Patchwork [v3] tilegx network driver: initial support

login
register
mail settings
Submitter Chris Metcalf
Date Sept. 17, 2001, 4 a.m.
Message ID <201205011759.q41HxR21020043@farm-0012.internal.tilera.com>
Download mbox | patch
Permalink /patch/156153/
State Changes Requested
Delegated to: David Miller
Headers show

Comments

Chris Metcalf - Sept. 17, 2001, 4 a.m.
This change adds support for the tilegx network driver based on the
GXIO IORPC support in the tilegx software stack, using the on-chip
mPIPE packet processing engine.

Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
---
This patch incoporates Arnd's comments about DEFINE_MUTEX(), per_cpu(),
and using module parameters.  I did not choose to break apart
tile_net_tx_tso() because when I did so, I ended up with sub-functions
requiring up to eleven parameters to carry the state around, and at
that point it seemed no better than the "one long function" model we
had before.

 drivers/net/ethernet/tile/Kconfig  |    1 +
 drivers/net/ethernet/tile/Makefile |    4 +-
 drivers/net/ethernet/tile/tilegx.c | 1949 ++++++++++++++++++++++++++++++++++++
 3 files changed, 1952 insertions(+), 2 deletions(-)
 create mode 100644 drivers/net/ethernet/tile/tilegx.c
David Miller - May 3, 2012, 5:41 a.m.
From: Chris Metcalf <cmetcalf@tilera.com>
Date: Mon, 17 Sep 2001 00:00:00 -0400

> +/* #define USE_SIM_PRINTF */
> +
> +#ifdef USE_SIM_PRINTF
> +
> +static __attribute__((unused, format (printf, 1, 2))) void
> +sim_printf(const char *format, ...)
 ...
> +/* HACK: Allow use of "sim_printf()" instead of "printk()". */
> +#define printk sim_printf
> +
> +#endif

This doesn't belong in a driver.

You want a debugging console driver that uses that special SIM output
facility instead.

Therefore, please remove this sim_printf stuff completely.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Chris Metcalf - May 3, 2012, 3:45 p.m.
On 5/3/2012 1:41 AM, David Miller wrote:
> From: Chris Metcalf <cmetcalf@tilera.com>
> Date: Mon, 17 Sep 2001 00:00:00 -0400
>
>> +/* #define USE_SIM_PRINTF */
>> +
>> +#ifdef USE_SIM_PRINTF
>> +
>> +static __attribute__((unused, format (printf, 1, 2))) void
>> +sim_printf(const char *format, ...)
>  ...
>> +/* HACK: Allow use of "sim_printf()" instead of "printk()". */
>> +#define printk sim_printf
>> +
>> +#endif
> This doesn't belong in a driver.
>
> You want a debugging console driver that uses that special SIM output
> facility instead.
>
> Therefore, please remove this sim_printf stuff completely.

Thanks, I've removed it from my branch.  (Since it's a trivial update, I
won't repost the change on LKML unless I get any more feedback that needs
addressing.)

I've checked in support for a "sim_console" boot flag that modifies the
behavior of the tile-specific console driver to use the simulator output
facility instead.  I'll plan to push that to LKML with the next batch of
changes I post.
David Miller - May 3, 2012, 5:07 p.m.
From: Chris Metcalf <cmetcalf@tilera.com>
Date: Thu, 3 May 2012 11:45:32 -0400

> Thanks, I've removed it from my branch.  (Since it's a trivial update, I
> won't repost the change on LKML unless I get any more feedback that needs
> addressing.)

Sorry, this approach doesn't work, you should post a new version
more expediently.

When I see something so terrible like I saw this time, I just provide
feedback on that major issue and stop reviewing.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Chris Metcalf - May 3, 2012, 5:25 p.m.
On 5/3/2012 1:07 PM, David Miller wrote:
> From: Chris Metcalf <cmetcalf@tilera.com>
> Date: Thu, 3 May 2012 11:45:32 -0400
>
>> Thanks, I've removed it from my branch.  (Since it's a trivial update, I
>> won't repost the change on LKML unless I get any more feedback that needs
>> addressing.)
> Sorry, this approach doesn't work, you should post a new version
> more expediently.
>
> When I see something so terrible like I saw this time, I just provide
> feedback on that major issue and stop reviewing.

Thanks, good to know.  Will do.

Patch

diff --git a/drivers/net/ethernet/tile/Kconfig b/drivers/net/ethernet/tile/Kconfig
index 2d9218f..9184b61 100644
--- a/drivers/net/ethernet/tile/Kconfig
+++ b/drivers/net/ethernet/tile/Kconfig
@@ -7,6 +7,7 @@  config TILE_NET
 	depends on TILE
 	default y
 	select CRC32
+	select TILE_GXIO_MPIPE if TILEGX
 	---help---
 	  This is a standard Linux network device driver for the
 	  on-chip Tilera Gigabit Ethernet and XAUI interfaces.
diff --git a/drivers/net/ethernet/tile/Makefile b/drivers/net/ethernet/tile/Makefile
index f634f14..0ef9eef 100644
--- a/drivers/net/ethernet/tile/Makefile
+++ b/drivers/net/ethernet/tile/Makefile
@@ -4,7 +4,7 @@ 
 
