #include "br_private.h"
 
+/* Don't forward packets to originating port or forwarding diasabled */
 static inline int should_deliver(const struct net_bridge_port *p, 
 				 const struct sk_buff *skb)
 {
-	if (skb->dev == p->dev ||
+	return (skb->dev != p->dev && p->state == BR_STATE_FORWARDING);
-	    p->state != BR_STATE_FORWARDING)
-		return 0;
-
-	return 1;
 }
 
 static inline unsigned packet_length(const struct sk_buff *skb)
@@ -55,10 +52,9 @@ int br_dev_queue_push_xmit(struct sk_buff *skb)
 
 int br_forward_finish(struct sk_buff *skb)
 {
-	NF_HOOK(PF_BRIDGE, NF_BR_POST_ROUTING, skb, NULL, skb->dev,
+	return NF_HOOK(PF_BRIDGE, NF_BR_POST_ROUTING, skb, NULL, skb->dev,
 		       br_dev_queue_push_xmit);
 
-	return 0;
 }
 
 static void __br_deliver(const struct net_bridge_port *to, struct sk_buff *skb)

net/bridge/br_if.c

@@ -372,12 +372,17 @@ void br_features_recompute(struct net_bridge *br)
 	struct net_bridge_port *p;
 	unsigned long features, checksum;
 
-	features = br->feature_mask &~ NETIF_F_IP_CSUM;
+	checksum = br->feature_mask & NETIF_F_ALL_CSUM ? NETIF_F_NO_CSUM : 0;
-	checksum = br->feature_mask & NETIF_F_IP_CSUM;
+	features = br->feature_mask & ~NETIF_F_ALL_CSUM;
 
 	list_for_each_entry(p, &br->port_list, list) {
-		if (!(p->dev->features 
+		if (checksum & NETIF_F_NO_CSUM &&
-		      & (NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM)))
+		    !(p->dev->features & NETIF_F_NO_CSUM))
+			checksum ^= NETIF_F_NO_CSUM | NETIF_F_HW_CSUM;
+		if (checksum & NETIF_F_HW_CSUM &&
+		    !(p->dev->features & NETIF_F_HW_CSUM))
+			checksum ^= NETIF_F_HW_CSUM | NETIF_F_IP_CSUM;
+		if (!(p->dev->features & NETIF_F_IP_CSUM))
 			checksum = 0;
 		features &= p->dev->features;
 	}

net/bridge/br_netfilter.c

@@ -407,12 +407,8 @@ static unsigned int br_nf_pre_routing_ipv6(unsigned int hook,
 	if (pkt_len || hdr->nexthdr != NEXTHDR_HOP) {
 		if (pkt_len + sizeof(struct ipv6hdr) > skb->len)
 			goto inhdr_error;
-		if (pkt_len + sizeof(struct ipv6hdr) < skb->len) {
+		if (pskb_trim_rcsum(skb, pkt_len + sizeof(struct ipv6hdr)))
-			if (__pskb_trim(skb, pkt_len + sizeof(struct ipv6hdr)))
 			goto inhdr_error;
-			if (skb->ip_summed == CHECKSUM_HW)
-				skb->ip_summed = CHECKSUM_NONE;
-		}
 	}
 	if (hdr->nexthdr == NEXTHDR_HOP && check_hbh_len(skb))
 		goto inhdr_error;
@@ -495,11 +491,7 @@ static unsigned int br_nf_pre_routing(unsigned int hook, struct sk_buff **pskb,
 	if (skb->len < len || len < 4 * iph->ihl)
 		goto inhdr_error;
 
-	if (skb->len > len) {
+	pskb_trim_rcsum(skb, len);
-		__pskb_trim(skb, len);
-		if (skb->ip_summed == CHECKSUM_HW)
-			skb->ip_summed = CHECKSUM_NONE;
-	}
 
 	nf_bridge_put(skb->nf_bridge);
 	if (!nf_bridge_alloc(skb))

