Hi,
Attached below is kernel patch with UPD large send offload which address the
sendfile() syscall also. This patch uses scatter-gather support of skb to
generate large UDP datagram.
Below is a "how-to" on changes required in network drivers to use the USO
interface.
UDP Large Send Offload (USO) Interface:
--------------------------------------
USO is a feature wherein the Linux kernel network stack will offload the IP
fragmentation functionality of large UDP datagram to hardware. This will reduce
the overhead of stack in fragmenting the large UDP datagram to MTU sized packets
1) Drivers indicate their capability of USO using
dev->features |= NETIF_F_USO | NETIF_F_HW_CSUM | NETIF_F_SG
NETIF_F_HW_CSUM is required for USO over ipv6.
2) USO packet will be submitted for transmission using driver xmit routine.
USO packet will have a non-zero value for
"skb_shinfo(skb)->uso_size"
skb_shinfo(skb)->uso_size will indicate the length of data part in each IP
fragment going out of the adapter after IP fragmentation by hardware.
skb->data will contain MAC/IP/UDP header and skb_shinfo(skb)->frags[]
contains the data payload. The skb->ip_summed will be set to CHECKSUM_HW
indicating that hardware has to do checksum calculation. Hardware should
compute the UDP checksum of complete datagram and also ip header checksum of
each fragmented IP packet.
For IPV6 the USO provides the fragment identification-id in
skb_shinfo(skb)->ip6_frag_id. The adapter should use this ID for generating
IPv6 fragments.
Signed-off-by: Ananda Raju <ananda.raju@xxxxxxxxxxxx>
---
diff -uNr linux-2.6.13/include/linux/ethtool.h
linux-2.6.13_uso/include/linux/ethtool.h
--- linux-2.6.13/include/linux/ethtool.h 2005-09-07 06:36:15.000000000
-0700
+++ linux-2.6.13_uso/include/linux/ethtool.h 2005-09-07 06:32:29.000000000
-0700
@@ -261,6 +261,8 @@
int ethtool_op_set_sg(struct net_device *dev, u32 data);
u32 ethtool_op_get_tso(struct net_device *dev);
int ethtool_op_set_tso(struct net_device *dev, u32 data);
+u32 ethtool_op_get_uso(struct net_device *dev);
+int ethtool_op_set_uso(struct net_device *dev, u32 data);
/**
* ðtool_ops - Alter and report network device settings
@@ -290,6 +292,8 @@
* set_sg: Turn scatter-gather on or off
* get_tso: Report whether TCP segmentation offload is enabled
* set_tso: Turn TCP segmentation offload on or off
+ * get_uso: Report whether UDP large send offload is enabled
+ * set_uso: Turn UDP large send offload on or off
* self_test: Run specified self-tests
* get_strings: Return a set of strings that describe the requested objects
* phys_id: Identify the device
@@ -354,6 +358,8 @@
void (*get_ethtool_stats)(struct net_device *, struct ethtool_stats
*, u64 *);
int (*begin)(struct net_device *);
void (*complete)(struct net_device *);
+ u32 (*get_uso)(struct net_device *);
+ int (*set_uso)(struct net_device *, u32);
};