 obj-$(CONFIG_TILE_NET) += tile_net.o
 ifdef CONFIG_TILEGX
-tile_net-objs := tilegx.o mpipe.o iorpc_mpipe.o dma_queue.o
+tile_net-y := tilegx.o
 else
-tile_net-objs := tilepro.o
+tile_net-y := tilepro.o
 endif
diff --git a/drivers/net/ethernet/tile/tilegx.c b/drivers/net/ethernet/tile/tilegx.c
new file mode 100644
index 0000000..6f7eaf4
--- /dev/null
+++ b/drivers/net/ethernet/tile/tilegx.c
@@ -0,0 +1,1949 @@ 
+/*
+ * Copyright 2012 Tilera 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, version 2.
+ *
+ *   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, GOOD TITLE or
+ *   NON INFRINGEMENT.  See the GNU General Public License for
+ *   more details.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/moduleparam.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>      /* printk() */
+#include <linux/slab.h>        /* kmalloc() */
+#include <linux/errno.h>       /* error codes */
+#include <linux/types.h>       /* size_t */
+#include <linux/interrupt.h>
+#include <linux/in.h>
+#include <linux/irq.h>
+#include <linux/netdevice.h>   /* struct device, and other headers */
+#include <linux/etherdevice.h> /* eth_type_trans */
+#include <linux/skbuff.h>
+#include <linux/ioctl.h>
+#include <linux/cdev.h>
+#include <linux/hugetlb.h>
+#include <linux/in6.h>
+#include <linux/timer.h>
+#include <linux/io.h>
+#include <linux/ctype.h>
+#include <asm/checksum.h>
+#include <asm/homecache.h>
+
+#include <gxio/mpipe.h>
+
+/* For TSO */
+#include <linux/ip.h>
+#include <linux/tcp.h>
+
+
+#include <arch/sim.h>
+
+
+/* #define USE_SIM_PRINTF */
+
+#ifdef USE_SIM_PRINTF
+
+static __attribute__((unused, format (printf, 1, 2))) void
+sim_printf(const char *format, ...)
+{
+	char *str;
+	char buf[1024];
+
+	va_list args;
+	va_start(args, format);
+	(void)vsnprintf(buf, sizeof(buf), format, args);
+	va_end(args);
+
+	/* NOTE: Copied from "sim_print()". */
+	for (str = buf; *str != '\0'; str++) {
+		__insn_mtspr(SPR_SIM_CONTROL, SIM_CONTROL_PUTC |
+			     (*str << _SIM_CONTROL_OPERATOR_BITS));
+	}
+	__insn_mtspr(SPR_SIM_CONTROL, SIM_CONTROL_PUTC |
+		     (SIM_PUTC_FLUSH_BINARY << _SIM_CONTROL_OPERATOR_BITS));
+}
+
+
+/* HACK: Allow use of "sim_printf()" instead of "printk()". */
+#define printk sim_printf
+
+#endif
+
+
+/* First, "tile_net_init_module()" initializes each network cpu to
+ * handle incoming packets, and initializes all the network devices.
+ *
+ * Then, "ifconfig DEVICE up" calls "tile_net_open()", which will
+ * turn on packet processing, if needed.
+ *
+ * If "ifconfig DEVICE down" is called, it uses "tile_net_stop()" to
+ * stop egress, and possibly turn off packet processing.
+ *
+ * We start out with the ingress IRQ enabled on each CPU.  When it
+ * fires, it is automatically disabled, and we call "napi_schedule()".
+ * This will cause "tile_net_poll()" to be called, which will pull
+ * packets from the netio queue, filtering them out, or passing them
+ * to "netif_receive_skb()".  If our budget is exhausted, we will
+ * return, knowing we will be called again later.  Otherwise, we
+ * reenable the ingress IRQ, and call "napi_complete()".
+ *
+ *
+ * NOTE: Failing to free completions for an arbitrarily long time
+ * (which is defined to be illegal) does in fact cause bizarre problems.
+ *
+ * NOTE: The egress code can be interrupted by the interrupt handler.
+ */
+
+
+/* HACK: Define to support GSO.
+ * ISSUE: This may actually hurt performance of the TCP blaster.
+ */
+#undef TILE_NET_GSO
+
+/* HACK: Define to support TSO. */
+#define TILE_NET_TSO
+
+/* Use 3000 to enable the Linux Traffic Control (QoS) layer, else 0. */
+#define TILE_NET_TX_QUEUE_LEN 0
+
+/* Define to dump packets (prints out the whole packet on tx and rx). */
+#undef TILE_NET_DUMP_PACKETS
+
+/* Define to use "round robin" distribution. */
+#undef TILE_NET_ROUND_ROBIN
+
+/* Default transmit lockup timeout period, in jiffies. */
+#define TILE_NET_TIMEOUT (5 * HZ)
+
+/* The maximum number of distinct channels (idesc.channel is 5 bits). */
+#define TILE_NET_CHANNELS 32
+
+/* Maximum number of idescs to handle per "poll". */
+#define TILE_NET_BATCH 128
+
+/* Maximum number of packets to handle per "poll". */
+#define TILE_NET_WEIGHT 64
+
+/* Number of entries in each iqueue. */
+#define IQUEUE_ENTRIES 512
+
+/* Number of entries in each equeue. */
+#define EQUEUE_ENTRIES 2048
+
+/* Total header bytes per equeue slot.  Must be big enough for 2 bytes
+ * of NET_IP_ALIGN alignment, plus 14 bytes (?) of L2 header, plus up to
+ * 60 bytes of actual TCP header.  We round up to align to cache lines.
+ */
+#define HEADER_BYTES 128
+
+/* Maximum completions per cpu per device (must be a power of two).
+ * ISSUE: What is the right number here?
+ */
+#define TILE_NET_MAX_COMPS 64
+
+
+#define ROUND_UP(n, align) (((n) + (align) - 1) & -(align))
+
+
+#define MAX_FRAGS (65536 / PAGE_SIZE + 2 + 1)
+
+
+MODULE_AUTHOR("Tilera Corporation");
+MODULE_LICENSE("GPL");
+
+
+
+/* A "packet fragment" (a chunk of memory). */
+struct frag {
+	void *buf;
+	size_t length;
+};
+
+
+/* A single completion. */
+struct tile_net_comp {
+	/* The "complete_count" when the completion will be complete. */
+	s64 when;
+	/* The buffer to be freed when the completion is complete. */
+	struct sk_buff *skb;
+};
+
+
+/* The completions for a given cpu and device. */
+struct tile_net_comps {
+	/* The completions. */
+	struct tile_net_comp comp_queue[TILE_NET_MAX_COMPS];
+	/* The number of completions used. */
+	unsigned long comp_next;
+	/* The number of completions freed. */
+	unsigned long comp_last;
+};
+
+
+/* Info for a specific cpu. */
+struct tile_net_info {
+	/* The NAPI struct. */
+	struct napi_struct napi;
+	/* Packet queue. */
+	gxio_mpipe_iqueue_t iqueue;
+	/* Our cpu. */
+	int my_cpu;
+	/* True if iqueue is valid. */
+	bool has_iqueue;
+	/* NAPI flags. */
+	bool napi_added;
+	bool napi_enabled;
+	/* Number of small sk_buffs which must still be provided. */
+	unsigned int num_needed_small_buffers;
+	/* Number of large sk_buffs which must still be provided. */
+	unsigned int num_needed_large_buffers;
+	/* A timer for handling egress completions. */
+	struct timer_list egress_timer;
+	/* True if "egress_timer" is scheduled. */
+	bool egress_timer_scheduled;
+	/* Comps for each egress channel. */
+	struct tile_net_comps *comps_for_echannel[TILE_NET_CHANNELS];
+};
+
+
+/* Info for egress on a particular egress channel. */
+struct tile_net_egress {
+	/* The "equeue". */
+	gxio_mpipe_equeue_t *equeue;
+	/* The headers for TSO. */
+	unsigned char *headers;
+};
+
+
+/* Info for a specific device. */
+struct tile_net_priv {
+	/* Our network device. */
+	struct net_device *dev;
+	/* The primary link. */
+	gxio_mpipe_link_t link;
+	/* The primary channel, if open, else -1. */
+	int channel;
+	/* The "loopify" egress link, if needed. */
+	gxio_mpipe_link_t loopify_link;
+	/* The "loopify" egress channel, if open, else -1. */
+	int loopify_channel;
+	/* The egress channel (channel or loopify_channel). */
+	int echannel;
+	/* Total stats. */
+	struct net_device_stats stats;
+};
+
+
+/* Egress info, indexed by "priv->echannel" (lazily created as needed). */
+static struct tile_net_egress egress_for_echannel[TILE_NET_CHANNELS];
+
+/* Devices currently associated with each channel.
+ * NOTE: The array entry can become NULL after ifconfig down, but
+ * we do not free the underlying net_device structures, so it is
+ * safe to use a pointer after reading it from this array.
+ */