net/bridge/br_netlink.c

@@ -0,0 +1,199 @@
+/*
+ *	Bridge netlink control interface
+ *
+ *	Authors:
+ *	Stephen Hemminger		<shemminger@osdl.org>
+ *
+ *	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.
+ */
+
+#include <linux/kernel.h>
+#include <linux/rtnetlink.h>
+#include "br_private.h"
+
+/*
+ * Create one netlink message for one interface
+ * Contains port and master info as well as carrier and bridge state.
+ */
+static int br_fill_ifinfo(struct sk_buff *skb, const struct net_bridge_port *port,
+			  u32 pid, u32 seq, int event, unsigned int flags)
+{
+	const struct net_bridge *br = port->br;
+	const struct net_device *dev = port->dev;
+	struct ifinfomsg *r;
+	struct nlmsghdr *nlh;
+	unsigned char *b = skb->tail;
+	u32 mtu = dev->mtu;
+	u8 operstate = netif_running(dev) ? dev->operstate : IF_OPER_DOWN;
+	u8 portstate = port->state;
+
+	pr_debug("br_fill_info event %d port %s master %s\n",
+		 event, dev->name, br->dev->name);
+
+	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*r), flags);
+	r = NLMSG_DATA(nlh);
+	r->ifi_family = AF_BRIDGE;
+	r->__ifi_pad = 0;
+	r->ifi_type = dev->type;
+	r->ifi_index = dev->ifindex;
+	r->ifi_flags = dev_get_flags(dev);
+	r->ifi_change = 0;
+
+	RTA_PUT(skb, IFLA_IFNAME, strlen(dev->name)+1, dev->name);
+
+	RTA_PUT(skb, IFLA_MASTER, sizeof(int), &br->dev->ifindex);
+
+	if (dev->addr_len)
+		RTA_PUT(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr);
+
+	RTA_PUT(skb, IFLA_MTU, sizeof(mtu), &mtu);
+	if (dev->ifindex != dev->iflink)
+		RTA_PUT(skb, IFLA_LINK, sizeof(int), &dev->iflink);
+
+
+	RTA_PUT(skb, IFLA_OPERSTATE, sizeof(operstate), &operstate);
+
+	if (event == RTM_NEWLINK)
+		RTA_PUT(skb, IFLA_PROTINFO, sizeof(portstate), &portstate);
+
+	nlh->nlmsg_len = skb->tail - b;
+
+	return skb->len;
+
+nlmsg_failure:
+rtattr_failure:
+
+	skb_trim(skb, b - skb->data);
+	return -EINVAL;
+}
+
+/*
+ * Notify listeners of a change in port information
+ */
+void br_ifinfo_notify(int event, struct net_bridge_port *port)
+{
+	struct sk_buff *skb;
+	int err = -ENOMEM;
+
+	pr_debug("bridge notify event=%d\n", event);
+	skb = alloc_skb(NLMSG_SPACE(sizeof(struct ifinfomsg) + 128),
+			GFP_ATOMIC);
+	if (!skb)
+		goto err_out;
+
+	err = br_fill_ifinfo(skb, port, current->pid, 0, event, 0);
+	if (err)
+		goto err_kfree;
+
+	NETLINK_CB(skb).dst_group = RTNLGRP_LINK;
+	netlink_broadcast(rtnl, skb, 0, RTNLGRP_LINK, GFP_ATOMIC);
+	return;
+
+err_kfree:
+	kfree_skb(skb);
+err_out:
+	netlink_set_err(rtnl, 0, RTNLGRP_LINK, err);
+}
+
+/*
+ * Dump information about all ports, in response to GETLINK
+ */
+static int br_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	struct net_device *dev;
+	int idx;
+	int s_idx = cb->args[0];
+	int err = 0;
+
+	read_lock(&dev_base_lock);
+	for (dev = dev_base, idx = 0; dev; dev = dev->next) {
+		struct net_bridge_port *p = dev->br_port;
+
+		/* not a bridge port */
+		if (!p)
+			continue;
+
+		if (idx < s_idx)
+			continue;
+
+		err = br_fill_ifinfo(skb, p, NETLINK_CB(cb->skb).pid,
+				     cb->nlh->nlmsg_seq, RTM_NEWLINK, NLM_F_MULTI);
+		if (err <= 0)
+			break;
+		++idx;
+	}
+	read_unlock(&dev_base_lock);
+
+	cb->args[0] = idx;
+
+	return skb->len;
+}
+
+/*
+ * Change state of port (ie from forwarding to blocking etc)
+ * Used by spanning tree in user space.
+ */
+static int br_rtm_setlink(struct sk_buff *skb,  struct nlmsghdr *nlh, void *arg)
+{
+	struct rtattr  **rta = arg;
+	struct ifinfomsg *ifm = NLMSG_DATA(nlh);
+	struct net_device *dev;
+	struct net_bridge_port *p;
+	u8 new_state;
+
+	if (ifm->ifi_family != AF_BRIDGE)
+		return -EPFNOSUPPORT;
+
+	/* Must pass valid state as PROTINFO */
+	if (rta[IFLA_PROTINFO-1]) {
+		u8 *pstate = RTA_DATA(rta[IFLA_PROTINFO-1]);
+		new_state = *pstate;
+	} else
+		return -EINVAL;
+
+	if (new_state > BR_STATE_BLOCKING)
+		return -EINVAL;
+
+	/* Find bridge port */
+	dev = __dev_get_by_index(ifm->ifi_index);
+	if (!dev)
+		return -ENODEV;
+
+	p = dev->br_port;
+	if (!p)
+		return -EINVAL;
+
+	/* if kernel STP is running, don't allow changes */
+	if (p->br->stp_enabled)
+		return -EBUSY;
+
+	if (!netif_running(dev))
+		return -ENETDOWN;
+
+	if (!netif_carrier_ok(dev) && new_state != BR_STATE_DISABLED)
+		return -ENETDOWN;
+
+	p->state = new_state;
+	br_log_state(p);
+	return 0;
+}
+
+
+static struct rtnetlink_link bridge_rtnetlink_table[RTM_NR_MSGTYPES] = {
+	[RTM_GETLINK - RTM_BASE] = { .dumpit	= br_dump_ifinfo, },
+	[RTM_SETLINK - RTM_BASE] = { .doit      = br_rtm_setlink, },
+};
+
+void __init br_netlink_init(void)
+{
+	rtnetlink_links[PF_BRIDGE] = bridge_rtnetlink_table;
+}
+
+void __exit br_netlink_fini(void)
+{
+	rtnetlink_links[PF_BRIDGE] = NULL;
+}
+

net/bridge/br_notify.c

@@ -14,6 +14,7 @@
  */
 
 #include <linux/kernel.h>
+#include <linux/rtnetlink.h>
 
 #include "br_private.h"
 
