[PATCH v16 01/17] Add a new structure for skb buffer from external.

[PATCH v16 01/17] Add a new structure for skb buffer from external.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 include/linux/skbuff.h |    9 +++++++++
 1 files changed, 9 insertions(+), 0 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 77eb60d..696e690 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -211,6 +211,15 @@ struct skb_shared_info {
 	skb_frag_t	frags[MAX_SKB_FRAGS];
 };
 
+/* The structure is for a skb which pages may point to
+ * an external buffer, which is not allocated from kernel space.
+ * It also contains a destructor for itself.
+ */
+struct skb_ext_page {
+	struct		page *page;
+	void		(*dtor)(struct skb_ext_page *);
+};
+
 /* We divide dataref into two halves.  The higher 16 bits hold references
  * to the payload part of skb->data.  The lower 16 bits hold references to
  * the entire skb->data.  A clone of a headerless skb holds the length of
-- 
1.7.3

[PATCH v16 02/17]Add a new struct for device to manipulate external buffer.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

    Add a structure in structure net_device, the new field is
    named as mp_port. It's for mediate passthru (zero-copy).
    It contains the capability for the net device driver,
    a socket, and an external buffer creator, external means
    skb buffer belongs to the device may not be allocated from
    kernel space.

    Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
    Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
    Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 include/linux/netdevice.h |   25 ++++++++++++++++++++++++-
 1 files changed, 24 insertions(+), 1 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 46c36ff..f6b1870 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -325,6 +325,28 @@ enum netdev_state_t {
 	__LINK_STATE_DORMANT,
 };
 
+/*The structure for mediate passthru(zero-copy). */
+struct mp_port	{
+	/* the header len */
+	int		hdr_len;
+	/* the max payload len for one descriptor */
+	int		data_len;
+	/* the pages for DMA in one time */
+	int		npages;
+	/* the socket bind to */
+	struct socket	*sock;
+	/* the header len for virtio-net */
+	int		vnet_hlen;
+	/* the external buffer page creator */
+	struct skb_ext_page *(*ctor)(struct mp_port *,
+				struct sk_buff *, int);
+	/* the hash function attached to find according
+	 * backend ring descriptor info for one external
+	 * buffer page.
+	 */
+	struct skb_ext_page *(*hash)(struct net_device *,
+				struct page *);
+};
 
 /*
  * This structure holds at boot time configured netdevice settings. They
@@ -1045,7 +1067,8 @@ struct net_device {
 
 	/* GARP */
 	struct garp_port	*garp_port;
-
+	/* mpassthru */
+	struct mp_port		*mp_port;
 	/* class/net/name entry */
 	struct device		dev;
 	/* space for optional device, statistics, and wireless sysfs groups */
-- 
1.7.3

Re: [PATCH v16 02/17]Add a new struct for device to manipulate external buffer.

[David Miller]

From: xiaohui.xin@intel.com
Date: Wed,  1 Dec 2010 16:08:13 +0800

> From: Xin Xiaohui <xiaohui.xin@intel.com>
> 
>     Add a structure in structure net_device, the new field is
>     named as mp_port. It's for mediate passthru (zero-copy).
>     It contains the capability for the net device driver,
>     a socket, and an external buffer creator, external means
>     skb buffer belongs to the device may not be allocated from
>     kernel space.
> 
>     Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
>     Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
>     Reviewed-by: Jeff Dike <jdike@linux.intel.com>

Please eliminate whatever is causing this indentation of your
commit messages.

There should be no special indentation of the commit message.

[PATCH v16 03/17] Add a ndo_mp_port_prep pointer to net_device_ops.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

    If the driver want to allocate external buffers,
    then it can export it's capability, as the skb
    buffer header length, the page length can be DMA, etc.
    The external buffers owner may utilize this.

    Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
    Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
    Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 include/linux/netdevice.h |   10 ++++++++++
 1 files changed, 10 insertions(+), 0 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f6b1870..575777f 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -723,6 +723,12 @@ struct netdev_rx_queue {
  * int (*ndo_set_vf_port)(struct net_device *dev, int vf,
  *			  struct nlattr *port[]);
  * int (*ndo_get_vf_port)(struct net_device *dev, int vf, struct sk_buff *skb);
+ *
+ * int (*ndo_mp_port_prep)(struct net_device *dev, struct mp_port *port);
+ *	If the driver want to allocate external buffers,
+ *	then it can export it's capability, as the skb
+ *	buffer header length, the page length can be DMA, etc.
+ *	The external buffers owner may utilize this.
  */
 #define HAVE_NET_DEVICE_OPS
 struct net_device_ops {
@@ -795,6 +801,10 @@ struct net_device_ops {
 	int			(*ndo_fcoe_get_wwn)(struct net_device *dev,
 						    u64 *wwn, int type);
 #endif
+#if defined(CONFIG_MEDIATE_PASSTHRU) || defined(CONFIG_MEDIATE_PASSTHRU_MODULE)
+	int			(*ndo_mp_port_prep)(struct net_device *dev,
+						struct mp_port *port);
+#endif
 };
 
 /*
-- 
1.7.3

Re: [PATCH v16 03/17] Add a ndo_mp_port_prep pointer to net_device_ops.

[David Miller]

From: xiaohui.xin@intel.com
Date: Wed,  1 Dec 2010 16:08:14 +0800

> +#if defined(CONFIG_MEDIATE_PASSTHRU) || defined(CONFIG_MEDIATE_PASSTHRU_MODULE)
> +	int			(*ndo_mp_port_prep)(struct net_device *dev,
> +						struct mp_port *port);
> +#endif

Please rename this config option so that it is clear, by name, that
this option is for a networking facility.

F.e. CONFIG_NET_MEDIATE_PASSTHRU

[PATCH v16 04/17]Add a function make external buffer owner to query capability.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

The external buffer owner can use the functions to get
the capability of the underlying NIC driver.

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
Signed-off-by: Zhao Yu <yzhaonew@gmail.com>
Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 include/linux/netdevice.h |    2 ++
 net/core/dev.c            |   41 +++++++++++++++++++++++++++++++++++++++++
 2 files changed, 43 insertions(+), 0 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 575777f..8dcf6de 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1736,6 +1736,8 @@ extern gro_result_t	napi_frags_finish(struct napi_struct *napi,
 					  gro_result_t ret);
 extern struct sk_buff *	napi_frags_skb(struct napi_struct *napi);
 extern gro_result_t	napi_gro_frags(struct napi_struct *napi);
+extern int netdev_mp_port_prep(struct net_device *dev,
+				struct mp_port *port);
 
 static inline void napi_free_frags(struct napi_struct *napi)
 {
diff --git a/net/core/dev.c b/net/core/dev.c
index 660dd41..84fbb83 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2942,6 +2942,47 @@ out:
 	return ret;
 }
 
+/* To support meidate passthru(zero-copy) with NIC driver,
+ * we'd better query NIC driver for the capability it can
+ * provide, especially for packet split mode, now we only
+ * query for the header size, and the payload a descriptor
+ * may carry.
+ * Now, it's only called by mpassthru device.
+ */
+#if defined(CONFIG_MEDIATE_PASSTHRU) || defined(CONFIG_MEDIATE_PASSTHRU_MODULE)
+int netdev_mp_port_prep(struct net_device *dev,
+		struct mp_port *port)
+{
+	int rc;
+	int npages, data_len;
+	const struct net_device_ops *ops = dev->netdev_ops;
+
+	if (ops->ndo_mp_port_prep) {
+		rc = ops->ndo_mp_port_prep(dev, port);
+		if (rc)
+			return rc;
+	} else
+		return -EINVAL;
+
+	if (port->hdr_len <= 0)
+		goto err;
+
+	npages = port->npages;
+	data_len = port->data_len;
+	if (npages <= 0 || npages > MAX_SKB_FRAGS ||
+			(data_len < PAGE_SIZE * (npages - 1) ||
+			 data_len > PAGE_SIZE * npages))
+		goto err;
+
+	return 0;
+err:
+	dev_warn(&dev->dev, "invalid page constructor parameters\n");
+
+	return -EINVAL;
+}
+EXPORT_SYMBOL(netdev_mp_port_prep);
+#endif
+
 /**
  *	netif_receive_skb - process receive buffer from network
  *	@skb: buffer to process
-- 
1.7.3

[PATCH v16 05/17] Add a function to indicate if device use external buffer.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 include/linux/netdevice.h |    5 +++++
 1 files changed, 5 insertions(+), 0 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 8dcf6de..f91d9bb 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1739,6 +1739,11 @@ extern gro_result_t	napi_gro_frags(struct napi_struct *napi);
 extern int netdev_mp_port_prep(struct net_device *dev,
 				struct mp_port *port);
 
+static inline bool dev_is_mpassthru(struct net_device *dev)
+{
+	return dev && dev->mp_port;
+}
+
 static inline void napi_free_frags(struct napi_struct *napi)
 {
 	kfree_skb(napi->skb);
-- 
1.7.3

[PATCH v16 06/17]move member destructor_arg before member dataref

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

Then we can clear destructor_arg when __alloc_skb().