+static struct net_device *tile_net_devs_for_channel[TILE_NET_CHANNELS];
+
+/* A mutex for "tile_net_devs_for_channel". */
+static DEFINE_MUTEX(tile_net_devs_for_channel_mutex);
+
+/* The per-cpu info. */
+static DEFINE_PER_CPU(struct tile_net_info, per_cpu_info);
+
+/* The "context" for all devices. */
+static gxio_mpipe_context_t context;
+
+/* The small/large "buffer stacks". */
+static int small_buffer_stack = -1;
+static int large_buffer_stack = -1;
+
+/* The buckets. */
+static int first_bucket = -1;
+static int num_buckets = 1;
+
+/* The ingress irq. */
+static int ingress_irq = -1;
+
+
+/* Text value of tile_net.cpus if passed as a module parameter. */
+static char *network_cpus_string;
+
+/* The actual cpus in "network_cpus". */
+static struct cpumask network_cpus_map;
+
+
+/* If "loopify=LINK" was specified, this is "LINK". */
+static char *loopify_link_name;
+
+
+/* The "tile_net.cpus" argument specifies the cpus that are dedicated
+ * to handle ingress packets.
+ *
+ * The parameter should be in the form "tile_net.cpus=m-n[,x-y]", where
+ * m, n, x, y are integer numbers that represent the cpus that can be
+ * neither a dedicated cpu nor a dataplane cpu.
+ */
+static bool network_cpus_init(void)
+{
+	char buf[1024];
+	int rc;
+
+	if (network_cpus_string == NULL)
+		return false;
+
+	rc = cpulist_parse_crop(network_cpus_string, &network_cpus_map);
+	if (rc != 0) {
+		pr_warning("tile_net.cpus=%s: malformed cpu list\n",
+		       network_cpus_string);
+		return false;
+	}
+
+	/* Remove dedicated cpus. */
+	cpumask_and(&network_cpus_map, &network_cpus_map, cpu_possible_mask);
+
+
+	if (cpumask_empty(&network_cpus_map)) {
+		pr_warning("Ignoring empty tile_net.cpus='%s'.\n",
+			   network_cpus_string);
+		return false;
+	}
+
+	cpulist_scnprintf(buf, sizeof(buf), &network_cpus_map);
+	pr_info("Linux network CPUs: %s\n", buf);
+	return true;
+}
+
+module_param_named(cpus, network_cpus_string, charp, 0444);
+MODULE_PARM_DESC(cpus, "cpulist of cores that handle network interrupts");
+
+
+/* The "tile_net.loopify=LINK" argument causes the named device to
+ * actually use "loop0" for ingress, and "loop1" for egress.  This
+ * allows an app to sit between the actual link and linux, passing
+ * (some) packets along to linux, and forwarding (some) packets sent
+ * out by linux.
+ */
+module_param_named(loopify, loopify_link_name, charp, 0444);
+MODULE_PARM_DESC(loopify, "name the device to use loop0/1 for ingress/egress");
+
+
+#ifdef TILE_NET_DUMP_PACKETS
+/* Dump a packet. */
+static void dump_packet(unsigned char *data, unsigned long length, char *s)
+{
+	unsigned long i;
+	static unsigned int count;
+	char buf[128];
+
+	pr_info("Dumping %s packet of 0x%lx bytes at %p [%d]\n",
+	       s, length, data, count++);
+
+	pr_info("\n");
+
+	for (i = 0; i < length; i++) {
+		if ((i & 0xf) == 0)
+			sprintf(buf, "%8.8lx:", i);
+		sprintf(buf + strlen(buf), " %02x", data[i]);
+		if ((i & 0xf) == 0xf || i == length - 1)
+			pr_info("%s\n", buf);
+	}
+
+	pr_info("\n");
+}
+#endif
+
+
+/* Allocate and push a buffer. */
+static bool tile_net_provide_buffer(bool small)
+{
+	int stack = small ? small_buffer_stack : large_buffer_stack;
+
+	/* Buffers must be aligned. */
+	const unsigned long align = 128;
+
+	/* Note that "dev_alloc_skb()" adds NET_SKB_PAD more bytes,
+	 * and also "reserves" that many bytes.
+	 */
+	int len = sizeof(struct sk_buff **) + align + (small ? 128 : 1664);
+
+	/* Allocate (or fail). */
+	struct sk_buff *skb = dev_alloc_skb(len);
+	if (skb == NULL)
+		return false;
+
+	/* Make room for a back-pointer to 'skb'. */
+	skb_reserve(skb, sizeof(struct sk_buff **));
+
+	/* Make sure we are aligned. */
+	skb_reserve(skb, -(long)skb->data & (align - 1));
+
+	/* Save a back-pointer to 'skb'. */
+	*(struct sk_buff **)(skb->data - sizeof(struct sk_buff **)) = skb;
+
+	/* Make sure "skb" and the back-pointer have been flushed. */
+	wmb();
+
+	gxio_mpipe_push_buffer(&context, stack,
+			       (void *)va_to_tile_io_addr(skb->data));
+
+	return true;
+}
+
+
+/* Provide linux buffers to mPIPE. */
+static void tile_net_provide_needed_buffers(struct tile_net_info *info)
+{
+	while (info->num_needed_small_buffers != 0) {
+		if (!tile_net_provide_buffer(true))
+			goto oops;
+		info->num_needed_small_buffers--;
+	}
+
+	while (info->num_needed_large_buffers != 0) {
+		if (!tile_net_provide_buffer(false))
+			goto oops;
+		info->num_needed_large_buffers--;
+	}
+
+	return;
+
+oops:
+
+	/* Add a description to the page allocation failure dump. */
+	pr_notice("Tile %d still needs some buffers\n", info->my_cpu);
+}
+
+
+/* Handle a packet.  Return true if "processed", false if "filtered". */
+static bool tile_net_handle_packet(struct tile_net_info *info,
+				    gxio_mpipe_idesc_t *idesc)
+{
+	struct net_device *dev = tile_net_devs_for_channel[idesc->channel];
+
+	void *va;
+
+	uint8_t l2_offset = gxio_mpipe_idesc_get_l2_offset(idesc);
+
+	void *buf;
+	unsigned long len;
+
+	int filter = 0;
+
+	/* Drop packets for which no buffer was available.
+	 * NOTE: This happens under heavy load.
+	 */
+	if (idesc->be) {
+		gxio_mpipe_iqueue_consume(&info->iqueue, idesc);
+		if (net_ratelimit())
+			pr_info("Dropping packet (insufficient buffers).\n");
+		return false;
+	}
+
+	/* Get the raw buffer VA. */
+	va = tile_io_addr_to_va((unsigned long)gxio_mpipe_idesc_get_va(idesc));
+
+	/* Get the actual packet start/length. */
+	buf = va + l2_offset;
+	len = gxio_mpipe_idesc_get_l2_length(idesc);
+
+	/* Point "va" at the raw buffer. */
+	va -= NET_IP_ALIGN;
+
+#ifdef TILE_NET_DUMP_PACKETS
+	dump_packet(buf, len, "rx");
+#endif /* TILE_NET_DUMP_PACKETS */
+
+	if (dev != NULL) {
+		/* ISSUE: Is this needed? */
+		dev->last_rx = jiffies;
+	}
+
+	if (dev == NULL || !(dev->flags & IFF_UP)) {
+		/* Filter packets received before we're up. */
+		filter = 1;
+	} else if (!(dev->flags & IFF_PROMISC)) {
+		/* ISSUE: "eth_type_trans()" implies that "IFF_PROMISC"
+		 * is set for "all silly devices", however, it appears
+		 * to NOT be set for us, so this code here DOES run.
+		 * FIXME: The classifier will soon detect "multicast".
+		 */
+		if (!is_multicast_ether_addr(buf)) {
+			/* Filter packets not for our address. */
+			const u8 *mine = dev->dev_addr;
+			filter = compare_ether_addr(mine, buf);
+		}
+	}
+
+	if (filter) {
+
+		/* ISSUE: Update "drop" statistics? */
+
+		gxio_mpipe_iqueue_drop(&info->iqueue, idesc);
+
+	} else {
+
+		struct tile_net_priv *priv = netdev_priv(dev);
+
+		/* Acquire the associated "skb". */
+		struct sk_buff **skb_ptr = va - sizeof(*skb_ptr);
+		struct sk_buff *skb = *skb_ptr;
+
+		/* Paranoia. */
+		if (skb->data != va)
+			panic("Corrupt linux buffer! "
+			      "buf=%p, skb=%p, skb->data=%p\n",
+			      buf, skb, skb->data);
+
+		/* Skip headroom, and any custom header. */
+		skb_reserve(skb, NET_IP_ALIGN + l2_offset);
+
+		/* Encode the actual packet length. */
+		skb_put(skb, len);
+
+		/* NOTE: This call also sets "skb->dev = dev".
+		 * ISSUE: The classifier provides us with "eth_type"
+		 * (aka "eth->h_proto"), which is basically the value
+		 * returned by "eth_type_trans()".
+		 * Note that "eth_type_trans()" computes "skb->pkt_type",
+		 * which would be useful for the "filter" check above,
+		 * if we had a (modifiable) "skb" to work with.
+		 */
+		skb->protocol = eth_type_trans(skb, dev);
+
+		/* Acknowledge "good" hardware checksums. */
+		if (idesc->cs && idesc->csum_seed_val == 0xFFFF)
