Skip to content

Commit

Permalink
[IPv4/IPv6]: UFO Scatter-gather approach
Browse files Browse the repository at this point in the history
Attached is kernel patch for UDP Fragmentation Offload (UFO) feature.

1. This patch incorporate the review comments by Jeff Garzik.
2. Renamed USO as UFO (UDP Fragmentation Offload)
3. udp sendfile support with UFO

This patches uses scatter-gather feature of skb to generate large UDP
datagram. Below is a "how-to" on changes required in network device
driver to use the UFO interface.

UDP Fragmentation Offload (UFO) Interface:
-------------------------------------------
UFO 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 UFO using
dev->features |= NETIF_F_UFO | NETIF_F_HW_CSUM | NETIF_F_SG

NETIF_F_HW_CSUM is required for UFO over ipv6.

2) UFO packet will be submitted for transmission using driver xmit routine.
UFO packet will have a non-zero value for

"skb_shinfo(skb)->ufo_size"

skb_shinfo(skb)->ufo_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 UFO 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@neterion.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (forwarded)
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
  • Loading branch information
Ananda Raju authored and Arnaldo Carvalho de Melo committed Oct 28, 2005
1 parent de51441 commit e89e9cf
Show file tree
Hide file tree
Showing 8 changed files with 306 additions and 6 deletions.
8 changes: 8 additions & 0 deletions include/linux/ethtool.h
Expand Up @@ -269,6 +269,8 @@ u32 ethtool_op_get_tso(struct net_device *dev);
int ethtool_op_set_tso(struct net_device *dev, u32 data);
int ethtool_op_get_perm_addr(struct net_device *dev,
struct ethtool_perm_addr *addr, u8 *data);
u32 ethtool_op_get_ufo(struct net_device *dev);
int ethtool_op_set_ufo(struct net_device *dev, u32 data);

/**
* &ethtool_ops - Alter and report network device settings
Expand Down Expand Up @@ -298,6 +300,8 @@ int ethtool_op_get_perm_addr(struct net_device *dev,
* 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_ufo: Report whether UDP fragmentation offload is enabled
* set_ufo: Turn UDP fragmentation 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
Expand Down Expand Up @@ -364,6 +368,8 @@ struct ethtool_ops {
int (*get_perm_addr)(struct net_device *, struct ethtool_perm_addr *, u8 *);
int (*begin)(struct net_device *);
void (*complete)(struct net_device *);
u32 (*get_ufo)(struct net_device *);
int (*set_ufo)(struct net_device *, u32);
};

/* CMDs currently supported */
Expand Down Expand Up @@ -400,6 +406,8 @@ struct ethtool_ops {
#define ETHTOOL_GTSO 0x0000001e /* Get TSO enable (ethtool_value) */
#define ETHTOOL_STSO 0x0000001f /* Set TSO enable (ethtool_value) */
#define ETHTOOL_GPERMADDR 0x00000020 /* Get permanent hardware address */
#define ETHTOOL_GUFO 0x00000021 /* Get UFO enable (ethtool_value) */
#define ETHTOOL_SUFO 0x00000022 /* Set UFO enable (ethtool_value) */

/* compatibility with older code */
#define SPARC_ETH_GSET ETHTOOL_GSET
Expand Down
1 change: 1 addition & 0 deletions include/linux/netdevice.h
Expand Up @@ -308,6 +308,7 @@ struct net_device
#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_UFO 8192 /* Can offload UDP Large Send*/

struct net_device *next_sched;

Expand Down
7 changes: 7 additions & 0 deletions include/linux/skbuff.h
Expand Up @@ -137,6 +137,8 @@ struct skb_shared_info {
unsigned int nr_frags;
unsigned short tso_size;
unsigned short tso_segs;
unsigned short ufo_size;
unsigned int ip6_frag_id;
struct sk_buff *frag_list;
skb_frag_t frags[MAX_SKB_FRAGS];
};
Expand Down Expand Up @@ -341,6 +343,11 @@ extern void skb_over_panic(struct sk_buff *skb, int len,
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;
Expand Down
14 changes: 14 additions & 0 deletions net/core/dev.c
Expand Up @@ -2717,6 +2717,20 @@ int register_netdevice(struct net_device *dev)
dev->name);
dev->features &= ~NETIF_F_TSO;
}
if (dev->features & NETIF_F_UFO) {
if (!(dev->features & NETIF_F_HW_CSUM)) {
printk(KERN_ERR "%s: Dropping NETIF_F_UFO since no "
"NETIF_F_HW_CSUM feature.\n",
dev->name);
dev->features &= ~NETIF_F_UFO;
}
if (!(dev->features & NETIF_F_SG)) {
printk(KERN_ERR "%s: Dropping NETIF_F_UFO since no "
"NETIF_F_SG feature.\n",
dev->name);
dev->features &= ~NETIF_F_UFO;
}
}