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 include/linux/skbuff.h |    7 ++++---
 1 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 696e690..6e1e991 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -199,14 +199,15 @@ struct skb_shared_info {
 	struct sk_buff	*frag_list;
 	struct skb_shared_hwtstamps hwtstamps;
 
+	/* Intermediate layers must ensure that destructor_arg
+	 * remains valid until skb destructor */
+	void *		destructor_arg;
+
 	/*
 	 * Warning : all fields before dataref are cleared in __alloc_skb()
 	 */
 	atomic_t	dataref;
 
-	/* Intermediate layers must ensure that destructor_arg
-	 * remains valid until skb destructor */
-	void *		destructor_arg;
 	/* must be last field, see pskb_expand_head() */
 	skb_frag_t	frags[MAX_SKB_FRAGS];
 };
-- 
1.7.3

[PATCH v16 07/17]Modify netdev_alloc_page() to get external buffer

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

    Currently, it can get external buffers from mp device.

    Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
    Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
    Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 net/core/skbuff.c |   27 +++++++++++++++++++++++++++
 1 files changed, 27 insertions(+), 0 deletions(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index c83b421..b9858c7 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -261,11 +261,38 @@ struct sk_buff *__netdev_alloc_skb(struct net_device *dev,
 }
 EXPORT_SYMBOL(__netdev_alloc_skb);
 
+struct page *netdev_alloc_ext_pages(struct net_device *dev, int npages)
+{
+	struct mp_port *port;
+	struct skb_ext_page *ext_page = NULL;
+
+	port = dev->mp_port;
+	if (!port)
+		goto out;
+	ext_page = port->ctor(port, NULL, npages);
+	if (ext_page)
+		return ext_page->page;
+out:
+	return NULL;
+
+}
+EXPORT_SYMBOL(netdev_alloc_ext_pages);
+
+struct page *netdev_alloc_ext_page(struct net_device *dev)
+{
+	return netdev_alloc_ext_pages(dev, 1);
+
+}
+EXPORT_SYMBOL(netdev_alloc_ext_page);
+
 struct page *__netdev_alloc_page(struct net_device *dev, gfp_t gfp_mask)
 {
 	int node = dev->dev.parent ? dev_to_node(dev->dev.parent) : -1;
 	struct page *page;
 
+	if (dev_is_mpassthru(dev))
+		return netdev_alloc_ext_page(dev);
+
 	page = alloc_pages_node(node, gfp_mask, 0);
 	return page;
 }
-- 
1.7.3

[PATCH v16 08/17]Modify netdev_free_page() to release external buffer

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

Currently, it can get external buffers from mp device.

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 include/linux/skbuff.h |    4 +++-
 net/core/skbuff.c      |   24 ++++++++++++++++++++++++
 2 files changed, 27 insertions(+), 1 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 6e1e991..6309ce6 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1586,9 +1586,11 @@ static inline struct page *netdev_alloc_page(struct net_device *dev)
 	return __netdev_alloc_page(dev, GFP_ATOMIC);
 }
 
+extern void __netdev_free_page(struct net_device *dev, struct page *page);
+
 static inline void netdev_free_page(struct net_device *dev, struct page *page)
 {
-	__free_page(page);
+	__netdev_free_page(dev, page);
 }
 
 /**
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index b9858c7..d3ece5c 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -298,6 +298,30 @@ struct page *__netdev_alloc_page(struct net_device *dev, gfp_t gfp_mask)
 }
 EXPORT_SYMBOL(__netdev_alloc_page);
 
+void netdev_free_ext_page(struct net_device *dev, struct page *page)
+{
+	struct skb_ext_page *ext_page = NULL;
+	if (dev_is_mpassthru(dev) && dev->mp_port->hash) {
+		ext_page = dev->mp_port->hash(dev, page);
+		if (ext_page)
+			ext_page->dtor(ext_page);
+		else
+			__free_page(page);
+	}
+}
+EXPORT_SYMBOL(netdev_free_ext_page);
+
+void __netdev_free_page(struct net_device *dev, struct page *page)
+{
+	if (dev_is_mpassthru(dev)) {
+		netdev_free_ext_page(dev, page);
+		return;
+	}
+
+	__free_page(page);
+}
+EXPORT_SYMBOL(__netdev_free_page);
+
 void skb_add_rx_frag(struct sk_buff *skb, int i, struct page *page, int off,
 		int size)
 {
-- 
1.7.3

[PATCH v16 09/17] Don't do skb recycle, if device use external buffer.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 net/core/skbuff.c |    6 ++++++
 1 files changed, 6 insertions(+), 0 deletions(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index d3ece5c..11833b4 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -550,6 +550,12 @@ bool skb_recycle_check(struct sk_buff *skb, int skb_size)
 	if (skb_shared(skb) || skb_cloned(skb))
 		return false;
 
+	/* if the device wants to do mediate passthru, the skb may
+	 * get external buffer, so don't recycle
+	 */
+	if (dev_is_mpassthru(skb->dev))
+		return 0;
+
 	skb_release_head_state(skb);
 
 	shinfo = skb_shinfo(skb);
-- 
1.7.3

[PATCH v16 10/17] If device is in zero-copy mode first, bonding will fail.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

If device is in this zero-copy mode first, we cannot handle this,
so fail it. This patch is for this.

If bonding is created first, and one of the device will be in zero-copy
mode, this will be handled by mp device. It will first check if all the
slaves have the zero-copy capability. If no, fail too. Otherwise make
all the slaves in zero-copy mode.

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
---
 drivers/net/bonding/bond_main.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 3b16f62..dfb6a2c 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1428,6 +1428,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 			   bond_dev->name);
 	}
 
+	/* if the device is in zero-copy mode before bonding, fail it. */
+	if (dev_is_mpassthru(slave_dev))
+		return -EBUSY;
+
 	/* already enslaved */
 	if (slave_dev->flags & IFF_SLAVE) {
 		pr_debug("Error, Device was already enslaved\n");
-- 
1.7.3

[PATCH v16 11/17]Add a hook to intercept external buffers from NIC driver.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

The hook is called in __netif_receive_skb().

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 net/core/dev.c |   40 ++++++++++++++++++++++++++++++++++++++++
 1 files changed, 40 insertions(+), 0 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 84fbb83..bdad1c8 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2814,6 +2814,40 @@ int __skb_bond_should_drop(struct sk_buff *skb, struct net_device *master)
 }
 EXPORT_SYMBOL(__skb_bond_should_drop);
 
+#if defined(CONFIG_MEDIATE_PASSTHRU) || defined(CONFIG_MEDIATE_PASSTHRU_MODULE)
+/* Add a hook to intercept mediate passthru(zero-copy) packets,
+ * and insert it to the socket queue owned by mp_port specially.
+ */
+static inline struct sk_buff *handle_mpassthru(struct sk_buff *skb,
+					       struct packet_type **pt_prev,
+					       int *ret,
+					       struct net_device *orig_dev)
+{
+	struct mp_port *mp_port = NULL;
+	struct sock *sk = NULL;
+
+	if (!dev_is_mpassthru(skb->dev) && !dev_is_mpassthru(orig_dev))
+		return skb;
+	if (dev_is_mpassthru(skb->dev))
+		mp_port = skb->dev->mp_port;
+	else if (orig_dev->master == skb->dev && dev_is_mpassthru(orig_dev))
+		mp_port = orig_dev->mp_port;
+
+	if (*pt_prev) {
+		*ret = deliver_skb(skb, *pt_prev, orig_dev);
+		*pt_prev = NULL;
+	}
+
+	sk = mp_port->sock->sk;
+	skb_queue_tail(&sk->sk_receive_queue, skb);
+	sk->sk_state_change(sk);
+
+	return NULL;
+}
+#else
+#define handle_mpassthru(skb, pt_prev, ret, orig_dev)     (skb)
+#endif
+
 static int __netif_receive_skb(struct sk_buff *skb)
 {
 	struct packet_type *ptype, *pt_prev;
@@ -2891,6 +2925,11 @@ static int __netif_receive_skb(struct sk_buff *skb)
 ncls:
 #endif
 
+	/* To intercept mediate passthru(zero-copy) packets here */
+	skb = handle_mpassthru(skb, &pt_prev, &ret, orig_dev);
+	if (!skb)
+		goto out;
+
 	/* Handle special case of bridge or macvlan */
 	rx_handler = rcu_dereference(skb->dev->rx_handler);
 	if (rx_handler) {
@@ -2983,6 +3022,7 @@ err:
 EXPORT_SYMBOL(netdev_mp_port_prep);
 #endif
 
+
 /**
  *	netif_receive_skb - process receive buffer from network
  *	@skb: buffer to process
-- 
1.7.3

Re: [PATCH v16 11/17]Add a hook to intercept external buffers from NIC driver.