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+		netif_receive_skb(skb);
+
+		/* Update stats. */
+		atomic_add(1, (atomic_t *)&priv->stats.rx_packets);
+		atomic_add(len, (atomic_t *)&priv->stats.rx_bytes);
+
+		/* Need a new buffer. */
+		if (idesc->size == GXIO_MPIPE_BUFFER_SIZE_128)
+			info->num_needed_small_buffers++;
+		else
+			info->num_needed_large_buffers++;
+	}
+
+	gxio_mpipe_iqueue_consume(&info->iqueue, idesc);
+
+	return !filter;
+}
+
+
+/* Handle some packets for the current CPU.
+ *
+ * This function handles up to TILE_NET_BATCH idescs per call.
+ *
+ * ISSUE: Since we do not provide new buffers until this function is
+ * complete, we must initially provide enough buffers for each network
+ * cpu to fill its iqueue and also its batched idescs.
+ *
+ * ISSUE: The "rotting packet" race condition occurs if a packet
+ * arrives after the queue appears to be empty, and before the
+ * hypervisor interrupt is re-enabled.
+ */
+static int tile_net_poll(struct napi_struct *napi, int budget)
+{
+	struct tile_net_info *info = &__get_cpu_var(per_cpu_info);
+
+	unsigned int work = 0;
+
+	gxio_mpipe_idesc_t *idesc;
+	int i, n;
+
+	/* Process packets. */
+	while ((n = gxio_mpipe_iqueue_try_peek(&info->iqueue, &idesc)) > 0) {
+		for (i = 0; i < n; i++) {
+			if (i == TILE_NET_BATCH)
+				goto done;
+			if (tile_net_handle_packet(info, idesc + i)) {
+				if (++work >= budget)
+					goto done;
+			}
+		}
+	}
+
+	/* There are no packets left. */
+	napi_complete(&info->napi);
+
+	/* Re-enable hypervisor interrupts. */
+	gxio_mpipe_enable_notif_ring_interrupt(&context, info->iqueue.ring);
+
+	/* HACK: Avoid the "rotting packet" problem. */
+	if (gxio_mpipe_iqueue_try_peek(&info->iqueue, &idesc) > 0)
+		napi_schedule(&info->napi);
+
+	/* ISSUE: Handle completions? */
+
+done:
+
+	tile_net_provide_needed_buffers(info);
+
+	return work;
+}
+
+
+/* Handle an ingress interrupt on the current cpu. */
+static irqreturn_t tile_net_handle_ingress_irq(int irq, void *unused)
+{
+	struct tile_net_info *info = &__get_cpu_var(per_cpu_info);
+	napi_schedule(&info->napi);
+	return IRQ_HANDLED;
+}
+
+
+/* Free some completions.  This must be called with interrupts blocked. */
+static void tile_net_free_comps(gxio_mpipe_equeue_t* equeue,
+				struct tile_net_comps *comps,
+				int limit, bool force_update)
+{
+	int n = 0;
+	while (comps->comp_last < comps->comp_next) {
+		unsigned int cid = comps->comp_last % TILE_NET_MAX_COMPS;
+		struct tile_net_comp *comp = &comps->comp_queue[cid];
+		if (!gxio_mpipe_equeue_is_complete(equeue, comp->when,
+						   force_update || n == 0))
+			return;
+		dev_kfree_skb_irq(comp->skb);
+		comps->comp_last++;
+		if (++n == limit)
+			return;
+	}
+}
+
+
+/* Make sure the egress timer is scheduled.
+ *
+ * Note that we use "schedule if not scheduled" logic instead of the more
+ * obvious "reschedule" logic, because "reschedule" is fairly expensive.
+ */
+static void tile_net_schedule_egress_timer(struct tile_net_info *info)
+{
+	if (!info->egress_timer_scheduled) {
+		mod_timer_pinned(&info->egress_timer, jiffies + 1);
+		info->egress_timer_scheduled = true;
+	}
+}
+
+
+/* The "function" for "info->egress_timer".
+ *
+ * This timer will reschedule itself as long as there are any pending
+ * completions expected for this tile.
+ */
+static void tile_net_handle_egress_timer(unsigned long arg)
+{
+	struct tile_net_info *info = (struct tile_net_info *)arg;
+
+	unsigned int i;
+
+	bool pending = false;
+
+	unsigned long irqflags;
+
+	local_irq_save(irqflags);
+
+	/* The timer is no longer scheduled. */
+	info->egress_timer_scheduled = false;
+
+	/* Free all possible comps for this tile. */
+	for (i = 0; i < TILE_NET_CHANNELS; i++) {
+		struct tile_net_egress *egress = &egress_for_echannel[i];
+		struct tile_net_comps *comps = info->comps_for_echannel[i];
+		if (comps->comp_last >= comps->comp_next)
+			continue;
+		tile_net_free_comps(egress->equeue, comps, -1, true);
+		pending = pending || (comps->comp_last < comps->comp_next);
+	}
+
+	/* Reschedule timer if needed. */
+	if (pending)
+		tile_net_schedule_egress_timer(info);
+
+	local_irq_restore(irqflags);
+}
+
+
+/* Prepare each CPU. */
+static void tile_net_prepare_cpu(void *unused)
+{
+	struct tile_net_info *info = &__get_cpu_var(per_cpu_info);
+
+	int my_cpu = smp_processor_id();
+
+	info->has_iqueue = false;
+
+	info->my_cpu = my_cpu;
+
+	/* Initialize the egress timer. */
+	init_timer(&info->egress_timer);
+	info->egress_timer.data = (long)info;
+	info->egress_timer.function = tile_net_handle_egress_timer;
+}
+
+
+/* Helper function for "tile_net_update()". */
+static void tile_net_update_cpu(void *arg)
+{
+	struct net_device *dev = arg;
+
+	struct tile_net_info *info = &__get_cpu_var(per_cpu_info);
+
+	if (info->has_iqueue) {
+		if (dev != NULL) {
+			if (!info->napi_added) {
+				/* FIXME: HACK: We use one of the devices.
+				 * ISSUE: We never call "netif_napi_del()".
+				 */
+				netif_napi_add(dev, &info->napi,
+					       tile_net_poll, TILE_NET_WEIGHT);
+				info->napi_added = true;
+			}
+			if (!info->napi_enabled) {
+				napi_enable(&info->napi);
+				info->napi_enabled = true;
+			}
+			enable_percpu_irq(ingress_irq, 0);
+		} else {
+			disable_percpu_irq(ingress_irq);
+			if (info->napi_enabled) {
+				napi_disable(&info->napi);
+				info->napi_enabled = false;
+			}
+			/* FIXME: Drain the iqueue. */
+		}
+	}
+}
+
+
+/* Helper function for tile_net_open() and tile_net_stop(). */
+static int tile_net_update(void)
+{
+	struct net_device *dev = NULL;
+	int channel;
+	long count = 0;
+	int cpu;
+
+	/* HACK: This is too big for the linux stack. */
+	static gxio_mpipe_rules_t rules;
+
+	gxio_mpipe_rules_init(&rules, &context);
+
+	/* TODO: Add support for "dmac" splitting? */
+	for (channel = 0; channel < TILE_NET_CHANNELS; channel++) {
+		if (tile_net_devs_for_channel[channel] == NULL)
+			continue;
+		if (dev == NULL) {
+			dev = tile_net_devs_for_channel[channel];
+			gxio_mpipe_rules_begin(&rules, first_bucket,
+					       num_buckets, NULL);
+			gxio_mpipe_rules_set_headroom(&rules, NET_IP_ALIGN);
+		}
+		gxio_mpipe_rules_add_channel(&rules, channel);
+	}
+
+	/* NOTE: This can happen if there is no classifier.
+	 * ISSUE: Can anything else cause it to happen?
+	 */
+	if (gxio_mpipe_rules_commit(&rules) != 0) {
+		pr_warning("Failed to update classifier rules!\n");
+		return -EIO;
+	}
+
+	/* Update all cpus, sequentially (to protect "netif_napi_add()"). */
+	for_each_online_cpu(cpu)
+		smp_call_function_single(cpu, tile_net_update_cpu, dev, 1);
+
+	/* HACK: Allow packets to flow. */
+	if (count != 0)
+		sim_enable_mpipe_links(0, -1);
+
+	return 0;
+}
+
+
+/* Helper function for "tile_net_init_cpus()". */
+static void tile_net_init_stacks(int network_cpus_count)
+{
+	int err;
+	int i;
+
+	gxio_mpipe_buffer_size_enum_t small_buf_size =
+		GXIO_MPIPE_BUFFER_SIZE_128;
+	gxio_mpipe_buffer_size_enum_t large_buf_size =
+		GXIO_MPIPE_BUFFER_SIZE_1664;
+
+	int num_buffers;
+
+	size_t stack_bytes;
+
+	pte_t pte = { 0 };
+
+	void *mem;
+
+	num_buffers =
+		network_cpus_count * (IQUEUE_ENTRIES + TILE_NET_BATCH);
+
+	/* Compute stack bytes, honoring the 64KB minimum alignment. */
+	stack_bytes = ROUND_UP(gxio_mpipe_calc_buffer_stack_bytes(num_buffers),
+			       64 * 1024);
+	if (stack_bytes > HPAGE_SIZE)
+		panic("Cannot allocate %d physically contiguous buffers.",
+		      num_buffers);
+
+#if 0
+	sim_printf("Using %d buffers for %d network cpus.\n",
+		   num_buffers, network_cpus_count);
