[PATCH 1/3] tun: Interface to query tun/tap features.

[PATCH 1/3] tun: Interface to query tun/tap features.

[Rusty Russell]

The problem with introducing checksum offload and gso to tun is they
need to set dev->features to enable GSO and/or checksumming, which is
supposed to be done before register_netdevice(), ie. as part of
TUNSETIFF.

Unfortunately, TUNSETIFF has always just ignored flags it doesn't
understand, so there's no good way of detecting whether the kernel
supports new IFF_ flags.

This patch implements a TUNGETFEATURES ioctl which returns all the valid IFF
flags.  It could be extended later to include other features.

Here's an example program which uses it:

#include <linux/if_tun.h>
#include <sys/types.h>
#include <sys/ioctl.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <err.h>
#include <stdio.h>

static struct {
	unsigned int flag;
	const char *name;
} known_flags[] = {
	{ IFF_TUN, "TUN" },
	{ IFF_TAP, "TAP" },
	{ IFF_NO_PI, "NO_PI" },
	{ IFF_ONE_QUEUE, "ONE_QUEUE" },
};

int main()
{
	unsigned int features, i;

	int netfd = open("/dev/net/tun", O_RDWR);
	if (netfd < 0)
		err(1, "Opening /dev/net/tun");

	if (ioctl(netfd, TUNGETFEATURES, &features) != 0) {
		printf("Kernel does not support TUNGETFEATURES, guessing\n");
		features = (IFF_TUN|IFF_TAP|IFF_NO_PI|IFF_ONE_QUEUE);
	}
	printf("Available features are: ");
	for (i = 0; i < sizeof(known_flags)/sizeof(known_flags[0]); i++) {
		if (features & known_flags[i].flag) {
			features &= ~known_flags[i].flag;
			printf("%s ", known_flags[i].name);
		}
	}
	if (features)
		printf("(UNKNOWN %#x)", features);
	printf("\n");
	return 0;
}

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/net/tun.c      |    8 ++++++++
 include/linux/if_tun.h |    1 +
 2 files changed, 9 insertions(+)

diff -r 8414a579e106 drivers/net/tun.c
--- a/drivers/net/tun.c	Tue Apr 22 07:36:45 2008 +1000
+++ b/drivers/net/tun.c	Tue Apr 22 07:37:33 2008 +1000
@@ -625,6 +625,14 @@ static int tun_chr_ioctl(struct inode *i
 		return 0;
 	}
 
+	if (cmd == TUNGETFEATURES) {
+		/* Currently this just means: "what IFF flags are valid?".
+		 * This is needed because we never checked for invalid flags on
+		 * TUNSETIFF. */
+		return put_user(IFF_TUN | IFF_TAP | IFF_NO_PI | IFF_ONE_QUEUE,
+				(unsigned int __user*)argp);
+	}
+
 	if (!tun)
 		return -EBADFD;
 
diff -r 8414a579e106 include/linux/if_tun.h
--- a/include/linux/if_tun.h	Tue Apr 22 07:36:45 2008 +1000
+++ b/include/linux/if_tun.h	Tue Apr 22 07:37:33 2008 +1000
@@ -42,6 +42,7 @@
 #define TUNSETOWNER   _IOW('T', 204, int)
 #define TUNSETLINK    _IOW('T', 205, int)
 #define TUNSETGROUP   _IOW('T', 206, int)
+#define TUNGETFEATURES _IOR('T', 207, unsigned int)
 
 /* TUNSETIFF ifr flags */
 #define IFF_TUN		0x0001

[PATCH 2/3] tun: TUNSETFEATURES to set gso features.

[Rusty Russell]

ethtool is useful for setting (some) device fields, but it's
root-only.  Finer feature control is available through a tun-specific
ioctl.