[Changli Gao]

On Wed, Dec 1, 2010 at 4:08 PM,  <xiaohui.xin@intel.com> wrote:
> From: Xin Xiaohui <xiaohui.xin@intel.com>
> @@ -2891,6 +2925,11 @@ static int __netif_receive_skb(struct sk_buff *skb)
>  ncls:
>  #endif
>
> +       /* To intercept mediate passthru(zero-copy) packets here */
> +       skb = handle_mpassthru(skb, &pt_prev, &ret, orig_dev);
> +       if (!skb)
> +               goto out;
> +
>        /* Handle special case of bridge or macvlan */

I think it won't work if the skbs is captured by the previous ptype.
We need to trace the skb pages(skb_shared_info.frags[*].page), but
currently, there isn't a easy way to do that. skb pages are treated as
normal pages, and anyone can get it freely. And it is the problem in
the way to fix the potential data corruption bug.

Eric thinks af_packets(mmap) isn't worth fixing. But if this patch
serial has the same problem. It will be worth.

My idea is adding a new function dtor pointer  and some other args
pointers to the struct skb_shared_info. If skb_shared_info.dtor
exists, the pages in skb_shared_info.frags are private to this
skb->head. Anyone who wants to get these pages, should copy them
instead.

-- 
Regards,
Changli Gao(xiaosuo@gmail.com)

[PATCH v16 12/17] Add header file for mp device.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 include/linux/mpassthru.h |  133 +++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 133 insertions(+), 0 deletions(-)
 create mode 100644 include/linux/mpassthru.h

diff --git a/include/linux/mpassthru.h b/include/linux/mpassthru.h
new file mode 100644
index 0000000..1115f55
--- /dev/null
+++ b/include/linux/mpassthru.h
@@ -0,0 +1,133 @@
+#ifndef __MPASSTHRU_H
+#define __MPASSTHRU_H
+
+#include <linux/types.h>
+#include <linux/if_ether.h>
+#include <linux/ioctl.h>
+
+/* ioctl defines */
+#define MPASSTHRU_BINDDEV      _IOW('M', 213, int)
+#define MPASSTHRU_UNBINDDEV    _IO('M', 214)
+#define MPASSTHRU_SET_MEM_LOCKED       _IOW('M', 215, unsigned long)
+#define MPASSTHRU_GET_MEM_LOCKED_NEED  _IOR('M', 216, unsigned long)
+
+#define COPY_THRESHOLD (L1_CACHE_BYTES * 4)
+#define COPY_HDR_LEN   (L1_CACHE_BYTES < 64 ? 64 : L1_CACHE_BYTES)
+
+#define DEFAULT_NEED   ((8192*2*2)*4096)
+
+struct frag {
+	u16     offset;
+	u16     size;
+};
+
+#define HASH_BUCKETS    (8192*2)
+struct page_info {
+	struct list_head        list;
+	struct page_info        *next;
+	struct page_info        *prev;
+	struct page             *pages[MAX_SKB_FRAGS];
+	struct sk_buff          *skb;
+	struct page_pool        *pool;
+
+	/* The pointer relayed to skb, to indicate
+	 * it's a external allocated skb or kernel
+	 */
+	struct skb_ext_page    ext_page;
+	/* flag to indicate read or write */
+#define INFO_READ                      0
+#define INFO_WRITE                     1
+	unsigned                flags;
+	/* exact number of locked pages */
+	unsigned                pnum;
+
+	/* The fields after that is for backend
+	 * driver, now for vhost-net.
+	 */
+	/* the kiocb structure related to */
+	struct kiocb            *iocb;
+	/* the ring descriptor index */
+	unsigned int            desc_pos;
+	/* the iovec coming from backend, we only
+	 * need few of them */
+	struct iovec            hdr[2];
+	struct iovec            iov[2];
+};
+
+struct page_pool {
+	/* the queue for rx side */
+	struct list_head        readq;
+	/* the lock to protect readq */
+	spinlock_t              read_lock;
+	/* record the orignal rlimit */
+	struct rlimit           o_rlim;
+	/* userspace wants to locked */
+	int                     locked_pages;
+	/* currently locked pages */
+	int                     cur_pages;
+	/* the memory locked before */
+	unsigned long		orig_locked_vm;
+	/* the device according to */
+	struct net_device       *dev;
+	/* the mp_port according to dev */
+	struct mp_port          port;
+	/* the hash_table list to find each locked page */
+	struct page_info        **hash_table;
+};
+
+static struct kmem_cache *ext_page_info_cache;
+
+#ifdef __KERNEL__
+#if defined(CONFIG_MEDIATE_PASSTHRU) || defined(CONFIG_MEDIATE_PASSTHRU_MODULE)
+struct socket *mp_get_socket(struct file *);
+struct page_pool *page_pool_create(struct net_device *dev,
+				   struct socket *sock);
+int async_recvmsg(struct kiocb *iocb, struct page_pool *pool,
+		  struct iovec *iov, int count, int flags);
+int async_sendmsg(struct sock *sk, struct kiocb *iocb,
+		  struct page_pool *pool, struct iovec *iov,
+		  int count);
+void async_data_ready(struct sock *sk, struct page_pool *pool);
+void dev_change_state(struct net_device *dev);
+void page_pool_destroy(struct mm_struct *mm, struct page_pool *pool);
+#else
+#include <linux/err.h>
+#include <linux/errno.h>
+struct file;
+struct socket;
+static inline struct socket *mp_get_socket(struct file *f)
+{
+	return ERR_PTR(-EINVAL);
+}
+static inline struct page_pool *page_pool_create(struct net_device *dev,
+		struct socket *sock)
+{
+	return ERR_PTR(-EINVAL);
+}
+static inline int async_recvmsg(struct kiocb *iocb, struct page_pool *pool,
+		struct iovec *iov, int count, int flags)
+{
+	return -EINVAL;
+}
+static inline int async_sendmsg(struct sock *sk, struct kiocb *iocb,
+		struct page_pool *pool, struct iovec *iov,
+		int count)
+{
+	return -EINVAL;
+}
+static inline void async_data_ready(struct sock *sk, struct page_pool *pool)
+{
+	return;
+}
+static inline void dev_change_state(struct net_device *dev)
+{
+	return;
+}
+static inline void page_pool_destroy(struct mm_struct *mm,
+				     struct page_pool *pool)
+{
+	return;
+}
+#endif /* CONFIG_MEDIATE_PASSTHRU */
+#endif /* __KERNEL__ */
+#endif /* __MPASSTHRU_H */
-- 
1.7.3

[PATCH v16 13/17] Add mp(mediate passthru) device.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

The patch add mp(mediate passthru) device, which now
based on vhost-net backend driver and provides proto_ops
to send/receive guest buffers data from/to guest vitio-net
driver.
It also exports async functions which can be used by other
drivers like macvtap to utilize zero-copy too.