@@ -49,6 +50,7 @@ static int br_device_event(struct notifier_block *unused, unsigned long event, v
 
 	case NETDEV_CHANGEADDR:
 		br_fdb_changeaddr(p, dev->dev_addr);
+		br_ifinfo_notify(RTM_NEWLINK, p);
 		br_stp_recalculate_bridge_id(br);
 		break;
 

net/bridge/br_private.h

@@ -29,7 +29,7 @@
 
 #define BR_PORT_DEBOUNCE (HZ/10)
 
-#define BR_VERSION	"2.1"
+#define BR_VERSION	"2.2"
 
 typedef struct bridge_id bridge_id;
 typedef struct mac_addr mac_addr;
@@ -192,8 +192,13 @@ extern int br_dev_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
 extern int br_ioctl_deviceless_stub(unsigned int cmd, void __user *arg);
 
 /* br_netfilter.c */
+#ifdef CONFIG_BRIDGE_NETFILTER
 extern int br_netfilter_init(void);
 extern void br_netfilter_fini(void);
+#else
+#define br_netfilter_init()	(0)
+#define br_netfilter_fini()	do { } while(0)
+#endif
 
 /* br_stp.c */
 extern void br_log_state(const struct net_bridge_port *p);
@@ -232,6 +237,11 @@ extern struct net_bridge_fdb_entry *(*br_fdb_get_hook)(struct net_bridge *br,
 extern void (*br_fdb_put_hook)(struct net_bridge_fdb_entry *ent);
 
 
+/* br_netlink.c */
+extern void br_netlink_init(void);
+extern void br_netlink_fini(void);
+extern void br_ifinfo_notify(int event, struct net_bridge_port *port);
+
 #ifdef CONFIG_SYSFS
 /* br_sysfs_if.c */
 extern struct sysfs_ops brport_sysfs_ops;

net/bridge/br_stp_if.c

@@ -16,6 +16,7 @@
 #include <linux/kernel.h>
 #include <linux/smp_lock.h>
 #include <linux/etherdevice.h>
+#include <linux/rtnetlink.h>
 
 #include "br_private.h"
 #include "br_private_stp.h"
@@ -86,6 +87,7 @@ void br_stp_disable_bridge(struct net_bridge *br)
 void br_stp_enable_port(struct net_bridge_port *p)
 {
 	br_init_port(p);
+	br_ifinfo_notify(RTM_NEWLINK, p);
 	br_port_state_selection(p->br);
 }
 
@@ -99,6 +101,8 @@ void br_stp_disable_port(struct net_bridge_port *p)
 	printk(KERN_INFO "%s: port %i(%s) entering %s state\n",
 	       br->dev->name, p->port_no, p->dev->name, "disabled");
 
+	br_ifinfo_notify(RTM_DELLINK, p);
+
 	wasroot = br_is_root_bridge(br);
 	br_become_designated_port(p);
 	p->state = BR_STATE_DISABLED;

net/core/Makefile

@@ -16,3 +16,4 @@ obj-$(CONFIG_NET_DIVERT) += dv.o
 obj-$(CONFIG_NET_PKTGEN) += pktgen.o
 obj-$(CONFIG_WIRELESS_EXT) += wireless.o
 obj-$(CONFIG_NETPOLL) += netpoll.o
+obj-$(CONFIG_NET_DMA) += user_dma.o

net/core/dev.c

@@ -115,6 +115,7 @@
 #include <net/iw_handler.h>
 #include <asm/current.h>
 #include <linux/audit.h>
+#include <linux/dmaengine.h>
 
 /*
  *	The list of packet types we will receive (as opposed to discard)
@@ -148,6 +149,12 @@ static DEFINE_SPINLOCK(ptype_lock);
 static struct list_head ptype_base[16];	/* 16 way hashed list */
 static struct list_head ptype_all;		/* Taps */
 
+#ifdef CONFIG_NET_DMA
+static struct dma_client *net_dma_client;
+static unsigned int net_dma_count;
+static spinlock_t net_dma_event_lock;
+#endif
+
 /*
  * The @dev_base list is protected by @dev_base_lock and the rtnl
  * semaphore.
@@ -1215,75 +1222,15 @@ static inline int illegal_highdma(struct net_device *dev, struct sk_buff *skb)
 #define illegal_highdma(dev, skb)	(0)
 #endif
 