/* CMDs currently supported */
@@ -389,6 +395,8 @@
#define ETHTOOL_GSTATS 0x0000001d /* get NIC-specific statistics */
#define ETHTOOL_GTSO 0x0000001e /* Get TSO enable (ethtool_value) */
#define ETHTOOL_STSO 0x0000001f /* Set TSO enable (ethtool_value) */
+#define ETHTOOL_GUSO 0x00000020 /* Get USO enable (ethtool_value) */
+#define ETHTOOL_SUSO 0x00000021 /* Set USO enable (ethtool_value) */
/* compatibility with older code */
#define SPARC_ETH_GSET ETHTOOL_GSET
diff -uNr linux-2.6.13/include/linux/netdevice.h
linux-2.6.13_uso/include/linux/netdevice.h
--- linux-2.6.13/include/linux/netdevice.h 2005-09-07 04:20:51.000000000
-0700
+++ linux-2.6.13_uso/include/linux/netdevice.h 2005-09-07 04:22:51.000000000
-0700
@@ -408,6 +408,7 @@
#define NETIF_F_VLAN_CHALLENGED 1024 /* Device cannot handle VLAN
packets */
#define NETIF_F_TSO 2048 /* Can offload TCP/IP segmentation */
#define NETIF_F_LLTX 4096 /* LockLess TX */
+#define NETIF_F_USO 8192 /* Can offload UDP Large Send*/
/* Called after device is detached from network. */
void (*uninit)(struct net_device *dev);
diff -uNr linux-2.6.13/include/linux/skbuff.h
linux-2.6.13_uso/include/linux/skbuff.h
--- linux-2.6.13/include/linux/skbuff.h 2005-09-07 04:20:56.000000000 -0700
+++ linux-2.6.13_uso/include/linux/skbuff.h 2005-09-07 04:22:58.000000000
-0700
@@ -137,6 +137,8 @@
unsigned int nr_frags;
unsigned short tso_size;
unsigned short tso_segs;
+ unsigned short uso_size;
+ unsigned int ip6_frag_id;
struct sk_buff *frag_list;
skb_frag_t frags[MAX_SKB_FRAGS];
};
@@ -327,6 +329,11 @@
extern void skb_under_panic(struct sk_buff *skb, int len,
void *here);
+extern int skb_append_datato_frags(struct sock *sk, struct sk_buff *skb,
+ int getfrag(void *from, char *to, int offset,
+ int len,int odd, struct sk_buff *skb),
+ void *from, int length);
+
struct skb_seq_state
{
__u32 lower_offset;
diff -uNr linux-2.6.13/net/core/dev.c linux-2.6.13_uso/net/core/dev.c
--- linux-2.6.13/net/core/dev.c 2005-09-07 06:36:25.000000000 -0700
+++ linux-2.6.13_uso/net/core/dev.c 2005-09-07 06:32:02.000000000 -0700
@@ -2706,6 +2706,25 @@
dev->name);
dev->features &= ~NETIF_F_TSO;
}
+ /* TSO requires that SG is present as well. */
+ if ((dev->features & NETIF_F_TSO) &&
+ !(dev->features & NETIF_F_SG)) {
+ printk("%s: Dropping NETIF_F_TSO since no SG feature.\n",
+ dev->name);
+ dev->features &= ~NETIF_F_TSO;
+ }
+ if (dev->features & NETIF_F_USO) {
+ if (!(dev->features & NETIF_F_HW_CSUM)) {
+ printk("%s: Dropping NETIF_F_USO since no ", dev->name);
+ printk("NETIF_F_HW_CSUM feature.\n");
+ dev->features &= ~NETIF_F_USO;
+ }
+ if (!(dev->features & NETIF_F_SG)) {
+ printk("%s: Dropping NETIF_F_USO since no ", dev->name);
+ printk("NETIF_F_SG feature.\n");
+ dev->features &= ~NETIF_F_USO;
+ }
+ }
/*
* nil rebuild_header routine,
diff -uNr linux-2.6.13/net/core/ethtool.c linux-2.6.13_uso/net/core/ethtool.c