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
Signed-off-by: Zhao Yu <yzhao81new@gmail.com>
Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 drivers/vhost/mpassthru.c | 1495 +++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 1495 insertions(+), 0 deletions(-)
 create mode 100644 drivers/vhost/mpassthru.c

diff --git a/drivers/vhost/mpassthru.c b/drivers/vhost/mpassthru.c
new file mode 100644
index 0000000..868200a
--- /dev/null
+++ b/drivers/vhost/mpassthru.c
@@ -0,0 +1,1495 @@
+/*
+ *  MPASSTHRU - Mediate passthrough device.
+ *  Copyright (C) 2009 ZhaoYu, XinXiaohui, Dike, Jeffery G
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ *  GNU General Public License for more details.
+ *
+ */
+
+#define DRV_NAME        "mpassthru"
+#define DRV_DESCRIPTION "Mediate passthru device driver"
+#define DRV_COPYRIGHT   "(C) 2009 ZhaoYu, XinXiaohui, Dike, Jeffery G"
+
+#include <linux/compat.h>
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/major.h>
+#include <linux/slab.h>
+#include <linux/smp_lock.h>
+#include <linux/poll.h>
+#include <linux/fcntl.h>
+#include <linux/init.h>
+#include <linux/aio.h>
+
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/miscdevice.h>
+#include <linux/ethtool.h>
+#include <linux/rtnetlink.h>
+#include <linux/if.h>
+#include <linux/if_arp.h>
+#include <linux/if_ether.h>
+#include <linux/crc32.h>
+#include <linux/nsproxy.h>
+#include <linux/uaccess.h>
+#include <linux/virtio_net.h>
+#include <linux/mpassthru.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+#include <net/rtnetlink.h>
+#include <net/sock.h>
+
+#include <asm/system.h>
+#include "../net/bonding/bonding.h"
+
+struct mp_struct {
+	struct mp_file		*mfile;
+	struct net_device       *dev;
+	struct page_pool	*pool;
+	struct socket           socket;
+	struct socket_wq	wq;
+	struct mm_struct	*mm;
+};
+
+struct mp_file {
+	atomic_t count;
+	struct mp_struct *mp;
+	struct net *net;
+};
+
+struct mp_sock {
+	struct sock		sk;
+	struct mp_struct	*mp;
+};
+
+/* The main function to allocate external buffers */
+static struct skb_ext_page *page_ctor(struct mp_port *port,
+				      struct sk_buff *skb,
+				      int npages)
+{
+	int i;
+	unsigned long flags;
+	struct page_pool *pool;
+	struct page_info *info = NULL;
+
+	if (npages != 1)
+		BUG();
+	pool = container_of(port, struct page_pool, port);
+
+	spin_lock_irqsave(&pool->read_lock, flags);
+	if (!list_empty(&pool->readq)) {
+		info = list_first_entry(&pool->readq, struct page_info, list);
+		list_del(&info->list);
+	}
+	spin_unlock_irqrestore(&pool->read_lock, flags);
+	if (!info)
+		return NULL;
+
+	for (i = 0; i < info->pnum; i++)
+		get_page(info->pages[i]);
+	info->skb = skb;
+	return &info->ext_page;
+}
+
+static struct page_info *mp_hash_lookup(struct page_pool *pool,
+					struct page *page);
+static struct page_info *mp_hash_delete(struct page_pool *pool,
+					struct page_info *info);
+
+static struct skb_ext_page *mp_lookup(struct net_device *dev,
+				      struct page *page)
+{
+	struct mp_struct *mp =
+		container_of(dev->mp_port->sock->sk, struct mp_sock, sk)->mp;
+	struct page_pool *pool = mp->pool;
+	struct page_info *info;
+
+	info = mp_hash_lookup(pool, page);
+	if (!info)
+		return NULL;
+	return &info->ext_page;
+}
+
+struct page_pool *page_pool_create(struct net_device *dev,
+				  struct socket *sock)
+{
+	struct page_pool *pool;
+	struct net_device *master;
+	struct slave *slave;
+	struct bonding *bond;
+	int i;
+	int rc;
+
+	pool = kzalloc(sizeof(*pool), GFP_KERNEL);
+	if (!pool)
+		return NULL;
+
+	/* How to deal with bonding device:
+	 * check if all the slaves are capable of zero-copy.
+	 * if not, fail.
+	 */
+	master = dev->master;
+	if (master) {
+		bond = netdev_priv(master);
+		read_lock(&bond->lock);
+		bond_for_each_slave(bond, slave, i) {
+			rc = netdev_mp_port_prep(slave->dev, &pool->port);
+			if (rc)
+				break;
+		}
+		read_unlock(&bond->lock);
+	} else
+		rc = netdev_mp_port_prep(dev, &pool->port);
+	if (rc)
+		goto fail;
+
+	INIT_LIST_HEAD(&pool->readq);
+	spin_lock_init(&pool->read_lock);
+	pool->hash_table =
+		kzalloc(sizeof(struct page_info *) * HASH_BUCKETS, GFP_KERNEL);
+	if (!pool->hash_table)
+		goto fail;
+
+	pool->dev = dev;
+	pool->port.ctor = page_ctor;
+	pool->port.sock = sock;
+	pool->port.hash = mp_lookup;
+	pool->locked_pages = 0;
+	pool->cur_pages = 0;
+	pool->orig_locked_vm = 0;
+
+	/* for bonding device, assign all the slaves the same page_pool */
+	if (master) {
+		read_lock(&bond->lock);
+		bond_for_each_slave(bond, slave, i) {
+			dev_hold(slave->dev);
+			slave->dev->mp_port = &pool->port;
+		}
+		read_unlock(&bond->lock);
+	} else {
+		dev_hold(dev);
+		dev->mp_port = &pool->port;
+	}
+
+	return pool;
+fail:
+	kfree(pool);
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(page_pool_create);
+
+void dev_bond_hold(struct net_device *dev)
+{
+	struct net_device *master;
+	struct bonding *bond;
+	struct slave *slave;
+	int i;
+
+	master = dev->master;
+	if (master) {
+		bond = netdev_priv(master);
+		read_lock(&bond->lock);
+		bond_for_each_slave(bond, slave, i) {
+			if (slave->dev != dev)
+				dev_hold(slave->dev);
+		}
+		read_unlock(&bond->lock);
+	}
+}
+
+void dev_bond_put(struct net_device *dev)
+{
+	struct net_device *master;
+	struct bonding *bond;
+	struct slave *slave;
+	int i;
+
+	master = dev->master;
+	if (master) {
+		bond = netdev_priv(master);
+		read_lock(&bond->lock);
+		bond_for_each_slave(bond, slave, i) {
+			if (slave->dev != dev)
+				dev_put(slave->dev);
+		}
+		read_unlock(&bond->lock);
+	}
+}
+
+void dev_change_state(struct net_device *dev)
+{
+	struct net_device *master;
+	struct bonding *bond;
+	struct slave *slave;
+	int i;
+
+	master = dev->master;
+	if (master) {
+		bond = netdev_priv(master);
+		read_lock(&bond->lock);
+		bond_for_each_slave(bond, slave, i) {
+			dev_change_flags(slave->dev,
+					 slave->dev->flags & (~IFF_UP));
+			dev_change_flags(slave->dev,
+					 slave->dev->flags | IFF_UP);
+		}
+		read_unlock(&bond->lock);
+	} else {
+		dev_change_flags(dev, dev->flags & (~IFF_UP));
+		dev_change_flags(dev, dev->flags | IFF_UP);
+	}
+}
+EXPORT_SYMBOL_GPL(dev_change_state);
+
+static int mp_page_pool_attach(struct mp_struct *mp, struct page_pool *pool)
+{
+	int rc = 0;
+	/* should be protected by mp_mutex */
+	if (mp->pool) {
+		rc = -EBUSY;
+		goto fail;
+	}