-/* Keep head the same: replace data */
-int __skb_linearize(struct sk_buff *skb, gfp_t gfp_mask)
-{
-	unsigned int size;
-	u8 *data;
-	long offset;
-	struct skb_shared_info *ninfo;
-	int headerlen = skb->data - skb->head;
-	int expand = (skb->tail + skb->data_len) - skb->end;
-
-	if (skb_shared(skb))
-		BUG();
-
-	if (expand <= 0)
-		expand = 0;
-
-	size = skb->end - skb->head + expand;
-	size = SKB_DATA_ALIGN(size);
-	data = kmalloc(size + sizeof(struct skb_shared_info), gfp_mask);
-	if (!data)
-		return -ENOMEM;
-
-	/* Copy entire thing */
-	if (skb_copy_bits(skb, -headerlen, data, headerlen + skb->len))
-		BUG();
-
-	/* Set up shinfo */
-	ninfo = (struct skb_shared_info*)(data + size);
-	atomic_set(&ninfo->dataref, 1);
-	ninfo->tso_size = skb_shinfo(skb)->tso_size;
-	ninfo->tso_segs = skb_shinfo(skb)->tso_segs;
-	ninfo->nr_frags = 0;
-	ninfo->frag_list = NULL;
-
-	/* Offset between the two in bytes */
-	offset = data - skb->head;
-
-	/* Free old data. */
-	skb_release_data(skb);
-
-	skb->head = data;
-	skb->end  = data + size;
-
-	/* Set up new pointers */
-	skb->h.raw   += offset;
-	skb->nh.raw  += offset;
-	skb->mac.raw += offset;
-	skb->tail    += offset;
-	skb->data    += offset;
-
-	/* We are no longer a clone, even if we were. */
-	skb->cloned    = 0;
-
-	skb->tail     += skb->data_len;
-	skb->data_len  = 0;
-	return 0;
-}
-
 #define HARD_TX_LOCK(dev, cpu) {			\
 	if ((dev->features & NETIF_F_LLTX) == 0) {	\
-		spin_lock(&dev->xmit_lock);		\
+		netif_tx_lock(dev);			\
-		dev->xmit_lock_owner = cpu;		\
 	}						\
 }
 
 #define HARD_TX_UNLOCK(dev) {				\
 	if ((dev->features & NETIF_F_LLTX) == 0) {	\
-		dev->xmit_lock_owner = -1;		\
+		netif_tx_unlock(dev);			\
-		spin_unlock(&dev->xmit_lock);		\
 	}						\
 }
 
@@ -1321,7 +1268,7 @@ int dev_queue_xmit(struct sk_buff *skb)
 
 	if (skb_shinfo(skb)->frag_list &&
 	    !(dev->features & NETIF_F_FRAGLIST) &&
-	    __skb_linearize(skb, GFP_ATOMIC))
+	    __skb_linearize(skb))
 		goto out_kfree_skb;
 
 	/* Fragmented skb is linearized if device does not support SG,
@@ -1330,14 +1277,14 @@ int dev_queue_xmit(struct sk_buff *skb)
 	 */
 	if (skb_shinfo(skb)->nr_frags &&
 	    (!(dev->features & NETIF_F_SG) || illegal_highdma(dev, skb)) &&
-	    __skb_linearize(skb, GFP_ATOMIC))
+	    __skb_linearize(skb))
 		goto out_kfree_skb;
 
 	/* If packet is not checksummed and device does not support
 	 * checksumming for this protocol, complete checksumming here.
 	 */
 	if (skb->ip_summed == CHECKSUM_HW &&
-	    (!(dev->features & (NETIF_F_HW_CSUM | NETIF_F_NO_CSUM)) &&
+	    (!(dev->features & NETIF_F_GEN_CSUM) &&
 	     (!(dev->features & NETIF_F_IP_CSUM) ||
 	      skb->protocol != htons(ETH_P_IP))))
 	      	if (skb_checksum_help(skb, 0))
@@ -1382,8 +1329,8 @@ int dev_queue_xmit(struct sk_buff *skb)
 	/* The device has no queue. Common case for software devices:
 	   loopback, all the sorts of tunnels...
 
-	   Really, it is unlikely that xmit_lock protection is necessary here.
+	   Really, it is unlikely that netif_tx_lock protection is necessary
-	   (f.e. loopback and IP tunnels are clean ignoring statistics
+	   here.  (f.e. loopback and IP tunnels are clean ignoring statistics
 	   counters.)
 	   However, it is possible, that they rely on protection
 	   made by us here.
@@ -1846,6 +1793,19 @@ static void net_rx_action(struct softirq_action *h)
 		}
 	}
 out:
+#ifdef CONFIG_NET_DMA
+	/*
+	 * There may not be any more sk_buffs coming right now, so push
+	 * any pending DMA copies to hardware
+	 */
+	if (net_dma_client) {
+		struct dma_chan *chan;
+		rcu_read_lock();
+		list_for_each_entry_rcu(chan, &net_dma_client->channels, client_node)
+			dma_async_memcpy_issue_pending(chan);
+		rcu_read_unlock();
+	}
+#endif
 	local_irq_enable();
 	return;
 