+#endif
+
+	/* Allocate two buffer stacks. */
+	small_buffer_stack = gxio_mpipe_alloc_buffer_stacks(&context, 2, 0, 0);
+	if (small_buffer_stack < 0)
+		panic("Failure in 'gxio_mpipe_alloc_buffer_stacks()'");
+	large_buffer_stack = small_buffer_stack + 1;
+
+	/* Allocate the small memory stack. */
+	mem = alloc_pages_exact(stack_bytes, GFP_KERNEL);
+	if (mem == NULL)
+		panic("Could not allocate buffer memory!");
+	err = gxio_mpipe_init_buffer_stack(&context, small_buffer_stack,
+					   small_buf_size,
+					   mem, stack_bytes, 0);
+	if (err != 0)
+		panic("Error %d in 'gxio_mpipe_init_buffer_stack()'.", err);
+
+	/* Allocate the large buffer stack. */
+	mem = alloc_pages_exact(stack_bytes, GFP_KERNEL);
+	if (mem == NULL)
+		panic("Could not allocate buffer memory!");
+	err = gxio_mpipe_init_buffer_stack(&context, large_buffer_stack,
+					   large_buf_size,
+					   mem, stack_bytes, 0);
+	if (err != 0)
+		panic("Error %d in 'gxio_mpipe_init_buffer_stack()'.", err);
+
+	/* Pin all the client memory. */
+	pte = pte_set_home(pte, PAGE_HOME_HASH);
+	err = gxio_mpipe_register_client_memory(&context, small_buffer_stack,
+						pte, 0);
+	if (err != 0)
+		panic("Error %d in 'gxio_mpipe_register_buffer_memory()'.",
+		      err);
+	err = gxio_mpipe_register_client_memory(&context, large_buffer_stack,
+						pte, 0);
+	if (err != 0)
+		panic("Error %d in 'gxio_mpipe_register_buffer_memory()'.",
+		      err);
+
+	/* Provide initial buffers. */
+	for (i = 0; i < num_buffers; i++) {
+		if (!tile_net_provide_buffer(true))
+			panic("Cannot provide initial buffers!");
+	}
+	for (i = 0; i < num_buffers; i++) {
+		if (!tile_net_provide_buffer(false))
+			panic("Cannot provide initial buffers!");
+	}
+}
+
+
+/* Actually initialize the mPIPE state. */
+static int tile_net_init_cpus(void)
+{
+	int network_cpus_count;
+
+	int ring;
+	int group;
+
+	int next_ring;
+
+	int cpu;
+
+	int i;
+
+#ifdef TILE_NET_ROUND_ROBIN
+	gxio_mpipe_bucket_mode_t mode = GXIO_MPIPE_BUCKET_ROUND_ROBIN;
+#else
+	/* Use random rebalancing. */
+	gxio_mpipe_bucket_mode_t mode = GXIO_MPIPE_BUCKET_STICKY_FLOW_LOCALITY;
+#endif
+
+	if (!hash_default) {
+		pr_warning("Networking requires hash_default!\n");
+		goto fail;
+	}
+
+	if (gxio_mpipe_init(&context, 0) != 0) {
+		pr_warning("Failed to initialize mPIPE!\n");
+		goto fail;
+	}
+
+	network_cpus_count = cpus_weight(network_cpus_map);
+
+	/* ISSUE: Handle failures more gracefully. */
+	tile_net_init_stacks(network_cpus_count);
+
+	/* Allocate one NotifRing for each network cpu. */
+	ring = gxio_mpipe_alloc_notif_rings(&context, network_cpus_count,
+					    0, 0);
+	if (ring < 0) {
+		pr_warning("Failed to allocate notif rings.\n");
+		goto fail;
+	}
+
+	/* ISSUE: Handle failures below more cleanly. */
+
+	/* Init NotifRings. */
+	next_ring = ring;
+
+	for_each_online_cpu(cpu) {
+
+		size_t notif_ring_size =
+			IQUEUE_ENTRIES * sizeof(gxio_mpipe_idesc_t);
+
+		int order;
+		struct page *page;
+		void *addr;
+
+		struct tile_net_info *info = &per_cpu(per_cpu_info, cpu);
+
+		/* ISSUE: This is overkill. */
+		size_t comps_size =
+			TILE_NET_CHANNELS * sizeof(struct tile_net_comps);
+
+		/* Allocate the "comps". */
+		order = get_order(comps_size);
+		page = homecache_alloc_pages(GFP_KERNEL, order, cpu);
+		if (page == NULL)
+			panic("Failed to allocate comps memory.");
+		addr = pfn_to_kaddr(page_to_pfn(page));
+		/* ISSUE: Is this needed? */
+		memset(addr, 0, comps_size);
+		for (i = 0; i < TILE_NET_CHANNELS; i++)
+			info->comps_for_echannel[i] =
+				addr + i * sizeof(struct tile_net_comps);
+
+		/* Only network cpus can receive packets. */
+		if (!cpu_isset(cpu, network_cpus_map))
+			continue;
+
+		/* Allocate the actual idescs array. */
+		order = get_order(notif_ring_size);
+		page = homecache_alloc_pages(GFP_KERNEL, order, cpu);
+		if (page == NULL)
+			panic("Failed to allocate iqueue memory.");
+		addr = pfn_to_kaddr(page_to_pfn(page));
+
+		if (gxio_mpipe_iqueue_init(&info->iqueue, &context, next_ring,
+					   addr, notif_ring_size, 0) != 0)
+			panic("Failure in 'gxio_mpipe_iqueue_init()'.");
+
+		info->has_iqueue = true;
+
+		next_ring++;
+	}
+
+	/* Allocate one NotifGroup. */
+	group = gxio_mpipe_alloc_notif_groups(&context, 1, 0, 0);
+	if (group < 0)
+		panic("Failure in 'gxio_mpipe_alloc_notif_groups()'.");
+
+#ifndef TILE_NET_ROUND_ROBIN
+	if (network_cpus_count > 4)
+		num_buckets = 256;
+	else if (network_cpus_count > 1)
+		num_buckets = 16;
+#endif
+
+	/* Allocate some buckets. */
+	first_bucket = gxio_mpipe_alloc_buckets(&context, num_buckets, 0, 0);
+	if (first_bucket < 0)
+		panic("Failure in 'gxio_mpipe_alloc_buckets()'.");
+
+	/* Init group and buckets. */
+	if (gxio_mpipe_init_notif_group_and_buckets(&context, group, ring,
+						    network_cpus_count,
+						    first_bucket, num_buckets,
+						    mode) != 0)
+		panic("Fail in 'gxio_mpipe_init_notif_group_and_buckets().");
+
+
+	/* Create an irq and register it. */
+	ingress_irq = create_irq();
+	if (ingress_irq < 0)
+		panic("Failed to create irq for ingress.");
+	tile_irq_activate(ingress_irq, TILE_IRQ_PERCPU);
+	BUG_ON(request_irq(ingress_irq, tile_net_handle_ingress_irq,
+			   0, NULL, NULL) != 0);
+
+	for_each_online_cpu(cpu) {
+
+		struct tile_net_info *info = &per_cpu(per_cpu_info, cpu);
+
+		int ring = info->iqueue.ring;
+
+		if (!info->has_iqueue)
+			continue;
+
+		gxio_mpipe_request_notif_ring_interrupt(&context,
+							cpu_x(cpu), cpu_y(cpu),
+							1, ingress_irq, ring);
+	}
+
+	return 0;
+
+fail:
+	return -EIO;
+}
+
+
+/* Create persistent egress info for a given egress channel.
+ *
+ * Note that this may be shared between, say, "gbe0" and "xgbe0".
+ *
+ * ISSUE: Defer header allocation until TSO is actually needed?
+ */
+static int tile_net_init_egress(int echannel)
+{
+	size_t headers_order;
+	struct page *headers_page;
+	unsigned char* headers;
+
+	size_t edescs_size;
+	int edescs_order;
+	struct page *edescs_page;
+	gxio_mpipe_edesc_t* edescs;
+
+	int equeue_order;
+	struct page *equeue_page;
+	gxio_mpipe_equeue_t* equeue;
+	int edma;
+
+	/* Only initialize once. */
+	if (egress_for_echannel[echannel].equeue != NULL)
+		return 0;
+
+	/* Allocate memory for the "headers". */
+	headers_order = get_order(EQUEUE_ENTRIES * HEADER_BYTES);
+	headers_page = alloc_pages(GFP_KERNEL, headers_order);
+	if (headers_page == NULL) {
+		pr_warning("Could not allocate memory for TSO headers.\n");
+		goto fail;
+	}
+	headers = pfn_to_kaddr(page_to_pfn(headers_page));
+
+	/* Allocate memory for the "edescs". */
+	edescs_size = EQUEUE_ENTRIES * sizeof(*edescs);
+	edescs_order = get_order(edescs_size);
+	edescs_page = alloc_pages(GFP_KERNEL, edescs_order);
+	if (edescs_page == NULL) {
+		pr_warning("Could not allocate memory for eDMA ring.\n");
+		goto fail_headers;
+	}
+	edescs = pfn_to_kaddr(page_to_pfn(edescs_page));
+
+	/* Allocate memory for the "equeue". */
+	equeue_order = get_order(sizeof(*equeue));
+	equeue_page = alloc_pages(GFP_KERNEL, equeue_order);
+	if (equeue_page == NULL) {
+		pr_warning("Could not allocate memory for equeue info.\n");
+		goto fail_edescs;
+	}
+	equeue = pfn_to_kaddr(page_to_pfn(equeue_page));
+
+	/* Allocate an edma ring. */
+	edma = gxio_mpipe_alloc_edma_rings(&context, 1, 0, 0);
+	if (edma < 0) {
+		pr_warning("Could not allocate edma ring.\n");
+		goto fail_equeue;
+	}
+
+	/* Initialize the equeue.  This should not fail. */
+	if (gxio_mpipe_equeue_init(equeue, &context, edma, echannel,