+	if (mp->dev != pool->dev) {
+		rc = -EFAULT;
+		goto fail;
+	}
+	mp->pool = pool;
+	return 0;
+fail:
+	kfree(pool->hash_table);
+	kfree(pool);
+	return rc;
+}
+
+struct page_info *info_dequeue(struct page_pool *pool)
+{
+	unsigned long flags;
+	struct page_info *info = NULL;
+	spin_lock_irqsave(&pool->read_lock, flags);
+	if (!list_empty(&pool->readq)) {
+		info = list_first_entry(&pool->readq,
+				struct page_info, list);
+		list_del(&info->list);
+	}
+	spin_unlock_irqrestore(&pool->read_lock, flags);
+	return info;
+}
+
+static void mp_ki_dtor(struct kiocb *iocb)
+{
+	struct page_info *info = (struct page_info *)(iocb->private);
+	int i;
+
+	if (info->flags == INFO_READ) {
+		for (i = 0; i < info->pnum; i++) {
+			if (info->pages[i]) {
+				set_page_dirty_lock(info->pages[i]);
+				put_page(info->pages[i]);
+			}
+		}
+		mp_hash_delete(info->pool, info);
+		if (info->skb) {
+			info->skb->destructor = NULL;
+			kfree_skb(info->skb);
+		}
+	}
+	/* Decrement the number of locked pages */
+	info->pool->cur_pages -= info->pnum;
+	kmem_cache_free(ext_page_info_cache, info);
+
+	return;
+}
+
+static struct kiocb *create_iocb(struct page_info *info, int size)
+{
+	struct kiocb *iocb = NULL;
+
+	iocb = info->iocb;
+	if (!iocb)
+		return iocb;
+	iocb->ki_flags = 0;
+	iocb->ki_users = 1;
+	iocb->ki_key = 0;
+	iocb->ki_ctx = NULL;
+	iocb->ki_cancel = NULL;
+	iocb->ki_retry = NULL;
+	iocb->ki_eventfd = NULL;
+	iocb->ki_pos = info->desc_pos;
+	iocb->ki_nbytes = size;
+	iocb->ki_dtor(iocb);
+	iocb->private = (void *)info;
+	iocb->ki_dtor = mp_ki_dtor;
+
+	return iocb;
+}
+
+void page_pool_destroy(struct mm_struct *mm, struct page_pool *pool)
+{
+	struct page_info *info;
+	struct net_device *master;
+	struct slave *slave;
+	struct bonding *bond;
+	int i;
+
+	if (!pool)
+		return;
+
+	while ((info = info_dequeue(pool))) {
+		for (i = 0; i < info->pnum; i++)
+			if (info->pages[i])
+				put_page(info->pages[i]);
+		create_iocb(info, 0);
+		kmem_cache_free(ext_page_info_cache, info);
+	}
+	down_write(&mm->mmap_sem);
+	mm->locked_vm -= pool->locked_pages;
+	up_write(&mm->mmap_sem);
+
+	master = pool->dev->master;
+	if (master) {
+		bond = netdev_priv(master);
+		read_lock(&bond->lock);
+		bond_for_each_slave(bond, slave, i) {
+			slave->dev->mp_port = NULL;
+			dev_put(slave->dev);
+		}
+		read_unlock(&bond->lock);
+	} else {
+		pool->dev->mp_port = NULL;
+		dev_put(pool->dev);
+	}
+
+	kfree(pool->hash_table);
+	kfree(pool);
+}
+EXPORT_SYMBOL_GPL(page_pool_destroy);
+
+static void mp_page_pool_detach(struct mp_struct *mp)
+{
+	/* locked by mp_mutex */
+	if (mp->pool) {
+		page_pool_destroy(mp->mm, mp->pool);
+		mp->pool = NULL;
+	}
+}
+
+static void __mp_detach(struct mp_struct *mp)
+{
+	struct net_device *master;
+	struct bonding *bond;
+	struct slave *slave;
+	int i;
+
+	mp->mfile = NULL;
+	master = mp->dev->master;
+	if (master) {
+		bond = netdev_priv(master);
+		read_lock(&bond->lock);
+		bond_for_each_slave(bond, slave, i)
+			dev_change_flags(slave->dev,
+					 slave->dev->flags & ~IFF_UP);
+		read_unlock(&bond->lock);
+	} else
+		dev_change_flags(mp->dev, mp->dev->flags & ~IFF_UP);
+	mp_page_pool_detach(mp);
+	if (master) {
+		bond = netdev_priv(master);
+		read_lock(&bond->lock);
+		bond_for_each_slave(bond, slave, i)
+			dev_change_flags(slave->dev,
+					 slave->dev->flags | IFF_UP);
+		read_unlock(&bond->lock);
+	} else
+		dev_change_flags(mp->dev, mp->dev->flags | IFF_UP);
+}
+
+static DEFINE_MUTEX(mp_mutex);
+
+static void mp_detach(struct mp_struct *mp)
+{
+	mutex_lock(&mp_mutex);
+	__mp_detach(mp);
+	mutex_unlock(&mp_mutex);
+}
+
+static struct mp_struct *mp_get(struct mp_file *mfile)
+{
+	struct mp_struct *mp = NULL;
+	if (atomic_inc_not_zero(&mfile->count))
+		mp = mfile->mp;
+
+	return mp;
+}
+
+static void mp_put(struct mp_file *mfile)
+{
+	if (atomic_dec_and_test(&mfile->count)) {
+		if (mfile->mp)
+			return;
+		if (!rtnl_is_locked()) {
+			rtnl_lock();
+			mp_detach(mfile->mp);
+			rtnl_unlock();
+		} else
+			mp_detach(mfile->mp);
+	}
+}
+
+static void iocb_tag(struct kiocb *iocb)
+{
+	iocb->ki_flags = 1;
+}
+
+/* The callback to destruct the external buffers or skb */
+static void page_dtor(struct skb_ext_page *ext_page)
+{
+	struct page_info *info;
+	struct page_pool *pool;
+	struct sock *sk;
+	struct sk_buff *skb;
+
+	if (!ext_page)
+		return;
+	info = container_of(ext_page, struct page_info, ext_page);
+	if (!info)
+		return;
+	pool = info->pool;
+	skb = info->skb;
+
+	if (info->flags == INFO_READ) {
+		create_iocb(info, 0);
+		return;
+	}
+
+	/* For transmit, we should wait for the DMA finish by hardware.
+	 * Queue the notifier to wake up the backend driver
+	 */
+
+	iocb_tag(info->iocb);
+	sk = pool->port.sock->sk;
+	sk->sk_write_space(sk);
+
+	return;
+}
+
+/* For small exteranl buffers transmit, we don't need to call
+ * get_user_pages().
+ */
+static struct page_info *alloc_small_page_info(struct page_pool *pool,
+		struct kiocb *iocb, int total)
+{
+	struct page_info *info =
+		kmem_cache_alloc(ext_page_info_cache, GFP_KERNEL);
+
+	if (!info)
+		return NULL;
+	info->ext_page.dtor = page_dtor;
+	info->pool = pool;
+	info->flags = INFO_WRITE;
+	info->iocb = iocb;
+	info->pnum = 0;
+	return info;
+}
+
+typedef u32 key_mp_t;
+static inline key_mp_t mp_hash(struct page *page, int buckets)
+{
+	key_mp_t k;
+#if BITS_PER_LONG == 64
+	k = ((((unsigned long)page << 32UL) >> 32UL) /
+			sizeof(struct page)) % buckets ;
+#elif BITS_PER_LONG == 32
+	k = ((unsigned long)page / sizeof(struct page)) % buckets;
+#endif
+
+	return k;
+}
+
+static void mp_hash_insert(struct page_pool *pool,
+		struct page *page, struct page_info *page_info)
+{
+	struct page_info *tmp;
+	key_mp_t key = mp_hash(page, HASH_BUCKETS);
+	if (!pool->hash_table[key]) {
+		pool->hash_table[key] = page_info;
+		return;
+	}
+
+	tmp = pool->hash_table[key];
+	while (tmp->next)
+		tmp = tmp->next;
+
+	tmp->next = page_info;
+	page_info->prev = tmp;
+	return;
+}
+
+static struct page_info *mp_hash_delete(struct page_pool *pool,
+					struct page_info *info)
+{
+	key_mp_t key = mp_hash(info->pages[0], HASH_BUCKETS);
+	struct page_info *tmp = NULL;
+
+	tmp = pool->hash_table[key];
+	while (tmp) {
+		if (tmp == info) {
+			if (!tmp->prev) {
+				pool->hash_table[key] = tmp->next;
+				if (tmp->next)
+					tmp->next->prev = NULL;
+			} else {
+				tmp->prev->next = tmp->next;