(Includes Mark McLoughlin <markmc@redhat.com>'s fix to hold rtnl sem).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/net/tun.c      |   49 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/if_tun.h |    7 +++++++
 2 files changed, 56 insertions(+)

diff -r e08e6c5130ff drivers/net/tun.c
--- a/drivers/net/tun.c	Thu Jun 26 00:17:07 2008 +1000
+++ b/drivers/net/tun.c	Thu Jun 26 00:21:11 2008 +1000
@@ -596,6 +596,46 @@ static int tun_set_iff(struct net *net, 
 	return err;
 }
 
+/* This is like a cut-down ethtool ops, except done via tun fd so no
+ * privs required. */
+static int set_offload(struct net_device *dev, unsigned long arg)
+{
+	unsigned int old_features, features;
+
+	old_features = dev->features;
+	/* Unset features, set them as we chew on the arg. */
+	features = (old_features & ~(NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST
+				    |NETIF_F_TSO_ECN|NETIF_F_TSO|NETIF_F_TSO6));
+
+	if (arg & TUN_F_CSUM) {
+		features |= NETIF_F_HW_CSUM|NETIF_F_SG|NETIF_F_FRAGLIST;
+		arg &= ~TUN_F_CSUM;
+
+		if (arg & (TUN_F_TSO4|TUN_F_TSO6)) {
+			if (arg & TUN_F_TSO_ECN) {
+				features |= NETIF_F_TSO_ECN;
+				arg &= ~TUN_F_TSO_ECN;
+			}
+			if (arg & TUN_F_TSO4)
+				features |= NETIF_F_TSO;
+			if (arg & TUN_F_TSO6)
+				features |= NETIF_F_TSO6;
+			arg &= ~(TUN_F_TSO4|TUN_F_TSO6);
+		}
+	}
+
+	/* This gives the user a way to test for new features in future by
+	 * trying to set them. */
+	if (arg)
+		return -EINVAL;
+
+	dev->features = features;
+	if (old_features != dev->features)
+		netdev_features_change(dev);
+
+	return 0;
+}
+
 static int tun_chr_ioctl(struct inode *inode, struct file *file,
 			 unsigned int cmd, unsigned long arg)
 {
@@ -699,6 +739,15 @@ static int tun_chr_ioctl(struct inode *i
 		tun->debug = arg;
 		break;
 #endif
+
+	case TUNSETOFFLOAD:
+	{
+		int ret;
+		rtnl_lock();
+		ret = set_offload(tun->dev, arg);
+		rtnl_unlock();
+		return ret;
+	}
 
 	case SIOCGIFFLAGS:
 		ifr.ifr_flags = tun->if_flags;
diff -r e08e6c5130ff include/linux/if_tun.h
--- a/include/linux/if_tun.h	Thu Jun 26 00:17:07 2008 +1000
+++ b/include/linux/if_tun.h	Thu Jun 26 00:21:11 2008 +1000
@@ -43,12 +43,19 @@
 #define TUNSETLINK    _IOW('T', 205, int)
 #define TUNSETGROUP   _IOW('T', 206, int)
 #define TUNGETFEATURES _IOR('T', 207, unsigned int)
+#define TUNSETOFFLOAD _IOW('T', 208, unsigned int)
 
 /* TUNSETIFF ifr flags */
 #define IFF_TUN		0x0001
 #define IFF_TAP		0x0002
 #define IFF_NO_PI	0x1000
 #define IFF_ONE_QUEUE	0x2000
+
+/* Features for GSO (TUNSETOFFLOAD). */
+#define TUN_F_CSUM	0x01	/* You can hand me unchecksummed packets. */
+#define TUN_F_TSO4	0x02	/* I can handle TSO for IPv4 packets */
+#define TUN_F_TSO6	0x04	/* I can handle TSO for IPv6 packets */
+#define TUN_F_TSO_ECN	0x08	/* I can handle TSO with ECN bits. */
 
 struct tun_pi {
 	unsigned short flags;

[PATCH 3/3] tun: Allow GSO using virtio_net_hdr

[Rusty Russell]

Add a IFF_VNET_HDR flag.  This uses the same ABI as virtio_net (ie. prepending
struct virtio_net_hdr to packets) to indicate GSO and checksum information.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/net/tun.c      |   95 +++++++++++++++++++++++++++++++++++++++++++++++--
 include/linux/if_tun.h |    2 +
 2 files changed, 94 insertions(+), 3 deletions(-)

diff -r 2c95ece55f0a drivers/net/tun.c
--- a/drivers/net/tun.c	Thu Jun 26 02:07:18 2008 +1000
+++ b/drivers/net/tun.c	Thu Jun 26 14:35:03 2008 +1000
@@ -63,6 +63,7 @@
 #include <linux/if_tun.h>
 #include <linux/crc32.h>
 #include <linux/nsproxy.h>
+#include <linux/virtio_net.h>
 #include <net/net_namespace.h>
 #include <net/netns/generic.h>
 