/*
* nil rebuild_header routine,
Expand Down
53 changes: 53 additions & 0 deletions net/core/ethtool.c
Expand Up @@ -93,6 +93,20 @@ int ethtool_op_get_perm_addr(struct net_device *dev, struct ethtool_perm_addr *a
}


u32 ethtool_op_get_ufo(struct net_device *dev)
{
return (dev->features & NETIF_F_UFO) != 0;
}

int ethtool_op_set_ufo(struct net_device *dev, u32 data)
{
if (data)
dev->features |= NETIF_F_UFO;
else
dev->features &= ~NETIF_F_UFO;
return 0;
}

/* Handlers for each ethtool command */

static int ethtool_get_settings(struct net_device *dev, void __user *useraddr)
Expand Down Expand Up @@ -483,6 +497,11 @@ static int __ethtool_set_sg(struct net_device *dev, u32 data)
return err;
}

if (!data && dev->ethtool_ops->set_ufo) {
err = dev->ethtool_ops->set_ufo(dev, 0);
if (err)
return err;
}
return dev->ethtool_ops->set_sg(dev, data);
}

Expand Down Expand Up @@ -569,6 +588,32 @@ static int ethtool_set_tso(struct net_device *dev, char __user *useraddr)
return dev->ethtool_ops->set_tso(dev, edata.data);
}

static int ethtool_get_ufo(struct net_device *dev, char __user *useraddr)
{
struct ethtool_value edata = { ETHTOOL_GTSO };

if (!dev->ethtool_ops->get_ufo)
return -EOPNOTSUPP;
edata.data = dev->ethtool_ops->get_ufo(dev);
if (copy_to_user(useraddr, &edata, sizeof(edata)))
return -EFAULT;
return 0;
}
static int ethtool_set_ufo(struct net_device *dev, char __user *useraddr)
{
struct ethtool_value edata;

if (!dev->ethtool_ops->set_ufo)
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_ufo(dev, edata.data);
}

static int ethtool_self_test(struct net_device *dev, char __user *useraddr)
{
struct ethtool_test test;
Expand Down Expand Up @@ -854,6 +899,12 @@ int dev_ethtool(struct ifreq *ifr)
case ETHTOOL_GPERMADDR:
rc = ethtool_get_perm_addr(dev, useraddr);
break;
case ETHTOOL_GUFO:
rc = ethtool_get_ufo(dev, useraddr);
break;
case ETHTOOL_SUFO:
rc = ethtool_set_ufo(dev, useraddr);
break;
default:
rc = -EOPNOTSUPP;
}
Expand Down Expand Up @@ -882,3 +933,5 @@ EXPORT_SYMBOL(ethtool_op_set_sg);
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_ufo);
EXPORT_SYMBOL(ethtool_op_get_ufo);
75 changes: 75 additions & 0 deletions net/core/skbuff.c
Expand Up @@ -176,6 +176,8 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
skb_shinfo(skb)->tso_size = 0;
skb_shinfo(skb)->tso_segs = 0;
skb_shinfo(skb)->frag_list = NULL;
skb_shinfo(skb)->ufo_size = 0;
skb_shinfo(skb)->ip6_frag_id = 0;
out:
return skb;
nodata:
Expand Down Expand Up @@ -1696,6 +1698,78 @@ unsigned int skb_find_text(struct sk_buff *skb, unsigned int from,
return textsearch_find(config, state);
}