--- linux-2.6.13/net/core/ethtool.c 2005-09-07 06:36:34.000000000 -0700
+++ linux-2.6.13_uso/net/core/ethtool.c 2005-09-07 06:32:15.000000000 -0700
@@ -81,6 +81,20 @@
return 0;
}
+u32 ethtool_op_get_uso(struct net_device *dev)
+{
+ return (dev->features & NETIF_F_USO) != 0;
+}
+
+int ethtool_op_set_uso(struct net_device *dev, u32 data)
+{
+ if (data)
+ dev->features |= NETIF_F_USO;
+ else
+ dev->features &= ~NETIF_F_USO;
+ return 0;
+}
+
/* Handlers for each ethtool command */
static int ethtool_get_settings(struct net_device *dev, void __user *useraddr)
@@ -469,6 +483,9 @@
err = dev->ethtool_ops->set_tso(dev, 0);
if (err)
return err;
+ err = dev->ethtool_ops->set_uso(dev, 0);
+ if (err)
+ return err;
}
return dev->ethtool_ops->set_sg(dev, data);
@@ -557,6 +574,32 @@
return dev->ethtool_ops->set_tso(dev, edata.data);
}
+static int ethtool_get_uso(struct net_device *dev, char __user *useraddr)
+{
+ struct ethtool_value edata = { ETHTOOL_GTSO };
+
+ if (!dev->ethtool_ops->get_uso)
+ return -EOPNOTSUPP;
+ edata.data = dev->ethtool_ops->get_uso(dev);
+ if (copy_to_user(useraddr, &edata, sizeof(edata)))
+ return -EFAULT;
+ return 0;
+}
+static int ethtool_set_uso(struct net_device *dev, char __user *useraddr)
+{
+ struct ethtool_value edata;
+
+ if (!dev->ethtool_ops->set_uso)
+ return -EOPNOTSUPP;
+ if (copy_from_user(&edata, useraddr, sizeof(edata)))
+ return -EFAULT;
+ if (edata.data && !(dev->features & NETIF_F_SG))
+ return -EINVAL;
+ if (edata.data && !(dev->features & NETIF_F_HW_CSUM))
+ return -EINVAL;
+ return dev->ethtool_ops->set_uso(dev, edata.data);
+}
+
static int ethtool_self_test(struct net_device *dev, char __user *useraddr)
{
struct ethtool_test test;
@@ -806,6 +849,12 @@
case ETHTOOL_GSTATS:
rc = ethtool_get_stats(dev, useraddr);
break;
+ case ETHTOOL_GUSO:
+ rc = ethtool_get_uso(dev, useraddr);
+ break;
+ case ETHTOOL_SUSO:
+ rc = ethtool_set_uso(dev, useraddr);
+ break;
default:
rc = -EOPNOTSUPP;
}
@@ -833,3 +882,5 @@
EXPORT_SYMBOL(ethtool_op_set_tso);
EXPORT_SYMBOL(ethtool_op_set_tx_csum);
EXPORT_SYMBOL(ethtool_op_set_tx_hw_csum);
+EXPORT_SYMBOL(ethtool_op_set_uso);
+EXPORT_SYMBOL(ethtool_op_get_uso);
diff -uNr linux-2.6.13/net/core/skbuff.c linux-2.6.13_uso/net/core/skbuff.c
--- linux-2.6.13/net/core/skbuff.c 2005-09-07 04:21:30.000000000 -0700
+++ linux-2.6.13_uso/net/core/skbuff.c 2005-09-07 06:38:57.000000000 -0700
@@ -159,6 +159,8 @@
skb_shinfo(skb)->tso_size = 0;
skb_shinfo(skb)->tso_segs = 0;
skb_shinfo(skb)->frag_list = NULL;
+ skb_shinfo(skb)->uso_size = 0;
+ skb_shinfo(skb)->ip6_frag_id = 0;
out:
return skb;
nodata:
@@ -1654,6 +1656,64 @@
return textsearch_find(config, state);
}
+/*
+ * skb_append_datato_frags - append the user data to a skb,
+ * sk - sock structure which contains skbs for transmission
+ * getfrag - The function to be called to get the data from the user.