+				   edescs, edescs_size, 0) != 0)
+		panic("Failure in 'gxio_mpipe_equeue_init()'.");
+
+	/* Done. */
+	egress_for_echannel[echannel].equeue = equeue;
+	egress_for_echannel[echannel].headers = headers;
+	return 0;
+
+fail_equeue:
+	__free_pages(equeue_page, equeue_order);
+
+fail_edescs:
+	__free_pages(edescs_page, edescs_order);
+
+fail_headers:
+	__free_pages(headers_page, headers_order);
+
+fail:
+	return -EIO;
+}
+
+
+/* Help the kernel activate the given network interface. */
+static int tile_net_open(struct net_device *dev)
+{
+	struct tile_net_priv *priv = netdev_priv(dev);
+
+	/* Determine if this is the "loopify" device. */
+	bool loopify = (loopify_link_name != NULL) &&
+		!strcmp(dev->name, loopify_link_name);
+
+	int result;
+
+	mutex_lock(&tile_net_devs_for_channel_mutex);
+
+	if (ingress_irq < 0) {
+		result = tile_net_init_cpus();
+		if (result != 0)
+			goto fail;
+	}
+
+	if (priv->channel < 0) {
+		const char* ln = loopify ? "loop0" : dev->name;
+		if (gxio_mpipe_link_open(&priv->link, &context, ln, 0) < 0) {
+			netdev_err(dev, "Failed to open '%s'.\n", ln);
+			result = -EIO;
+			goto fail;
+		}
+		priv->channel = gxio_mpipe_link_channel(&priv->link);
+		BUG_ON(priv->channel < 0 ||
+		       priv->channel >= TILE_NET_CHANNELS);
+	}
+
+	if (loopify && priv->loopify_channel < 0) {
+		if (gxio_mpipe_link_open(&priv->loopify_link,
+					 &context, "loop1", 0) < 0) {
+			netdev_err(dev, "Failed to open 'loop1'.\n");
+			result = -EIO;
+			goto fail;
+		}
+		priv->loopify_channel =
+			gxio_mpipe_link_channel(&priv->loopify_link);
+		BUG_ON(priv->loopify_channel < 0 ||
+			priv->loopify_channel >= TILE_NET_CHANNELS);
+	}
+
+	priv->echannel =
+		((priv->loopify_channel >= 0) ?
+		 priv->loopify_channel : priv->channel);
+
+	/* Initialize egress info (if needed). */
+	result = tile_net_init_egress(priv->echannel);
+	if (result != 0)
+		goto fail;
+
+	tile_net_devs_for_channel[priv->channel] = dev;
+
+	result = tile_net_update();
+	if (result != 0)
+		goto fail_channel;
+
+	mutex_unlock(&tile_net_devs_for_channel_mutex);
+
+	/* Start our transmit queue. */
+	netif_start_queue(dev);
+
+	netif_carrier_on(dev);
+
+	return 0;
+
+fail_channel:
+	tile_net_devs_for_channel[priv->channel] = NULL;
+
+fail:
+	if (priv->loopify_channel >= 0) {
+		if (gxio_mpipe_link_close(&priv->loopify_link) != 0)
+			pr_warning("Failed to close loopify link!\n");
+		else
+			priv->loopify_channel = -1;
+	}
+	if (priv->channel >= 0) {
+		if (gxio_mpipe_link_close(&priv->link) != 0)
+			pr_warning("Failed to close link!\n");
+		else
+			priv->channel = -1;
+	}
+
+	priv->echannel = -1;
+
+	mutex_unlock(&tile_net_devs_for_channel_mutex);
+	return result;
+}
+
+
+
+/* Help the kernel deactivate the given network interface. */
+static int tile_net_stop(struct net_device *dev)
+{
+	struct tile_net_priv *priv = netdev_priv(dev);
+
+	/* Stop our transmit queue. */
+	netif_stop_queue(dev);
+
+	mutex_lock(&tile_net_devs_for_channel_mutex);
+
+	tile_net_devs_for_channel[priv->channel] = NULL;
+
+	(void)tile_net_update();
+
+	if (priv->loopify_channel >= 0) {
+		if (gxio_mpipe_link_close(&priv->loopify_link) != 0)
+			pr_warning("Failed to close loopify link!\n");
+		priv->loopify_channel = -1;
+	}
+
+	if (priv->channel >= 0) {
+		if (gxio_mpipe_link_close(&priv->link) != 0)
+			pr_warning("Failed to close link!\n");
+		priv->channel = -1;
+	}
+
+	priv->echannel = -1;
+
+	mutex_unlock(&tile_net_devs_for_channel_mutex);
+
+	return 0;
+}
+
+
+/* Determine the VA for a fragment. */
+static inline void *tile_net_frag_buf(skb_frag_t *f)
+{
+	unsigned long pfn = page_to_pfn(skb_frag_page(f));
+	return pfn_to_kaddr(pfn) + f->page_offset;
+}
+
+
+/* This function takes "skb", consisting of a header template and a
+ * (presumably) huge payload, and egresses it as one or more segments
+ * (aka packets), each consisting of a (possibly modified) copy of the
+ * header plus a piece of the payload, via "tcp segmentation offload".
+ *
+ * Usually, "data" will contain the header template, of size "sh_len",
+ * and "sh->frags" will contain "skb->data_len" bytes of payload, and
+ * there will be "sh->gso_segs" segments.
+ *
+ * Sometimes, if "sendfile()" requires copying, we will be called with
+ * "data" containing the header and payload, with "frags" being empty.
+ *
+ * Sometimes, for example when using NFS over TCP, a single segment can
+ * span 3 fragments.  This requires special care below.
+ *
+ * See "emulate_large_send_offload()" for some reference code, which
+ * does not handle checksumming.
+ */
+static int tile_net_tx_tso(struct sk_buff *skb, struct net_device *dev)
+{
+	struct tile_net_priv *priv = netdev_priv(dev);
+
+	struct tile_net_info *info = &__get_cpu_var(per_cpu_info);
+
+	struct tile_net_egress *egress = &egress_for_echannel[priv->echannel];
+	gxio_mpipe_equeue_t *equeue = egress->equeue;
+
+	struct tile_net_comps *comps =
+		info->comps_for_echannel[priv->echannel];
+
+	unsigned int len = skb->len;
+	unsigned char *data = skb->data;
+
+	/* The ip header follows the ethernet header. */
+	struct iphdr *ih = ip_hdr(skb);
+	unsigned int ih_len = ih->ihl * 4;
+
+	/* Note that "nh == iph", by definition. */
+	unsigned char *nh = skb_network_header(skb);
+	unsigned int eh_len = nh - data;
+
+	/* The tcp header follows the ip header. */
+	struct tcphdr *th = (struct tcphdr *)(nh + ih_len);
+	unsigned int th_len = th->doff * 4;
+
+	/* The total number of header bytes. */
+	unsigned int sh_len = eh_len + ih_len + th_len;
+
+	/* Help compute "jh->check". */
+	unsigned int isum_hack =
+		((0xFFFF - ih->check) +
+		 (0xFFFF - ih->tot_len) +
+		 (0xFFFF - ih->id));
+
+	/* Help compute "uh->check". */
+	unsigned int tsum_hack = th->check + (0xFFFF ^ htons(len));
+
+	struct skb_shared_info *sh = skb_shinfo(skb);
+
+	/* The maximum payload size. */
+	unsigned int gso_size = sh->gso_size;
+
+	/* The size of the initial segments (including header). */
+	unsigned int mtu = sh_len + gso_size;
+
+	/* The size of the final segment (including header). */
+	unsigned int mtu2 = len - ((sh->gso_segs - 1) * gso_size);
+
+	/* Track tx stats. */
+	unsigned int tx_packets = 0;
+	unsigned int tx_bytes = 0;
+
+	/* Which segment are we on. */
+	unsigned int segment;
+
+	/* Get the initial ip "id". */
+	u16 id = ntohs(ih->id);
+
+	/* Get the initial tcp "seq". */
+	u32 seq = ntohl(th->seq);
+
+	/* The id of the current fragment (or -1). */
+	long f_id;
+
+	/* The size of the current fragment (or -1). */
+	long f_size;
+
+	/* The bytes used from the current fragment (or -1). */
+	long f_used;
+
+	/* The size of the current piece of payload. */
+	long n;
+
+	/* Prepare checksum info. */
+	unsigned int csum_start = skb_checksum_start_offset(skb);
+
+	/* The header/payload edesc's. */
+	gxio_mpipe_edesc_t edesc_head = { { 0 } };
+	gxio_mpipe_edesc_t edesc_body = { { 0 } };
+
+	/* Total number of edescs needed. */
+	unsigned int num_edescs = 0;
+
+	unsigned long irqflags;
+
+	/* First reserved egress slot. */
+	s64 slot;
+
+	int cid;
+
+	/* Empty packets (etc) would cause trouble below. */
+	BUG_ON(skb->data_len == 0);
+	BUG_ON(sh->nr_frags == 0);
+	BUG_ON(sh->gso_segs == 0);
+
+	/* We assume the frags contain the entire payload. */
+	BUG_ON(skb_headlen(skb) != sh_len);
+	BUG_ON(len != sh_len + skb->data_len);
+
+	/* Implicitly verify "gso_segs" and "gso_size". */
+	BUG_ON(mtu2 > mtu);
+
+	/* We only have HEADER_BYTES for each header. */
+	BUG_ON(NET_IP_ALIGN + sh_len > HEADER_BYTES);