@@ -283,6 +284,7 @@ static __inline__ ssize_t tun_get_user(s
 	struct tun_pi pi = { 0, __constant_htons(ETH_P_IP) };
 	struct sk_buff *skb;
 	size_t len = count, align = 0;
+	struct virtio_net_hdr gso = { 0 };
 
 	if (!(tun->flags & TUN_NO_PI)) {
 		if ((len -= sizeof(pi)) > count)
@@ -290,6 +292,17 @@ static __inline__ ssize_t tun_get_user(s
 
 		if(memcpy_fromiovec((void *)&pi, iv, sizeof(pi)))
 			return -EFAULT;
+	}
+
+	if (tun->flags & TUN_VNET_HDR) {
+		if ((len -= sizeof(gso)) > count)
+			return -EINVAL;
+
+		if (memcpy_fromiovec((void *)&gso, iv, sizeof(gso)))
+			return -EFAULT;
+
+		if (gso.hdr_len > len)
+			return -EINVAL;
 	}
 
 	if ((tun->flags & TUN_TYPE_MASK) == TUN_TAP_DEV) {
@@ -311,6 +324,16 @@ static __inline__ ssize_t tun_get_user(s
 		return -EFAULT;
 	}
 
+	if (gso.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) {
+		if (!skb_partial_csum_set(skb, gso.csum_start,
+					  gso.csum_offset)) {
+			tun->dev->stats.rx_frame_errors++;
+			kfree_skb(skb);
+			return -EINVAL;
+		}
+	} else if (tun->flags & TUN_NOCHECKSUM)
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
 	switch (tun->flags & TUN_TYPE_MASK) {
 	case TUN_TUN_DEV:
 		skb_reset_mac_header(skb);
@@ -322,8 +345,35 @@ static __inline__ ssize_t tun_get_user(s
 		break;
 	};
 
-	if (tun->flags & TUN_NOCHECKSUM)
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	if (gso.gso_type != VIRTIO_NET_HDR_GSO_NONE) {
+		pr_debug("GSO!\n");
+		switch (gso.gso_type & ~VIRTIO_NET_HDR_GSO_ECN) {
+		case VIRTIO_NET_HDR_GSO_TCPV4:
+			skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
+			break;
+		case VIRTIO_NET_HDR_GSO_TCPV6:
+			skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6;
+			break;
+		default:
+			tun->dev->stats.rx_frame_errors++;
+			kfree_skb(skb);
+			return -EINVAL;
+		}
+
+		if (gso.gso_type & VIRTIO_NET_HDR_GSO_ECN)
+			skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN;
+
+		skb_shinfo(skb)->gso_size = gso.gso_size;
+		if (skb_shinfo(skb)->gso_size == 0) {
+			tun->dev->stats.rx_frame_errors++;
+			kfree_skb(skb);
+			return -EINVAL;
+		}
+
+		/* Header must be checked, and gso_segs computed. */
+		skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
+		skb_shinfo(skb)->gso_segs = 0;
+	}
 
 	netif_rx_ni(skb);
 	tun->dev->last_rx = jiffies;
@@ -367,6 +417,39 @@ static __inline__ ssize_t tun_put_user(s
 		if (memcpy_toiovec(iv, (void *) &pi, sizeof(pi)))
 			return -EFAULT;
 		total += sizeof(pi);
+	}
+
+	if (tun->flags & TUN_VNET_HDR) {
+		struct virtio_net_hdr gso = { 0 }; /* no info leak */
+		if ((len -= sizeof(gso)) < 0)
+			return -EINVAL;
+
+		if (skb_is_gso(skb)) {
+			struct skb_shared_info *sinfo = skb_shinfo(skb);
+
+			/* This is a hint as to how much should be linear. */
+			gso.hdr_len = skb_headlen(skb);
+			gso.gso_size = sinfo->gso_size;
+			if (sinfo->gso_type & SKB_GSO_TCPV4)
+				gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
+			else if (sinfo->gso_type & SKB_GSO_TCPV6)
+				gso.gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
+			else
+				BUG();
+			if (sinfo->gso_type & SKB_GSO_TCP_ECN)
+				gso.gso_type |= VIRTIO_NET_HDR_GSO_ECN;
+		} else
+			gso.gso_type = VIRTIO_NET_HDR_GSO_NONE;
+
+		if (skb->ip_summed == CHECKSUM_PARTIAL) {
+			gso.flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
+			gso.csum_start = skb->csum_start - skb_headroom(skb);
+			gso.csum_offset = skb->csum_offset;
+		} /* else everything is zero */
+
+		if (unlikely(memcpy_toiovec(iv, (void *)&gso, sizeof(gso))))
+			return -EFAULT;
+		total += sizeof(gso);
 	}
 