+ * from - pointer to user message iov
+ * length - length of the iov message
+ *
+ * This procedure will allocate a skb enough to hold protocol headers and
+ * append the user data in the fragment part of the skb and add the skb to
+ * socket write queue
+ */
+int skb_append_datato_frags(struct sock *sk, struct sk_buff *skb,
+ int getfrag(void *from, char *to, int offset,
+ int len,int odd, struct sk_buff *skb),
+ void *from, int length)
+{
+ int frg_cnt = 0;
+ skb_frag_t *frag = NULL;
+ struct page *page = NULL;
+ int copy, left;
+ int offset = 0;
+ do {
+ frg_cnt = skb_shinfo(skb)->nr_frags;
+ if (frg_cnt >= MAX_SKB_FRAGS) {
+ kfree_skb(skb);
+ return -EFAULT;
+ }
+ page = alloc_pages(sk->sk_allocation, 0);
+ if (page == NULL) {
+ kfree_skb(skb);
+ return -ENOMEM;
+ }
+ sk->sk_sndmsg_page = page;
+ sk->sk_sndmsg_off = 0;
+ skb_fill_page_desc(skb, frg_cnt, page, 0, 0);
+ frg_cnt = skb_shinfo(skb)->nr_frags;
+ atomic_add(PAGE_SIZE, &sk->sk_wmem_alloc);
+ skb->truesize += PAGE_SIZE;
+ frag = &skb_shinfo(skb)->frags[frg_cnt - 1];
+ left = PAGE_SIZE - frag->page_offset;
+ copy = (length > left)? left : length;
+ if (getfrag(from, page_address(frag->page) +
+ frag->page_offset+frag->size,
+ offset, copy, 0, skb) < 0) {
+ kfree_skb(skb);
+ return -EFAULT;
+ }
+ sk->sk_sndmsg_off += copy;
+ frag->size += copy;
+ skb->len += copy;
+ skb->data_len += copy;
+ offset += copy;
+ length -= copy;
+ page = NULL;
+ } while (length > 0);
+ return 0;
+}
+
void __init skb_init(void)
{
skbuff_head_cache = kmem_cache_create("skbuff_head_cache",
@@ -1696,3 +1756,4 @@
EXPORT_SYMBOL(skb_seq_read);
EXPORT_SYMBOL(skb_abort_seq_read);
EXPORT_SYMBOL(skb_find_text);
+EXPORT_SYMBOL(skb_append_datato_frags);
diff -uNr linux-2.6.13/net/ipv4/ip_output.c
linux-2.6.13_uso/net/ipv4/ip_output.c
--- linux-2.6.13/net/ipv4/ip_output.c 2005-09-07 04:21:46.000000000 -0700
+++ linux-2.6.13_uso/net/ipv4/ip_output.c 2005-09-13 07:12:05.000000000
-0700
@@ -280,7 +280,8 @@
{
IP_INC_STATS(IPSTATS_MIB_OUTREQUESTS);
- if (skb->len > dst_mtu(skb->dst) && !skb_shinfo(skb)->tso_size)
+ if (skb->len > dst_mtu(skb->dst) &&
+ !(skb_shinfo(skb)->uso_size || skb_shinfo(skb)->tso_size))
return ip_fragment(skb, ip_finish_output);
else
return ip_finish_output(skb);
@@ -781,6 +782,46 @@
csummode = CHECKSUM_HW;
inet->cork.length += length;
+ if (((length > mtu) && (sk->sk_protocol == IPPROTO_UDP)) &&
+ (rt->u.dst.dev->features & NETIF_F_USO)) {
+ /* There is support for UDP large send offload by network
+ * device, so create one single skb packet containing complete
+ * udp datagram
+ */
+ if ((skb = skb_peek_tail(&sk->sk_write_queue)) == NULL) {
+ skb = sock_alloc_send_skb(sk,
+ hh_len + fragheaderlen + transhdrlen + 20,
+ (flags & MSG_DONTWAIT), &err);
+ if (skb == NULL)
+ goto error;
+ /* reserve space for Hardware header */
+ skb_reserve(skb, hh_len);
+ /* create space for UDP/IP header */
+ skb_put(skb,fragheaderlen + transhdrlen);
+ /* initialize network header pointer */
+ skb->nh.raw = skb->data;
+ /* initialize protocol header pointer */
+ skb->h.raw = skb->data + fragheaderlen;
+ skb->ip_summed = CHECKSUM_HW;
+ skb->csum = 0;
+ sk->sk_sndmsg_off = 0;
+ }
+ err = skb_append_datato_frags(sk,skb, getfrag, from,
+ (length - transhdrlen));
+ if (!err) {
+ /* specify the length of each IP datagram fragment*/
+ skb_shinfo(skb)->uso_size = (mtu - fragheaderlen);
+ __skb_queue_tail(&sk->sk_write_queue, skb);
+ return 0;
+ } else {
+ /* There is not enough support do UPD LSO,
+ * so follow normal path
+ */
+ kfree_skb(skb);
+ goto error;
+ }
+ }
+
/* So, what's going on in the loop below?