@@ -2785,7 +2745,7 @@ int register_netdevice(struct net_device *dev)
 	BUG_ON(dev->reg_state != NETREG_UNINITIALIZED);
 
 	spin_lock_init(&dev->queue_lock);
-	spin_lock_init(&dev->xmit_lock);
+	spin_lock_init(&dev->_xmit_lock);
 	dev->xmit_lock_owner = -1;
 #ifdef CONFIG_NET_CLS_ACT
 	spin_lock_init(&dev->ingress_lock);
@@ -2829,9 +2789,7 @@ int register_netdevice(struct net_device *dev)
 
 	/* Fix illegal SG+CSUM combinations. */
 	if ((dev->features & NETIF_F_SG) &&
-	    !(dev->features & (NETIF_F_IP_CSUM |
+	    !(dev->features & NETIF_F_ALL_CSUM)) {
-			       NETIF_F_NO_CSUM |
-			       NETIF_F_HW_CSUM))) {
 		printk("%s: Dropping NETIF_F_SG since no checksum feature.\n",
 		       dev->name);
 		dev->features &= ~NETIF_F_SG;
@@ -3300,6 +3258,88 @@ static int dev_cpu_callback(struct notifier_block *nfb,
 }
 #endif /* CONFIG_HOTPLUG_CPU */
 
+#ifdef CONFIG_NET_DMA
+/**
+ * net_dma_rebalance -
+ * This is called when the number of channels allocated to the net_dma_client
+ * changes.  The net_dma_client tries to have one DMA channel per CPU.
+ */
+static void net_dma_rebalance(void)
+{
+	unsigned int cpu, i, n;
+	struct dma_chan *chan;
+
+	lock_cpu_hotplug();
+
+	if (net_dma_count == 0) {
+		for_each_online_cpu(cpu)
+			rcu_assign_pointer(per_cpu(softnet_data.net_dma, cpu), NULL);
+		unlock_cpu_hotplug();
+		return;
+	}
+
+	i = 0;
+	cpu = first_cpu(cpu_online_map);
+
+	rcu_read_lock();
+	list_for_each_entry(chan, &net_dma_client->channels, client_node) {
+		n = ((num_online_cpus() / net_dma_count)
+		   + (i < (num_online_cpus() % net_dma_count) ? 1 : 0));
+
+		while(n) {
+			per_cpu(softnet_data.net_dma, cpu) = chan;
+			cpu = next_cpu(cpu, cpu_online_map);
+			n--;
+		}
+		i++;
+	}
+	rcu_read_unlock();
+
+	unlock_cpu_hotplug();
+}
+
+/**
+ * netdev_dma_event - event callback for the net_dma_client
+ * @client: should always be net_dma_client
+ * @chan:
+ * @event:
+ */
+static void netdev_dma_event(struct dma_client *client, struct dma_chan *chan,
+	enum dma_event event)
+{
+	spin_lock(&net_dma_event_lock);
+	switch (event) {
+	case DMA_RESOURCE_ADDED:
+		net_dma_count++;
+		net_dma_rebalance();
+		break;
+	case DMA_RESOURCE_REMOVED:
+		net_dma_count--;
+		net_dma_rebalance();
+		break;
+	default:
+		break;
+	}
+	spin_unlock(&net_dma_event_lock);
+}
+
+/**
+ * netdev_dma_regiser - register the networking subsystem as a DMA client
+ */
+static int __init netdev_dma_register(void)
+{
+	spin_lock_init(&net_dma_event_lock);
+	net_dma_client = dma_async_client_register(netdev_dma_event);
+	if (net_dma_client == NULL)
+		return -ENOMEM;
+
+	dma_async_client_chan_request(net_dma_client, num_online_cpus());
+	return 0;
+}
+
+#else
+static int __init netdev_dma_register(void) { return -ENODEV; }
+#endif /* CONFIG_NET_DMA */
 
 /*
  *	Initialize the DEV module. At boot time this walks the device list and
@@ -3353,6 +3393,8 @@ static int __init net_dev_init(void)
 		atomic_set(&queue->backlog_dev.refcnt, 1);
 	}
 
+	netdev_dma_register();
+
 	dev_boot_phase = 0;
 
 	open_softirq(NET_TX_SOFTIRQ, net_tx_action, NULL);
@@ -3371,7 +3413,6 @@ subsys_initcall(net_dev_init);
 EXPORT_SYMBOL(__dev_get_by_index);
 EXPORT_SYMBOL(__dev_get_by_name);
 EXPORT_SYMBOL(__dev_remove_pack);
-EXPORT_SYMBOL(__skb_linearize);
 EXPORT_SYMBOL(dev_valid_name);
 EXPORT_SYMBOL(dev_add_pack);
 EXPORT_SYMBOL(dev_alloc_name);

net/core/dev_mcast.c

@@ -62,7 +62,7 @@
  *	Device mc lists are changed by bh at least if IPv6 is enabled,
  *	so that it must be bh protected.
  *
- *	We block accesses to device mc filters with dev->xmit_lock.
+ *	We block accesses to device mc filters with netif_tx_lock.
  */
 