 	len = min_t(int, skb->len, len);
@@ -583,6 +666,11 @@ static int tun_set_iff(struct net *net, 
 	else
 		tun->flags &= ~TUN_ONE_QUEUE;
 
+	if (ifr->ifr_flags & IFF_VNET_HDR)
+		tun->flags |= TUN_VNET_HDR;
+	else
+		tun->flags &= ~TUN_VNET_HDR;
+
 	file->private_data = tun;
 	tun->attached = 1;
 	get_net(dev_net(tun->dev));
@@ -669,7 +757,8 @@ static int tun_chr_ioctl(struct inode *i
 		/* Currently this just means: "what IFF flags are valid?".
 		 * This is needed because we never checked for invalid flags on
 		 * TUNSETIFF. */
-		return put_user(IFF_TUN | IFF_TAP | IFF_NO_PI | IFF_ONE_QUEUE,
+		return put_user(IFF_TUN | IFF_TAP | IFF_NO_PI | IFF_ONE_QUEUE |
+				IFF_VNET_HDR,
 				(unsigned int __user*)argp);
 	}
 
diff -r 2c95ece55f0a include/linux/if_tun.h
--- a/include/linux/if_tun.h	Thu Jun 26 02:07:18 2008 +1000
+++ b/include/linux/if_tun.h	Thu Jun 26 14:35:03 2008 +1000
@@ -33,6 +33,7 @@
 #define TUN_NO_PI	0x0040
 #define TUN_ONE_QUEUE	0x0080
 #define TUN_PERSIST 	0x0100	
+#define TUN_VNET_HDR 	0x0200
 
 /* Ioctl defines */
 #define TUNSETNOCSUM  _IOW('T', 200, int) 
@@ -50,6 +51,7 @@
 #define IFF_TAP		0x0002
 #define IFF_NO_PI	0x1000
 #define IFF_ONE_QUEUE	0x2000
+#define IFF_VNET_HDR	0x4000
 
 /* Features for GSO (TUNSETOFFLOAD). */
 #define TUN_F_CSUM	0x01	/* You can hand me unchecksummed packets. */

Re: [PATCH 1/3] tun: Interface to query tun/tap features.

[David Miller]

From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 3 Jul 2008 11:32:12 +1000

> The problem with introducing checksum offload and gso to tun is they
> need to set dev->features to enable GSO and/or checksumming, which is
> supposed to be done before register_netdevice(), ie. as part of
> TUNSETIFF.
 ...
> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>

Applied to net-next-2.6, thanks!

Re: [PATCH 2/3] tun: TUNSETFEATURES to set gso features.

[David Miller]

From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 3 Jul 2008 11:33:11 +1000

> ethtool is useful for setting (some) device fields, but it's
> root-only.  Finer feature control is available through a tun-specific
> ioctl.
> 
> (Includes Mark McLoughlin <markmc@redhat.com>'s fix to hold rtnl sem).
> 
> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>

Applied to net-next-2.6

Re: [PATCH 3/3] tun: Allow GSO using virtio_net_hdr

[David Miller]

From: Rusty Russell <rusty@rustcorp.com.au>
Date: Thu, 3 Jul 2008 11:34:14 +1000

> Add a IFF_VNET_HDR flag.  This uses the same ABI as virtio_net (ie. prepending
> struct virtio_net_hdr to packets) to indicate GSO and checksum information.
> 
> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>

Also applied to net-next-2.6