+				if (tmp->next)
+					tmp->next->prev = tmp->prev;
+			}
+			return tmp;
+		}
+		tmp = tmp->next;
+	}
+	return tmp;
+}
+
+static struct page_info *mp_hash_lookup(struct page_pool *pool,
+					struct page *page)
+{
+	key_mp_t key = mp_hash(page, HASH_BUCKETS);
+	struct page_info *tmp = NULL;
+
+	int i;
+	tmp = pool->hash_table[key];
+	while (tmp) {
+		for (i = 0; i < tmp->pnum; i++) {
+			if (tmp->pages[i] == page)
+				return tmp;
+		}
+		tmp = tmp->next;
+	}
+	return tmp;
+}
+
+/* The main function to transform the guest user space address
+ * to host kernel address via get_user_pages(). Thus the hardware
+ * can do DMA directly to the external buffer address.
+ */
+static struct page_info *alloc_page_info(struct page_pool *pool,
+		struct kiocb *iocb, struct iovec *iov,
+		int count, struct frag *frags,
+		int npages, int total)
+{
+	int rc;
+	int i, j, n = 0;
+	int len;
+	unsigned long base;
+	struct page_info *info = NULL;
+
+	if (pool->cur_pages + count > pool->locked_pages) {
+		printk(KERN_INFO "Exceed memory lock rlimt.");
+		return NULL;
+	}
+
+	info = kmem_cache_alloc(ext_page_info_cache, GFP_KERNEL);
+
+	if (!info)
+		return NULL;
+	info->skb = NULL;
+	info->next = info->prev = NULL;
+
+	for (i = j = 0; i < count; i++) {
+		base = (unsigned long)iov[i].iov_base;
+		len = iov[i].iov_len;
+
+		if (!len)
+			continue;
+		n = ((base & ~PAGE_MASK) + len + ~PAGE_MASK) >> PAGE_SHIFT;
+
+		rc = get_user_pages_fast(base, n, npages ? 1 : 0,
+				&info->pages[j]);
+		if (rc != n)
+			goto failed;
+
+		while (n--) {
+			frags[j].offset = base & ~PAGE_MASK;
+			frags[j].size = min_t(int, len,
+					PAGE_SIZE - frags[j].offset);
+			len -= frags[j].size;
+			base += frags[j].size;
+			j++;
+		}
+	}
+
+#ifdef CONFIG_HIGHMEM
+	if (npages && !(dev->features & NETIF_F_HIGHDMA)) {
+		for (i = 0; i < j; i++) {
+			if (PageHighMem(info->pages[i]))
+				goto failed;
+		}
+	}
+#endif
+
+	info->ext_page.dtor = page_dtor;
+	info->ext_page.page = info->pages[0];
+	info->pool = pool;
+	info->pnum = j;
+	info->iocb = iocb;
+	if (!npages)
+		info->flags = INFO_WRITE;
+	else
+		info->flags = INFO_READ;
+
+	if (info->flags == INFO_READ) {
+		if (frags[0].offset == 0 && iocb->ki_iovec[0].iov_len) {
+			frags[0].offset = iocb->ki_iovec[0].iov_len;
+			pool->port.vnet_hlen = iocb->ki_iovec[0].iov_len;
+		}
+		for (i = 0; i < j; i++)
+			mp_hash_insert(pool, info->pages[i], info);
+	}
+	/* increment the number of locked pages */
+	pool->cur_pages += j;
+	return info;
+
+failed:
+	for (i = 0; i < j; i++)
+		put_page(info->pages[i]);
+
+	kmem_cache_free(ext_page_info_cache, info);
+
+	return NULL;
+}
+
+static void mp_sock_destruct(struct sock *sk)
+{
+	struct mp_struct *mp = container_of(sk, struct mp_sock, sk)->mp;
+	kfree(mp);
+}
+
+static void mp_sock_state_change(struct sock *sk)
+{
+	wait_queue_head_t *wqueue = sk_sleep(sk);
+	if (wqueue && waitqueue_active(wqueue))
+		wake_up_interruptible_sync_poll(wqueue, POLLIN);
+}
+
+static void mp_sock_write_space(struct sock *sk)
+{
+	wait_queue_head_t *wqueue = sk_sleep(sk);
+	if (wqueue && waitqueue_active(wqueue))
+		wake_up_interruptible_sync_poll(wqueue, POLLOUT);
+}
+
+void async_data_ready(struct sock *sk, struct page_pool *pool)
+{
+	struct sk_buff *skb = NULL;
+	struct page_info *info = NULL;
+	int len;
+
+	while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) {
+		struct page *page;
+		int off;
+		int size = 0, i = 0;
+		struct skb_shared_info *shinfo = skb_shinfo(skb);
+		struct skb_ext_page *ext_page =
+			(struct skb_ext_page *)(shinfo->destructor_arg);
+
+		struct virtio_net_hdr_mrg_rxbuf hdr = {
+			.hdr.flags = 0,
+			.hdr.gso_type = VIRTIO_NET_HDR_GSO_NONE
+		};
+
+		if (!ext_page) {
+			kfree_skb(skb);
+			continue;
+		}
+		if (skb->ip_summed == CHECKSUM_COMPLETE)
+			printk(KERN_INFO "Complete checksum occurs\n");
+
+		if (shinfo->frags[0].page == ext_page->page) {
+			info = container_of(ext_page,
+					    struct page_info,
+					    ext_page);
+			if (shinfo->nr_frags)
+				hdr.num_buffers = shinfo->nr_frags;
+			else
+				hdr.num_buffers = shinfo->nr_frags + 1;
+		} else {
+			info = container_of(ext_page,
+					    struct page_info,
+					    ext_page);
+			hdr.num_buffers = shinfo->nr_frags + 1;
+		}
+		skb_push(skb, ETH_HLEN);
+
+		if (skb_is_gso(skb)) {
+			hdr.hdr.hdr_len = skb_headlen(skb);
+			hdr.hdr.gso_size = shinfo->gso_size;
+			if (shinfo->gso_type & SKB_GSO_TCPV4)
+				hdr.hdr.gso_type = VIRTIO_NET_HDR_GSO_TCPV4;
+			else if (shinfo->gso_type & SKB_GSO_TCPV6)
+				hdr.hdr.gso_type = VIRTIO_NET_HDR_GSO_TCPV6;
+			else if (shinfo->gso_type & SKB_GSO_UDP)
+				hdr.hdr.gso_type = VIRTIO_NET_HDR_GSO_UDP;
+			else
+				BUG();
+			if (shinfo->gso_type & SKB_GSO_TCP_ECN)
+				hdr.hdr.gso_type |= VIRTIO_NET_HDR_GSO_ECN;
+
+		} else
+			hdr.hdr.gso_type = VIRTIO_NET_HDR_GSO_NONE;
+
+		if (skb->ip_summed == CHECKSUM_PARTIAL) {
+			hdr.hdr.flags = VIRTIO_NET_HDR_F_NEEDS_CSUM;
+			hdr.hdr.csum_start =
+				skb->csum_start - skb_headroom(skb);
+			hdr.hdr.csum_offset = skb->csum_offset;
+		}
+
+		off = info->hdr[0].iov_len;
+		len = memcpy_toiovec(info->iov, (unsigned char *)&hdr, off);
+		if (len) {
+			pr_debug("Unable to write vnet_hdr at addr '%p': '%d'\n",
+				info->iov, len);
+			goto clean;
+		}
+
+		memcpy_toiovec(info->iov, skb->data, skb_headlen(skb));
+
+		info->iocb->ki_left = hdr.num_buffers;
+		if (shinfo->frags[0].page == ext_page->page) {
+			size = shinfo->frags[0].size +
+				shinfo->frags[0].page_offset - off;
+			i = 1;
+		} else {
+			size = skb_headlen(skb);
+			i = 0;
+		}
+		create_iocb(info, off + size);
+		for (i = i; i < shinfo->nr_frags; i++) {
+			page = shinfo->frags[i].page;
+			info = mp_hash_lookup(pool, shinfo->frags[i].page);
+			create_iocb(info, shinfo->frags[i].size);
+		}
+		info->skb = skb;
+		shinfo->nr_frags = 0;
+		shinfo->destructor_arg = NULL;
+		continue;
+clean:
+		kfree_skb(skb);
+		for (i = 0; i < info->pnum; i++)
+			put_page(info->pages[i]);
+		kmem_cache_free(ext_page_info_cache, info);
+	}
+	return;
+}
+EXPORT_SYMBOL_GPL(async_data_ready);
+
+static void mp_sock_data_ready(struct sock *sk, int coming)
+{
+	struct mp_struct *mp = container_of(sk, struct mp_sock, sk)->mp;
+	struct page_pool *pool = NULL;
+
+	pool = mp->pool;
+	if (!pool)
+		return;
+	return async_data_ready(sk, pool);
+}
+
+static inline struct sk_buff *mp_alloc_skb(struct sock *sk, size_t prepad,