 /*
@@ -93,9 +93,9 @@ static void __dev_mc_upload(struct net_device *dev)
 
 void dev_mc_upload(struct net_device *dev)
 {
-	spin_lock_bh(&dev->xmit_lock);
+	netif_tx_lock_bh(dev);
 	__dev_mc_upload(dev);
-	spin_unlock_bh(&dev->xmit_lock);
+	netif_tx_unlock_bh(dev);
 }
 
 /*
@@ -107,7 +107,7 @@ int dev_mc_delete(struct net_device *dev, void *addr, int alen, int glbl)
 	int err = 0;
 	struct dev_mc_list *dmi, **dmip;
 
-	spin_lock_bh(&dev->xmit_lock);
+	netif_tx_lock_bh(dev);
 
 	for (dmip = &dev->mc_list; (dmi = *dmip) != NULL; dmip = &dmi->next) {
 		/*
@@ -139,13 +139,13 @@ int dev_mc_delete(struct net_device *dev, void *addr, int alen, int glbl)
 			 */
 			__dev_mc_upload(dev);
 			
-			spin_unlock_bh(&dev->xmit_lock);
+			netif_tx_unlock_bh(dev);
 			return 0;
 		}
 	}
 	err = -ENOENT;
 done:
-	spin_unlock_bh(&dev->xmit_lock);
+	netif_tx_unlock_bh(dev);
 	return err;
 }
 
@@ -160,7 +160,7 @@ int dev_mc_add(struct net_device *dev, void *addr, int alen, int glbl)
 
 	dmi1 = kmalloc(sizeof(*dmi), GFP_ATOMIC);
 
-	spin_lock_bh(&dev->xmit_lock);
+	netif_tx_lock_bh(dev);
 	for (dmi = dev->mc_list; dmi != NULL; dmi = dmi->next) {
 		if (memcmp(dmi->dmi_addr, addr, dmi->dmi_addrlen) == 0 &&
 		    dmi->dmi_addrlen == alen) {
@@ -176,7 +176,7 @@ int dev_mc_add(struct net_device *dev, void *addr, int alen, int glbl)
 	}
 
 	if ((dmi = dmi1) == NULL) {
-		spin_unlock_bh(&dev->xmit_lock);
+		netif_tx_unlock_bh(dev);
 		return -ENOMEM;
 	}
 	memcpy(dmi->dmi_addr, addr, alen);
@@ -189,11 +189,11 @@ int dev_mc_add(struct net_device *dev, void *addr, int alen, int glbl)
 
 	__dev_mc_upload(dev);
 	
-	spin_unlock_bh(&dev->xmit_lock);
+	netif_tx_unlock_bh(dev);
 	return 0;
 
 done:
-	spin_unlock_bh(&dev->xmit_lock);
+	netif_tx_unlock_bh(dev);
 	kfree(dmi1);
 	return err;
 }
@@ -204,7 +204,7 @@ done:
 
 void dev_mc_discard(struct net_device *dev)
 {
-	spin_lock_bh(&dev->xmit_lock);
+	netif_tx_lock_bh(dev);
 	
 	while (dev->mc_list != NULL) {
 		struct dev_mc_list *tmp = dev->mc_list;
@@ -215,7 +215,7 @@ void dev_mc_discard(struct net_device *dev)
 	}
 	dev->mc_count = 0;
 
-	spin_unlock_bh(&dev->xmit_lock);
+	netif_tx_unlock_bh(dev);
 }
 
 #ifdef CONFIG_PROC_FS
@@ -250,7 +250,7 @@ static int dev_mc_seq_show(struct seq_file *seq, void *v)
 	struct dev_mc_list *m;
 	struct net_device *dev = v;
 
-	spin_lock_bh(&dev->xmit_lock);
+	netif_tx_lock_bh(dev);
 	for (m = dev->mc_list; m; m = m->next) {
 		int i;
 
@@ -262,7 +262,7 @@ static int dev_mc_seq_show(struct seq_file *seq, void *v)
 
 		seq_putc(seq, '\n');
 	}
-	spin_unlock_bh(&dev->xmit_lock);
+	netif_tx_unlock_bh(dev);
 	return 0;
 }
 

net/core/ethtool.c

@@ -30,7 +30,7 @@ u32 ethtool_op_get_link(struct net_device *dev)
 
 u32 ethtool_op_get_tx_csum(struct net_device *dev)
 {
-	return (dev->features & (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM)) != 0;
+	return (dev->features & NETIF_F_ALL_CSUM) != 0;
 }
 
 int ethtool_op_set_tx_csum(struct net_device *dev, u32 data)
@@ -551,9 +551,7 @@ static int ethtool_set_sg(struct net_device *dev, char __user *useraddr)
 		return -EFAULT;
 
 	if (edata.data && 
-	    !(dev->features & (NETIF_F_IP_CSUM |
+	    !(dev->features & NETIF_F_ALL_CSUM))
-			       NETIF_F_NO_CSUM |
-			       NETIF_F_HW_CSUM)))
 		return -EINVAL;
 