+
+	/* Paranoia. */
+	BUG_ON(skb->protocol != htons(ETH_P_IP));
+	BUG_ON(ih->protocol != IPPROTO_TCP);
+	BUG_ON(skb->ip_summed != CHECKSUM_PARTIAL);
+	BUG_ON(csum_start != eh_len + ih_len);
+
+	/* NOTE: ".hwb = 0", so ".size" is unused.
+	 * NOTE: ".stack_idx" determines the TLB.
+	 */
+
+	/* Prepare to egress the headers. */
+	edesc_head.csum = 1;
+	edesc_head.csum_start = csum_start;
+	edesc_head.csum_dest = csum_start + skb->csum_offset;
+	edesc_head.xfer_size = sh_len;
+	edesc_head.stack_idx = large_buffer_stack;
+
+	/* Prepare to egress the body. */
+	edesc_body.stack_idx = large_buffer_stack;
+
+	/* Reset. */
+	f_id = f_size = f_used = -1;
+
+	/* Determine how many edesc's are needed. */
+	for (segment = 0; segment < sh->gso_segs; segment++) {
+
+		/* Detect the final segment. */
+		bool final = (segment == sh->gso_segs - 1);
+
+		/* The segment size (including header). */
+		unsigned int s_len = final ? mtu2 : mtu;
+
+		/* The size of the payload. */
+		unsigned int p_len = s_len - sh_len;
+
+		/* The bytes used from the payload. */
+		unsigned int p_used = 0;
+
+		/* One edesc for the header. */
+		num_edescs++;
+
+		/* One edesc for each piece of the payload. */
+		while (p_used < p_len) {
+
+			/* Advance as needed. */
+			while (f_used >= f_size) {
+				f_id++;
+				f_size = sh->frags[f_id].size;
+				f_used = 0;
+			}
+
+			/* Use bytes from the current fragment. */
+			n = p_len - p_used;
+			if (n > f_size - f_used)
+				n = f_size - f_used;
+			f_used += n;
+			p_used += n;
+
+			num_edescs++;
+		}
+	}
+
+	/* Verify all fragments consumed. */
+	BUG_ON(f_id + 1 != sh->nr_frags);
+	BUG_ON(f_used != f_size);
+
+	local_irq_save(irqflags);
+
+	/* Reserve slots, or return NETDEV_TX_BUSY if "full". */
+	slot = gxio_mpipe_equeue_try_reserve(equeue, num_edescs);
+	if (slot < 0) {
+		local_irq_restore(irqflags);
+		/* ISSUE: "Virtual device xxx asks to queue packet". */
+		return NETDEV_TX_BUSY;
+	}
+
+	/* Reset. */
+	f_id = f_size = f_used = -1;
+
+	/* Prepare all the headers. */
+	for (segment = 0; segment < sh->gso_segs; segment++) {
+
+		/* Detect the final segment. */
+		bool final = (segment == sh->gso_segs - 1);
+
+		/* The segment size (including header). */
+		unsigned int s_len = final ? mtu2 : mtu;
+
+		/* The size of the payload. */
+		unsigned int p_len = s_len - sh_len;
+
+		/* The bytes used from the payload. */
+		unsigned int p_used = 0;
+
+		/* Access the header memory for this segment. */
+		unsigned int bn = slot % EQUEUE_ENTRIES;
+		unsigned char *buf =
+			egress->headers + bn * HEADER_BYTES + NET_IP_ALIGN;
+
+		/* The soon-to-be copied "ip" header. */
+		struct iphdr *jh = (struct iphdr *)(buf + eh_len);
+
+		/* The soon-to-be copied "tcp" header. */
+		struct tcphdr *uh = (struct tcphdr *)(buf + eh_len + ih_len);
+
+		unsigned int jsum;
+
+		/* Copy the header. */
+		memcpy(buf, data, sh_len);
+
+		/* The packet size, not including ethernet header. */
+		jh->tot_len = htons(s_len - eh_len);
+
+		/* Update the ip "id". */
+		jh->id = htons(id);
+
+		/* Compute the "ip checksum". */
+		jsum = isum_hack + htons(s_len - eh_len) + htons(id);
+		jh->check = csum_long(jsum) ^ 0xffff;
+
+		/* Update the tcp "seq". */
+		uh->seq = htonl(seq);
+
+		/* Update some flags. */
+		if (!final)
+			uh->fin = uh->psh = 0;
+
+		/* Compute the tcp pseudo-header checksum. */
+		uh->check = csum_long(tsum_hack + htons(s_len));
+
+		/* Skip past the header. */
+		slot++;
+
+		/* Skip past the payload. */
+		while (p_used < p_len) {
+
+			/* Advance as needed. */
+			while (f_used >= f_size) {
+				f_id++;
+				f_size = sh->frags[f_id].size;
+				f_used = 0;
+			}
+
+			/* Use bytes from the current fragment. */
+			n = p_len - p_used;
+			if (n > f_size - f_used)
+				n = f_size - f_used;
+			f_used += n;
+			p_used += n;
+
+			slot++;
+		}
+
+		id++;
+		seq += p_len;
+	}
+
+	/* Reset "slot". */
+	slot -= num_edescs;
+
+	/* Flush the headers. */
+	wmb();
+
+	/* Reset. */
+	f_id = f_size = f_used = -1;
+
+	/* Egress all the edescs. */
+	for (segment = 0; segment < sh->gso_segs; segment++) {
+
+		/* Detect the final segment. */
+		bool final = (segment == sh->gso_segs - 1);
+
+		/* The segment size (including header). */
+		unsigned int s_len = final ? mtu2 : mtu;
+
+		/* The size of the payload. */
+		unsigned int p_len = s_len - sh_len;
+
+		/* The bytes used from the payload. */
+		unsigned int p_used = 0;
+
+		/* Access the header memory for this segment. */
+		unsigned int bn = slot % EQUEUE_ENTRIES;
+		unsigned char *buf =
+			egress->headers + bn * HEADER_BYTES + NET_IP_ALIGN;
+
+		void *va;
+
+		/* Egress the header. */
+		edesc_head.va = va_to_tile_io_addr(buf);
+		gxio_mpipe_equeue_put_at(equeue, edesc_head, slot);
+		slot++;
+
+		/* Egress the payload. */
+		while (p_used < p_len) {
+
+			/* Advance as needed. */
+			while (f_used >= f_size) {
+				f_id++;
+				f_size = sh->frags[f_id].size;
+				f_used = 0;
+			}
+
+			va = tile_net_frag_buf(&sh->frags[f_id]) + f_used;
+
+			/* Use bytes from the current fragment. */
+			n = p_len - p_used;
+			if (n > f_size - f_used)
+				n = f_size - f_used;
+			f_used += n;
+			p_used += n;
+
+			/* Egress a piece of the payload. */
+			edesc_body.va = va_to_tile_io_addr(va);
+			edesc_body.xfer_size = n;
+			edesc_body.bound = !(p_used < p_len);
+			gxio_mpipe_equeue_put_at(equeue, edesc_body, slot);
+			slot++;
+		}
+
+		tx_packets++;
+		tx_bytes += s_len;
+	}
+
+	/* Wait for a free completion entry.
+	 * ISSUE: Is this the best logic?
+	 * ISSUE: Can this cause undesirable "blocking"?
+	 */
+	while (comps->comp_next - comps->comp_last >= TILE_NET_MAX_COMPS - 1)
+		tile_net_free_comps(equeue, comps, 32, false);
+
+	/* Update the completions array. */
+	cid = comps->comp_next % TILE_NET_MAX_COMPS;
+	comps->comp_queue[cid].when = slot;
+	comps->comp_queue[cid].skb = skb;
+	comps->comp_next++;
+
+	/* Update stats. */
+	atomic_add(tx_packets, (atomic_t *)&priv->stats.tx_packets);
+	atomic_add(tx_bytes, (atomic_t *)&priv->stats.tx_bytes);
+
+	local_irq_restore(irqflags);
+
+	/* Make sure the egress timer is scheduled. */
+	tile_net_schedule_egress_timer(info);
+
+	return NETDEV_TX_OK;
+}
+
+
+/* Analyze the body and frags for a transmit request. */
+static unsigned int tile_net_tx_frags(struct frag *frags,
+				       struct sk_buff *skb,
+				       void *b_data, unsigned int b_len)
+{
+	unsigned int i, n = 0;
+
+	struct skb_shared_info *sh = skb_shinfo(skb);
+
+	if (b_len != 0) {
+		frags[n].buf = b_data;
+		frags[n++].length = b_len;
+	}
+
+	for (i = 0; i < sh->nr_frags; i++) {
+		skb_frag_t *f = &sh->frags[i];
+		frags[n].buf = tile_net_frag_buf(f);
+		frags[n++].length = skb_frag_size(f);
+	}
+
+	return n;
+}
+
+
+/* Help the kernel transmit a packet. */
+static int tile_net_tx(struct sk_buff *skb, struct net_device *dev)
+{
+	struct tile_net_priv *priv = netdev_priv(dev);
+
+	struct tile_net_info *info = &__get_cpu_var(per_cpu_info);
+
+	struct tile_net_egress *egress = &egress_for_echannel[priv->echannel];
+	gxio_mpipe_equeue_t *equeue = egress->equeue;
+
+	struct tile_net_comps *comps =
+		info->comps_for_echannel[priv->echannel];
+
+	struct skb_shared_info *sh = skb_shinfo(skb);
+
+	unsigned int len = skb->len;
+	unsigned char *data = skb->data;
+
+	unsigned int num_frags;
+	struct frag frags[MAX_FRAGS];
+	gxio_mpipe_edesc_t edescs[MAX_FRAGS];
+
+	unsigned int i;
+
+	int cid;
+
+	s64 slot;
+
+	unsigned long irqflags;
+
+	/* Save the timestamp. */