*
@@ -1012,14 +1053,23 @@
return -EINVAL;
inet->cork.length += size;
+ if ((sk->sk_protocol == IPPROTO_UDP) &&
+ (rt->u.dst.dev->features & NETIF_F_USO))
+ skb_shinfo(skb)->uso_size = (mtu - fragheaderlen);
+
while (size > 0) {
int i;
- /* Check if the remaining data fits into current packet. */
- len = mtu - skb->len;
- if (len < size)
- len = maxfraglen - skb->len;
+ if (skb_shinfo(skb)->uso_size) {
+ len = size;
+ } else {
+
+ /* Check if the remaining data fits into current
packet. */
+ len = mtu - skb->len;
+ if (len < size)
+ len = maxfraglen - skb->len;
+ }
if (len <= 0) {
struct sk_buff *skb_prev;
char *data;
diff -uNr linux-2.6.13/net/ipv6/ip6_output.c
linux-2.6.13_uso/net/ipv6/ip6_output.c
--- linux-2.6.13/net/ipv6/ip6_output.c 2005-09-07 04:21:57.000000000 -0700
+++ linux-2.6.13_uso/net/ipv6/ip6_output.c 2005-09-13 07:11:10.000000000
-0700
@@ -147,7 +147,8 @@
int ip6_output(struct sk_buff *skb)
{
- if (skb->len > dst_mtu(skb->dst) || dst_allfrag(skb->dst))
+ if ((skb->len > dst_mtu(skb->dst) && !skb_shinfo(skb)->uso_size) ||
+ dst_allfrag(skb->dst))
return ip6_fragment(skb, ip6_output2);
else
return ip6_output2(skb);
@@ -893,6 +894,50 @@
*/
inet->cork.length += length;
+ if (((length > mtu) && (sk->sk_protocol == IPPROTO_UDP)) &&
+ (rt->u.dst.dev->features & NETIF_F_USO)) {
+ /* There is support for UDP large send offload by network
+ * device, so create one single skb packet containing complete
+ * udp datagram
+ */
+ if ((skb = skb_peek_tail(&sk->sk_write_queue)) == NULL) {
+ skb = sock_alloc_send_skb(sk,
+ hh_len + fragheaderlen + transhdrlen + 20,
+ (flags & MSG_DONTWAIT), &err);
+ if (skb == NULL)
+ goto error;
+ /* reserve space for Hardware header */
+ skb_reserve(skb, hh_len);
+ /* create space for UDP/IP header */
+ skb_put(skb,fragheaderlen + transhdrlen);
+ /* initialize network header pointer */
+ skb->nh.raw = skb->data;
+ /* initialize protocol header pointer */
+ skb->h.raw = skb->data + fragheaderlen;
+ skb->ip_summed = CHECKSUM_HW;
+ skb->csum = 0;
+ sk->sk_sndmsg_off = 0;
+ }
+ err = skb_append_datato_frags(sk,skb, getfrag, from,
+ (length - transhdrlen));
+ if (!err) {
+ struct frag_hdr fhdr;
+
+ /* specify the length of each IP datagram fragment*/
+ skb_shinfo(skb)->uso_size = (mtu - fragheaderlen) -
+ sizeof(struct frag_hdr);
+ ipv6_select_ident(skb, &fhdr);
+ skb_shinfo(skb)->ip6_frag_id = fhdr.identification;
+ __skb_queue_tail(&sk->sk_write_queue, skb);
+ return 0;
+ } else {
+ /* There is not enough support do UPD LSO,
+ * so follow normal path
+ */
+ kfree_skb(skb);
+ goto error;
+ }
+ }
if ((skb = skb_peek_tail(&sk->sk_write_queue)) == NULL)
goto alloc_new_skb;
|