 	return __ethtool_set_sg(dev, edata.data);
@@ -591,7 +589,7 @@ static int ethtool_set_tso(struct net_device *dev, char __user *useraddr)
 
 static int ethtool_get_ufo(struct net_device *dev, char __user *useraddr)
 {
-	struct ethtool_value edata = { ETHTOOL_GTSO };
+	struct ethtool_value edata = { ETHTOOL_GUFO };
 
 	if (!dev->ethtool_ops->get_ufo)
 		return -EOPNOTSUPP;
@@ -600,6 +598,7 @@ static int ethtool_get_ufo(struct net_device *dev, char __user *useraddr)
 		 return -EFAULT;
 	return 0;
 }
+
 static int ethtool_set_ufo(struct net_device *dev, char __user *useraddr)
 {
 	struct ethtool_value edata;

net/core/netpoll.c

@@ -273,24 +273,21 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb)
 
 	do {
 		npinfo->tries--;
-		spin_lock(&np->dev->xmit_lock);
+		netif_tx_lock(np->dev);
-		np->dev->xmit_lock_owner = smp_processor_id();
 
 		/*
 		 * network drivers do not expect to be called if the queue is
 		 * stopped.
 		 */
 		if (netif_queue_stopped(np->dev)) {
-			np->dev->xmit_lock_owner = -1;
+			netif_tx_unlock(np->dev);
-			spin_unlock(&np->dev->xmit_lock);
 			netpoll_poll(np);
 			udelay(50);
 			continue;
 		}
 
 		status = np->dev->hard_start_xmit(skb, np->dev);
-		np->dev->xmit_lock_owner = -1;
+		netif_tx_unlock(np->dev);
-		spin_unlock(&np->dev->xmit_lock);
 
 		/* success */
 		if(!status) {

net/core/pktgen.c

@@ -2897,7 +2897,7 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 		}
 	}
 
-	spin_lock_bh(&odev->xmit_lock);
+	netif_tx_lock_bh(odev);
 	if (!netif_queue_stopped(odev)) {
 
 		atomic_inc(&(pkt_dev->skb->users));
@@ -2942,7 +2942,7 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev)
 		pkt_dev->next_tx_ns = 0;
 	}
 
-	spin_unlock_bh(&odev->xmit_lock);
+	netif_tx_unlock_bh(odev);
 
 	/* If pkt_dev->count is zero, then run forever */
 	if ((pkt_dev->count != 0) && (pkt_dev->sofar >= pkt_dev->count)) {

net/core/skbuff.c

@@ -464,7 +464,7 @@ struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask)
 	n->tc_verd = CLR_TC_MUNGED(n->tc_verd);
 	C(input_dev);
 #endif
-
+	skb_copy_secmark(n, skb);
 #endif
 	C(truesize);
 	atomic_set(&n->users, 1);
@@ -526,6 +526,7 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old)
 #endif
 	new->tc_index	= old->tc_index;
 #endif
+	skb_copy_secmark(new, old);
 	atomic_set(&new->users, 1);
 	skb_shinfo(new)->tso_size = skb_shinfo(old)->tso_size;
 	skb_shinfo(new)->tso_segs = skb_shinfo(old)->tso_segs;
@@ -800,12 +801,10 @@ struct sk_buff *skb_pad(struct sk_buff *skb, int pad)
 	return nskb;
 }	
  
-/* Trims skb to length len. It can change skb pointers, if "realloc" is 1.
+/* Trims skb to length len. It can change skb pointers.
- * If realloc==0 and trimming is impossible without change of data,
- * it is BUG().
  */
 
-int ___pskb_trim(struct sk_buff *skb, unsigned int len, int realloc)
+int ___pskb_trim(struct sk_buff *skb, unsigned int len)
 {
 	int offset = skb_headlen(skb);
 	int nfrags = skb_shinfo(skb)->nr_frags;
@@ -815,7 +814,6 @@ int ___pskb_trim(struct sk_buff *skb, unsigned int len, int realloc)
 		int end = offset + skb_shinfo(skb)->frags[i].size;
 		if (end > len) {
 			if (skb_cloned(skb)) {
-				BUG_ON(!realloc);
 				if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
 					return -ENOMEM;
 			}

net/core/sock.c

@@ -832,6 +832,9 @@ struct sock *sk_clone(const struct sock *sk, const gfp_t priority)
 		atomic_set(&newsk->sk_omem_alloc, 0);
 		skb_queue_head_init(&newsk->sk_receive_queue);
 		skb_queue_head_init(&newsk->sk_write_queue);
+#ifdef CONFIG_NET_DMA
+		skb_queue_head_init(&newsk->sk_async_wait_queue);
+#endif
 