+					   size_t len, size_t linear,
+					   int noblock, int *err)
+{
+	struct sk_buff *skb;
+
+	/* Under a page?  Don't bother with paged skb. */
+	if (prepad + len < PAGE_SIZE || !linear)
+		linear = len;
+
+	skb = sock_alloc_send_pskb(sk, prepad + linear, len - linear, noblock,
+			err);
+	if (!skb)
+		return NULL;
+
+	skb_reserve(skb, prepad);
+	skb_put(skb, linear);
+	skb->data_len = len - linear;
+	skb->len += len - linear;
+
+	return skb;
+}
+
+static int mp_skb_from_vnet_hdr(struct sk_buff *skb,
+		struct virtio_net_hdr *vnet_hdr)
+{
+	unsigned short gso_type = 0;
+	if (vnet_hdr->gso_type != VIRTIO_NET_HDR_GSO_NONE) {
+		switch (vnet_hdr->gso_type & ~VIRTIO_NET_HDR_GSO_ECN) {
+		case VIRTIO_NET_HDR_GSO_TCPV4:
+			gso_type = SKB_GSO_TCPV4;
+			break;
+		case VIRTIO_NET_HDR_GSO_TCPV6:
+			gso_type = SKB_GSO_TCPV6;
+			break;
+		case VIRTIO_NET_HDR_GSO_UDP:
+			gso_type = SKB_GSO_UDP;
+			break;
+		default:
+			return -EINVAL;
+		}
+
+		if (vnet_hdr->gso_type & VIRTIO_NET_HDR_GSO_ECN)
+			gso_type |= SKB_GSO_TCP_ECN;
+
+		if (vnet_hdr->gso_size == 0)
+			return -EINVAL;
+	}
+
+	if (vnet_hdr->flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) {
+		if (!skb_partial_csum_set(skb, vnet_hdr->csum_start,
+					vnet_hdr->csum_offset))
+			return -EINVAL;
+	}
+
+	if (vnet_hdr->gso_type != VIRTIO_NET_HDR_GSO_NONE) {
+		skb_shinfo(skb)->gso_size = vnet_hdr->gso_size;
+		skb_shinfo(skb)->gso_type = gso_type;
+
+		/* Header must be checked, and gso_segs computed. */
+		skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
+		skb_shinfo(skb)->gso_segs = 0;
+	}
+	return 0;
+}
+
+int async_sendmsg(struct sock *sk, struct kiocb *iocb, struct page_pool *pool,
+		  struct iovec *iov, int count)
+{
+	struct virtio_net_hdr vnet_hdr = {0};
+	int hdr_len = 0;
+	struct page_info *info = NULL;
+	struct frag frags[MAX_SKB_FRAGS];
+	struct sk_buff *skb;
+	int total = 0, header, n, i, len, rc;
+	unsigned long base;
+
+	total = iov_length(iov, count);
+
+	if (total < ETH_HLEN)
+		return -EINVAL;
+
+	if (total <= COPY_THRESHOLD)
+		goto copy;
+
+	n = 0;
+	for (i = 0; i < count; i++) {
+		base = (unsigned long)iov[i].iov_base;
+		len = iov[i].iov_len;
+		if (!len)
+			continue;
+		n += ((base & ~PAGE_MASK) + len + ~PAGE_MASK) >> PAGE_SHIFT;
+		if (n > MAX_SKB_FRAGS)
+			return -EINVAL;
+	}
+
+copy:
+	hdr_len = sizeof(vnet_hdr);
+	if ((total -= sizeof(struct virtio_net_hdr_mrg_rxbuf)) < 0)
+		return -EINVAL;
+
+	rc = memcpy_fromiovecend((void *)&vnet_hdr, iov, 0, hdr_len);
+	if (rc < 0)
+		return -EINVAL;
+
+	if ((vnet_hdr.flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) &&
+			vnet_hdr.csum_start + vnet_hdr.csum_offset + 2 >
+			vnet_hdr.hdr_len)
+		vnet_hdr.hdr_len = vnet_hdr.csum_start +
+			vnet_hdr.csum_offset + 2;
+
+	if (vnet_hdr.hdr_len > total)
+		return -EINVAL;
+
+	skb = mp_alloc_skb(sk, NET_IP_ALIGN, total,
+			vnet_hdr.hdr_len, 1, &rc);
+	if (!skb)
+		goto drop;
+
+        rc = skb_copy_datagram_from_iovec(skb, 0, iov,
+		sizeof(struct virtio_net_hdr_mrg_rxbuf), total);
+        if (rc)
+                goto drop;
+	skb_set_network_header(skb, ETH_HLEN);
+
+	skb_reset_mac_header(skb);
+	skb->protocol = eth_hdr(skb)->h_proto;
+
+	rc = mp_skb_from_vnet_hdr(skb, &vnet_hdr);
+	if (rc)
+		goto drop;
+
+	if (!pool->cur_pages)
+		sk->sk_state_change(sk);
+
+	skb->dev = pool->dev->master ? pool->dev->master : pool->dev;
+	dev_queue_xmit(skb);
+	return 0;
+
+drop:
+	kfree_skb(skb);
+	pool->dev->stats.tx_dropped++;
+	return -ENOMEM;
+}
+EXPORT_SYMBOL_GPL(async_sendmsg);
+
+static int mp_sendmsg(struct kiocb *iocb, struct socket *sock,
+		struct msghdr *m, size_t total_len)
+{
+	struct mp_struct *mp = container_of(sock->sk, struct mp_sock, sk)->mp;
+	struct page_pool *pool;
+	struct iovec *iov = m->msg_iov;
+	int count = m->msg_iovlen;
+
+	pool = mp->pool;
+	if (!pool)
+		return -ENODEV;
+	return async_sendmsg(sock->sk, iocb, pool, iov, count);
+}
+
+int async_recvmsg(struct kiocb *iocb, struct page_pool *pool,
+		  struct iovec *iov, int count, int flags)
+{
+	int npages, payload;
+	struct page_info *info;
+	struct frag frags[MAX_SKB_FRAGS];
+	unsigned long base;
+	int i, len;
+	unsigned long flag;
+
+	if (!(flags & MSG_DONTWAIT))
+		return -EINVAL;
+
+	if (!pool)
+		return -EINVAL;
+
+	/* Error detections in case invalid external buffer */
+	if (count > 2 && iov[1].iov_len < pool->port.hdr_len &&
+			pool->dev->features & NETIF_F_SG) {
+		return -EINVAL;
+	}
+
+	npages = pool->port.npages;
+	payload = pool->port.data_len;
+
+	/* If KVM guest virtio-net FE driver use SG feature */
+	if (count > 2) {
+		for (i = 2; i < count; i++) {
+			base = (unsigned long)iov[i].iov_base & ~PAGE_MASK;
+			len = iov[i].iov_len;
+			if (npages == 1)
+				len = min_t(int, len, PAGE_SIZE - base);
+			else if (base)
+				break;
+			payload -= len;
+			if (payload <= 0)
+				goto proceed;
+			if (npages == 1 || (len & ~PAGE_MASK))
+				break;
+		}
+	}
+
+	if ((((unsigned long)iov[1].iov_base & ~PAGE_MASK)
+				- NET_SKB_PAD - NET_IP_ALIGN) >= 0)
+		goto proceed;
+
+	return -EINVAL;
+proceed:
+	/* skip the virtnet head */
+	if (count > 1) {
+		iov++;
+		count--;
+	}
+
+	/* Translate address to kernel */
+	info = alloc_page_info(pool, iocb, iov, count, frags, npages, 0);
+	if (!info)
+		return -ENOMEM;
+	info->hdr[0].iov_base = iocb->ki_iovec[0].iov_base;
+	info->hdr[0].iov_len = iocb->ki_iovec[0].iov_len;
+	iocb->ki_iovec[0].iov_len = 0;
+	iocb->ki_left = 0;
+	info->desc_pos = iocb->ki_pos;
+
+	if (count > 1) {
+		iov--;
+		count++;
+	}
+
+	memcpy(info->iov, iov, sizeof(struct iovec) * count);
+
+	spin_lock_irqsave(&pool->read_lock, flag);
+	list_add_tail(&info->list, &pool->readq);
+	spin_unlock_irqrestore(&pool->read_lock, flag);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(async_recvmsg);
+
+static int mp_recvmsg(struct kiocb *iocb, struct socket *sock,
+		struct msghdr *m, size_t total_len,
+		int flags)
+{
+	struct mp_struct *mp = container_of(sock->sk, struct mp_sock, sk)->mp;
+	struct page_pool *pool;
+	struct iovec *iov = m->msg_iov;
+	int count = m->msg_iovlen;
+
+	pool = mp->pool;
+	if (!pool)
+		return -EINVAL;
+
+	return async_recvmsg(iocb, pool, iov, count, flags);
+}
+
+/* Ops structure to mimic raw sockets with mp device */
+static const struct proto_ops mp_socket_ops = {
+	.sendmsg = mp_sendmsg,
+	.recvmsg = mp_recvmsg,