/**
* skb_append_datato_frags: - append the user data to a skb
* @sk: sock structure
* @skb: skb structure to be appened with user data.
* @getfrag: call back function to be used for getting the user data
* @from: pointer to user message iov
* @length: length of the iov message
*
* Description: This procedure append the user data in the fragment part
* of the skb if any page alloc fails user this procedure returns -ENOMEM
*/
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;
int ret;

do {
/* Return error if we don't have space for new frag */
frg_cnt = skb_shinfo(skb)->nr_frags;
if (frg_cnt >= MAX_SKB_FRAGS)
return -EFAULT;

/* allocate a new page for next frag */
page = alloc_pages(sk->sk_allocation, 0);

/* If alloc_page fails just return failure and caller will
* free previous allocated pages by doing kfree_skb()
*/
if (page == NULL)
return -ENOMEM;

/* initialize the next frag */
sk->sk_sndmsg_page = page;
sk->sk_sndmsg_off = 0;
skb_fill_page_desc(skb, frg_cnt, page, 0, 0);
skb->truesize += PAGE_SIZE;
atomic_add(PAGE_SIZE, &sk->sk_wmem_alloc);

/* get the new initialized frag */
frg_cnt = skb_shinfo(skb)->nr_frags;
frag = &skb_shinfo(skb)->frags[frg_cnt - 1];

/* copy the user data to page */
left = PAGE_SIZE - frag->page_offset;
copy = (length > left)? left : length;

ret = getfrag(from, (page_address(frag->page) +
frag->page_offset + frag->size),
offset, copy, 0, skb);
if (ret < 0)
return -EFAULT;

/* copy was successful so update the size parameters */
sk->sk_sndmsg_off += copy;
frag->size += copy;
skb->len += copy;
skb->data_len += copy;
offset += copy;
length -= copy;

} while (length > 0);

return 0;
}

void __init skb_init(void)
{
skbuff_head_cache = kmem_cache_create("skbuff_head_cache",
Expand Down Expand Up @@ -1747,3 +1821,4 @@ EXPORT_SYMBOL(skb_prepare_seq_read);
EXPORT_SYMBOL(skb_seq_read);
EXPORT_SYMBOL(skb_abort_seq_read);
EXPORT_SYMBOL(skb_find_text);
EXPORT_SYMBOL(skb_append_datato_frags);
83 changes: 78 additions & 5 deletions net/ipv4/ip_output.c
Expand Up @@ -275,7 +275,8 @@ int ip_output(struct sk_buff *skb)
{
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)->ufo_size || skb_shinfo(skb)->tso_size))
return ip_fragment(skb, ip_finish_output);
else
return ip_finish_output(skb);
Expand Down Expand Up @@ -688,6 +689,60 @@ csum_page(struct page *page, int offset, int copy)
return csum;
}

inline int ip_ufo_append_data(struct sock *sk,
int getfrag(void *from, char *to, int offset, int len,
int odd, struct sk_buff *skb),
void *from, int length, int hh_len, int fragheaderlen,
int transhdrlen, int mtu,unsigned int flags)
{
struct sk_buff *skb;
int err;

/* There is support for UDP fragmentation 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)
return err;

/* 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)->ufo_size = (mtu - fragheaderlen);
__skb_queue_tail(&sk->sk_write_queue, skb);

return 0;
}
/* There is not enough support do UFO ,
* so follow normal path
*/
kfree_skb(skb);
return err;
}

/*
* ip_append_data() and ip_append_page() can make one large IP datagram
* from many pieces of data. Each pieces will be holded on the socket
Expand Down Expand Up @@ -777,6 +832,15 @@ int ip_append_data(struct sock *sk,
csummode = CHECKSUM_HW;

inet->cork.length += length;
if (((length > mtu) && (sk->sk_protocol == IPPROTO_UDP)) &&
(rt->u.dst.dev->features & NETIF_F_UFO)) {

if(ip_ufo_append_data(sk, getfrag, from, length, hh_len,
fragheaderlen, transhdrlen, mtu, flags))
goto error;

return 0;
}

/* So, what's going on in the loop below?
*
Expand Down Expand Up @@ -1008,14 +1072,23 @@ ssize_t ip_append_page(struct sock *sk, struct page *page,
return -EINVAL;

inet->cork.length += size;
if ((sk->sk_protocol == IPPROTO_UDP) &&
(rt->u.dst.dev->features & NETIF_F_UFO))
skb_shinfo(skb)->ufo_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)->ufo_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;
Expand Down

0 comments on commit e89e9cf

Please sign in to comment.