 		rwlock_init(&newsk->sk_dst_lock);
 		rwlock_init(&newsk->sk_callback_lock);
@@ -1383,6 +1386,9 @@ void sock_init_data(struct socket *sock, struct sock *sk)
 	skb_queue_head_init(&sk->sk_receive_queue);
 	skb_queue_head_init(&sk->sk_write_queue);
 	skb_queue_head_init(&sk->sk_error_queue);
+#ifdef CONFIG_NET_DMA
+	skb_queue_head_init(&sk->sk_async_wait_queue);
+#endif
 
 	sk->sk_send_head	=	NULL;
 

net/core/user_dma.c

@@ -0,0 +1,131 @@
+/*
+ * Copyright(c) 2004 - 2006 Intel Corporation. All rights reserved.
+ * Portions based on net/core/datagram.c and copyrighted by their authors.
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called COPYING.
+ */
+
+/*
+ * This code allows the net stack to make use of a DMA engine for
+ * skb to iovec copies.
+ */
+
+#include <linux/dmaengine.h>
+#include <linux/socket.h>
+#include <linux/rtnetlink.h> /* for BUG_TRAP */
+#include <net/tcp.h>
+
+#define NET_DMA_DEFAULT_COPYBREAK 4096
+
+int sysctl_tcp_dma_copybreak = NET_DMA_DEFAULT_COPYBREAK;
+
+/**
+ *	dma_skb_copy_datagram_iovec - Copy a datagram to an iovec.
+ *	@skb - buffer to copy
+ *	@offset - offset in the buffer to start copying from
+ *	@iovec - io vector to copy to
+ *	@len - amount of data to copy from buffer to iovec
+ *	@pinned_list - locked iovec buffer data
+ *
+ *	Note: the iovec is modified during the copy.
+ */
+int dma_skb_copy_datagram_iovec(struct dma_chan *chan,
+			struct sk_buff *skb, int offset, struct iovec *to,
+			size_t len, struct dma_pinned_list *pinned_list)
+{
+	int start = skb_headlen(skb);
+	int i, copy = start - offset;
+	dma_cookie_t cookie = 0;
+
+	/* Copy header. */
+	if (copy > 0) {
+		if (copy > len)
+			copy = len;
+		cookie = dma_memcpy_to_iovec(chan, to, pinned_list,
+		                            skb->data + offset, copy);
+		if (cookie < 0)
+			goto fault;
+		len -= copy;
+		if (len == 0)
+			goto end;
+		offset += copy;
+	}
+
+	/* 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);
+
+		end = start + skb_shinfo(skb)->frags[i].size;
+		copy = end - offset;
+		if ((copy = end - offset) > 0) {
+			skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+			struct page *page = frag->page;
+
+			if (copy > len)
+				copy = len;
+
+			cookie = dma_memcpy_pg_to_iovec(chan, to, pinned_list, page,
+					frag->page_offset + offset - start, copy);
+			if (cookie < 0)
+				goto fault;
+			len -= copy;
+			if (len == 0)
+				goto end;
+			offset += copy;
+		}
+		start = end;
+	}
+
+	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;
+			copy = end - offset;
+			if (copy > 0) {
+				if (copy > len)
+					copy = len;
+				cookie = dma_skb_copy_datagram_iovec(chan, list,
+				                offset - start, to, copy,
+				                pinned_list);
+				if (cookie < 0)
+					goto fault;
+				len -= copy;
+				if (len == 0)
+					goto end;
+				offset += copy;
+			}
+			start = end;
+		}
+	}
+
+end:
+	if (!len) {
+		skb->dma_cookie = cookie;
+		return cookie;
+	}
+
+fault:
+ 	return -EFAULT;
+}

net/dccp/proto.c

@@ -719,7 +719,7 @@ int dccp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 		}
 		dccp_pr_debug("packet_type=%s\n",
 			      dccp_packet_name(dh->dccph_type));
-		sk_eat_skb(sk, skb);
+		sk_eat_skb(sk, skb, 0);
 verify_sock_status:
 		if (sock_flag(sk, SOCK_DONE)) {
 			len = 0;
@@ -773,7 +773,7 @@ verify_sock_status:
 		}
 	found_fin_ok:
 		if (!(flags & MSG_PEEK))
-			sk_eat_skb(sk, skb);
+			sk_eat_skb(sk, skb, 0);
 		break;
 	} while (1);
 out:

net/decnet/dn_nsp_in.c

@@ -801,8 +801,7 @@ got_it:
 		 * We linearize everything except data segments here.
 		 */
 		if (cb->nsp_flags & ~0x60) {
-			if (unlikely(skb_is_nonlinear(skb)) &&
+			if (unlikely(skb_linearize(skb)))
-			    skb_linearize(skb, GFP_ATOMIC) != 0)
 				goto free_out;
 		}
 

net/decnet/dn_route.c

@@ -629,8 +629,7 @@ int dn_route_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type
 			padlen);
 
         if (flags & DN_RT_PKT_CNTL) {
-		if (unlikely(skb_is_nonlinear(skb)) &&
+		if (unlikely(skb_linearize(skb)))
-		    skb_linearize(skb, GFP_ATOMIC) != 0)
 			goto dump_it;
 
                 switch(flags & DN_RT_CNTL_MSK) {