+	dev->trans_start = jiffies;
+
+#ifdef TILE_NET_DUMP_PACKETS
+	/* ISSUE: Does not dump the "frags". */
+	dump_packet(data, skb_headlen(skb), "tx");
+#endif /* TILE_NET_DUMP_PACKETS */
+
+	if (sh->gso_size != 0)
+		return tile_net_tx_tso(skb, dev);
+
+	/* NOTE: This is usually 2, sometimes 3, for big writes. */
+	num_frags = tile_net_tx_frags(frags, skb, data, skb_headlen(skb));
+
+	/* Prepare the edescs. */
+	for (i = 0; i < num_frags; i++) {
+
+		/* NOTE: ".hwb = 0", so ".size" is unused.
+		 * NOTE: ".stack_idx" determines the TLB.
+		 */
+
+		gxio_mpipe_edesc_t edesc = { { 0 } };
+
+		/* Prepare the basic command. */
+		edesc.bound = (i == num_frags - 1);
+		edesc.xfer_size = frags[i].length;
+		edesc.va = va_to_tile_io_addr(frags[i].buf);
+		edesc.stack_idx = large_buffer_stack;
+
+		edescs[i] = edesc;
+	}
+
+	/* Add checksum info if needed. */
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		unsigned int csum_start = skb->csum_start - skb_headroom(skb);
+		edescs[0].csum = 1;
+		edescs[0].csum_start = csum_start;
+		edescs[0].csum_dest = csum_start + skb->csum_offset;
+	}
+
+	local_irq_save(irqflags);
+
+	/* Reserve slots, or return NETDEV_TX_BUSY if "full". */
+	slot = gxio_mpipe_equeue_try_reserve(equeue, num_frags);
+	if (slot < 0) {
+		local_irq_restore(irqflags);
+		/* ISSUE: "Virtual device xxx asks to queue packet". */
+		return NETDEV_TX_BUSY;
+	}
+
+	for (i = 0; i < num_frags; i++)
+		gxio_mpipe_equeue_put_at(equeue, edescs[i], slot + i);
+
+	/* Wait for a free completion entry.
+	 * ISSUE: Is this the best logic?
+	 * ISSUE: Can this cause undesirable "blocking"?
+	 */
+	while (comps->comp_next - comps->comp_last >= TILE_NET_MAX_COMPS - 1)
+		tile_net_free_comps(equeue, comps, 32, false);
+
+	/* Update the completions array. */
+	cid = comps->comp_next % TILE_NET_MAX_COMPS;
+	comps->comp_queue[cid].when = slot + num_frags;
+	comps->comp_queue[cid].skb = skb;
+	comps->comp_next++;
+
+	/* HACK: Track "expanded" size for short packets (e.g. 42 < 60). */
+	atomic_add(1, (atomic_t *)&priv->stats.tx_packets);
+	atomic_add((len >= ETH_ZLEN) ? len : ETH_ZLEN,
+		   (atomic_t *)&priv->stats.tx_bytes);
+
+	local_irq_restore(irqflags);
+
+	/* Make sure the egress timer is scheduled. */
+	tile_net_schedule_egress_timer(info);
+
+	return NETDEV_TX_OK;
+}
+
+
+/* Deal with a transmit timeout. */
+static void tile_net_tx_timeout(struct net_device *dev)
+{
+	/* ISSUE: This doesn't seem useful for us. */
+	netif_wake_queue(dev);
+}
+
+
+/* Ioctl commands. */
+static int tile_net_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
+{
+	return -EOPNOTSUPP;
+}
+
+
+/* Get System Network Statistics.
+ *
+ * Returns the address of the device statistics structure.
+ */
+static struct net_device_stats *tile_net_get_stats(struct net_device *dev)
+{
+	struct tile_net_priv *priv = netdev_priv(dev);
+	return &priv->stats;
+}
+
+
+/* Change the "mtu". */
+static int tile_net_change_mtu(struct net_device *dev, int new_mtu)
+{
+	/* Check ranges. */
+	if ((new_mtu < 68) || (new_mtu > 1500))
+		return -EINVAL;
+
+	/* Accept the value. */
+	dev->mtu = new_mtu;
+
+	return 0;
+}
+
+
+/* Change the Ethernet Address of the NIC.
+ *
+ * The hypervisor driver does not support changing MAC address.  However,
+ * the hardware does not do anything with the MAC address, so the address
+ * which gets used on outgoing packets, and which is accepted on incoming
+ * packets, is completely up to us.
+ *
+ * Returns 0 on success, negative on failure.
+ */
+static int tile_net_set_mac_address(struct net_device *dev, void *p)
+{
+	struct sockaddr *addr = p;
+
+	if (!is_valid_ether_addr(addr->sa_data))
+		return -EINVAL;
+
+	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
+
+	return 0;
+}
+
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+/* Polling 'interrupt' - used by things like netconsole to send skbs
+ * without having to re-enable interrupts. It's not called while
+ * the interrupt routine is executing.
+ */
+static void tile_net_netpoll(struct net_device *dev)
+{
+	disable_percpu_irq(ingress_irq);
+	tile_net_handle_ingress_irq(ingress_irq, NULL);
+	enable_percpu_irq(ingress_irq, 0);
+}
+#endif
+
+
+static const struct net_device_ops tile_net_ops = {
+	.ndo_open = tile_net_open,
+	.ndo_stop = tile_net_stop,
+	.ndo_start_xmit = tile_net_tx,
+	.ndo_do_ioctl = tile_net_ioctl,
+	.ndo_get_stats = tile_net_get_stats,
+	.ndo_change_mtu = tile_net_change_mtu,
+	.ndo_tx_timeout = tile_net_tx_timeout,
+	.ndo_set_mac_address = tile_net_set_mac_address,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller = tile_net_netpoll,
+#endif
+};
+
+/* The setup function.
+ *
+ * This uses ether_setup() to assign various fields in dev, including
+ * setting IFF_BROADCAST and IFF_MULTICAST, then sets some extra fields.
+ */
+static void tile_net_setup(struct net_device *dev)
+{
+	ether_setup(dev);
+
+	dev->netdev_ops      = &tile_net_ops;
+	dev->watchdog_timeo  = TILE_NET_TIMEOUT;
+
+	/* We want lockless xmit. */
+	dev->features |= NETIF_F_LLTX;
+
+	/* We support hardware tx checksums. */
+	dev->features |= NETIF_F_HW_CSUM;
+
+	/* We support scatter/gather. */
+	dev->features |= NETIF_F_SG;
+
+#ifdef TILE_NET_GSO
+	/* We support GSO. */
+	dev->features |= NETIF_F_GSO;
+#endif
+
+#ifdef TILE_NET_TSO
+	/* We support TSO. */
+	dev->features |= NETIF_F_TSO;
+#endif
+
+	dev->tx_queue_len = TILE_NET_TX_QUEUE_LEN;
+
+	dev->mtu = 1500;
+}
+
+
+/* Allocate the device structure, register the device, and obtain the
+ * MAC address from the hypervisor.
+ */
+static void tile_net_dev_init(const char *name, const uint8_t* mac)
+{
+	int ret;
+	int i;
+	int nz_addr = 0;
+	struct net_device *dev;
+	struct tile_net_priv *priv;
+
+	/* HACK: Ignore "loop" links. */
+	if (strncmp(name, "loop", 4) == 0)
+		return;
+
+	/* Allocate the device structure.  This allocates "priv", calls
+	 * tile_net_setup(), and saves "name".  Normally, "name" is a
+	 * template, instantiated by register_netdev(), but not for us.
+	 */
+	dev = alloc_netdev(sizeof(*priv), name, tile_net_setup);
+	if (!dev) {
+		pr_err("alloc_netdev(%s) failed\n", name);
+		return;
+	}
+
+	priv = netdev_priv(dev);
+
+	/* Initialize "priv". */
+
+	memset(priv, 0, sizeof(*priv));
+
+	priv->dev = dev;
+
+	priv->channel = priv->loopify_channel = priv->echannel = -1;
+
+	/* Register the network device. */
+	ret = register_netdev(dev);
+	if (ret) {
+		netdev_err(dev, "register_netdev failed %d\n", ret);
+		free_netdev(dev);
+		return;
+	}
+
+	/* Get the MAC address and set it in the device struct; this must
+	 * be done before the device is opened.  If the MAC is all zeroes,
+	 * we use a random address, since we're probably on the simulator.
+	 */
+	for (i = 0; i < 6; i++)
+		nz_addr |= mac[i];
+
+	if (nz_addr) {
+		memcpy(dev->dev_addr, mac, 6);
+		dev->addr_len = 6;
+	} else {
+		random_ether_addr(dev->dev_addr);
+	}
+}
+
+
+/* Module initialization. */
+static int __init tile_net_init_module(void)
+{
+	int i;
+	char name[GXIO_MPIPE_LINK_NAME_LEN];
+	uint8_t mac[6];
+
+	pr_info("Tilera Network Driver\n");
+
+	mutex_init(&tile_net_devs_for_channel_mutex);
+
+	/* Initialize each CPU. */
+	on_each_cpu(tile_net_prepare_cpu, NULL, 1);
+
+	/* Find out what devices we have, and initialize them. */
+	for (i = 0; gxio_mpipe_link_enumerate_mac(i, name, mac) >= 0; i++)
+		tile_net_dev_init(name, mac);
+
+	if (!network_cpus_init())
+		network_cpus_map = *cpu_online_mask;
+
+	return 0;
+}
+
+module_init(tile_net_init_module);