+};
+
+static struct proto mp_proto = {
+	.name           = "mp",
+	.owner          = THIS_MODULE,
+	.obj_size       = sizeof(struct mp_sock),
+};
+
+static int mp_chr_open(struct inode *inode, struct file * file)
+{
+	struct mp_file *mfile;
+	cycle_kernel_lock();
+
+	pr_debug("mp: mp_chr_open\n");
+	mfile = kzalloc(sizeof(*mfile), GFP_KERNEL);
+	if (!mfile)
+		return -ENOMEM;
+	atomic_set(&mfile->count, 0);
+	mfile->mp = NULL;
+	mfile->net = get_net(current->nsproxy->net_ns);
+	file->private_data = mfile;
+	return 0;
+}
+
+static int mp_attach(struct mp_struct *mp, struct file *file)
+{
+	struct mp_file *mfile = file->private_data;
+	int err;
+
+	netif_tx_lock_bh(mp->dev);
+
+	err = -EINVAL;
+
+	if (mfile->mp)
+		goto out;
+
+	err = -EBUSY;
+	if (mp->mfile)
+		goto out;
+
+	err = 0;
+	mfile->mp = mp;
+	mp->mfile = mfile;
+	mp->socket.file = file;
+	sock_hold(mp->socket.sk);
+	atomic_inc(&mfile->count);
+
+out:
+	netif_tx_unlock_bh(mp->dev);
+	return err;
+}
+
+static int do_unbind(struct mp_file *mfile)
+{
+	struct mp_struct *mp = mp_get(mfile);
+
+	if (mp) {
+		mp_detach(mp);
+		sock_put(mp->socket.sk);
+	}
+	mp_put(mfile);
+	return 0;
+}
+
+static long mp_chr_ioctl(struct file *file, unsigned int cmd,
+		unsigned long arg)
+{
+	struct mp_file *mfile = file->private_data;
+	struct mp_struct *mp;
+	struct net_device *dev;
+	struct page_pool *pool;
+	void __user* argp = (void __user *)arg;
+	unsigned long  __user *limitp = argp;
+	struct ifreq ifr;
+	struct sock *sk;
+	unsigned long limit, locked, lock_limit;
+	int ret;
+
+	ret = -EINVAL;
+
+	switch (cmd) {
+	case MPASSTHRU_BINDDEV:
+		ret = -EFAULT;
+		if (copy_from_user(&ifr, argp, sizeof ifr))
+			break;
+
+		ifr.ifr_name[IFNAMSIZ-1] = '\0';
+
+		ret = -ENODEV;
+
+		rtnl_lock();
+		dev = dev_get_by_name(mfile->net, ifr.ifr_name);
+		if (!dev) {
+			rtnl_unlock();
+			break;
+		}
+		dev_bond_hold(dev);
+		mutex_lock(&mp_mutex);
+
+		ret = -EBUSY;
+
+		/* the device can be only bind once */
+		if (dev_is_mpassthru(dev))
+			goto err_dev_put;
+
+		ret = -EFAULT;
+
+		if (!(dev->features & NETIF_F_SG)) {
+			pr_debug("The device has no SG features.\n");
+			goto err_dev_put;
+		}
+		mp = mfile->mp;
+		if (mp)
+			goto err_dev_put;
+
+		mp = kzalloc(sizeof(*mp), GFP_KERNEL);
+		if (!mp) {
+			ret = -ENOMEM;
+			goto err_dev_put;
+		}
+		mp->dev = dev;
+		mp->mm = get_task_mm(current);
+		ret = -ENOMEM;
+
+		sk = sk_alloc(mfile->net, AF_UNSPEC, GFP_KERNEL, &mp_proto);
+		if (!sk)
+			goto err_free_mp;
+
+		mp->socket.wq = &mp->wq;
+		init_waitqueue_head(&mp->wq.wait);
+		mp->socket.ops = &mp_socket_ops;
+		sock_init_data(&mp->socket, sk);
+		sk->sk_sndbuf = INT_MAX;
+		container_of(sk, struct mp_sock, sk)->mp = mp;
+
+		sk->sk_destruct = mp_sock_destruct;
+		sk->sk_data_ready = mp_sock_data_ready;
+		sk->sk_write_space = mp_sock_write_space;
+		sk->sk_state_change = mp_sock_state_change;
+
+		pool = page_pool_create(dev, &mp->socket);
+		if (!pool) {
+			ret = -EFAULT;
+			goto err_free_sk;
+		}
+
+		ret = mp_attach(mp, file);
+		if (ret < 0)
+			goto err_free_sk;
+
+		ret = mp_page_pool_attach(mp, pool);
+		if (ret < 0)
+			goto err_free_sk;
+		dev_bond_put(dev);
+		dev_put(dev);
+		dev_change_state(dev);
+out:
+		mutex_unlock(&mp_mutex);
+		rtnl_unlock();
+		break;
+err_free_sk:
+		sk_free(sk);
+err_free_mp:
+		mfile->mp = NULL;
+		kfree(mp);
+err_dev_put:
+		dev_bond_put(dev);
+		dev_put(dev);
+		goto out;
+
+	case MPASSTHRU_UNBINDDEV:
+		rtnl_lock();
+		ret = do_unbind(mfile);
+		rtnl_unlock();
+		break;
+
+	case MPASSTHRU_SET_MEM_LOCKED:
+		ret = copy_from_user(&limit, limitp, sizeof limit);
+		if (ret < 0)
+			return ret;
+
+		mp = mp_get(mfile);
+		if (!mp)
+			return -ENODEV;
+
+		mutex_lock(&mp_mutex);
+		if (mp->mm != current->mm) {
+			mutex_unlock(&mp_mutex);
+			return -EPERM;
+		}
+
+		limit = PAGE_ALIGN(limit) >> PAGE_SHIFT;
+		down_write(&mp->mm->mmap_sem);
+		if (!mp->pool->locked_pages)
+			mp->pool->orig_locked_vm = mp->mm->locked_vm;
+		locked = limit + mp->pool->orig_locked_vm;
+		lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
+
+		if ((locked > lock_limit) && !capable(CAP_IPC_LOCK)) {
+			up_write(&mp->mm->mmap_sem);
+			mutex_unlock(&mp_mutex);
+			mp_put(mfile);
+			return -ENOMEM;
+		}
+		mp->mm->locked_vm = locked;
+		up_write(&mp->mm->mmap_sem);
+
+		mp->pool->locked_pages = limit;
+		mutex_unlock(&mp_mutex);
+
+		mp_put(mfile);
+		return 0;
+
+	case MPASSTHRU_GET_MEM_LOCKED_NEED:
+		limit = DEFAULT_NEED;
+		return copy_to_user(limitp, &limit, sizeof limit);
+
+
+	default:
+		break;
+	}
+	return ret;
+}
+
+static unsigned int mp_chr_poll(struct file *file, poll_table * wait)
+{
+	struct mp_file *mfile = file->private_data;
+	struct mp_struct *mp = mp_get(mfile);
+	struct sock *sk;
+	unsigned int mask = 0;
+
+	if (!mp)
+		return POLLERR;
+
+	sk = mp->socket.sk;
+
+	poll_wait(file, &mp->wq.wait, wait);
+
+	if (!skb_queue_empty(&sk->sk_receive_queue))
+		mask |= POLLIN | POLLRDNORM;
+
+	if (sock_writeable(sk) ||
+		(!test_and_set_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags) &&
+			 sock_writeable(sk)))
+		mask |= POLLOUT | POLLWRNORM;
+
+	if (mp->dev->reg_state != NETREG_REGISTERED)
+		mask = POLLERR;
+
+	mp_put(mfile);
+	return mask;
+}
+
+static ssize_t mp_chr_aio_write(struct kiocb *iocb, const struct iovec *iov,
+				unsigned long count, loff_t pos)
+{
+	struct file *file = iocb->ki_filp;
+	struct mp_struct *mp = mp_get(file->private_data);
+	struct sock *sk = mp->socket.sk;
+	struct sk_buff *skb;
+	int len, err;
+	ssize_t result = 0;
+
+	if (!mp)
+		return -EBADFD;
+
+	/* currently, async is not supported.
+	 * but we may support real async aio from user application,
+	 * maybe qemu virtio-net backend.
+	 */
+	if (!is_sync_kiocb(iocb))
+		return -EFAULT;
+
+	len = iov_length(iov, count);
+
+	if (unlikely(len < ETH_HLEN))
+		return -EINVAL;
+
+	skb = sock_alloc_send_skb(sk, len + NET_IP_ALIGN,
+				  file->f_flags & O_NONBLOCK, &err);
+
+	if (!skb)
+		return -ENOMEM;
+
+	skb_reserve(skb, NET_IP_ALIGN);
+	skb_put(skb, len);
+
+	if (skb_copy_datagram_from_iovec(skb, 0, iov, 0, len)) {
+		kfree_skb(skb);
+		return -EAGAIN;
+	}
+
+	skb->protocol = eth_type_trans(skb, mp->dev);
+	skb->dev = mp->dev;
+
+	dev_queue_xmit(skb);
+
+	mp_put(file->private_data);
+	return result;
+}
+
+static int mp_chr_close(struct inode *inode, struct file *file)
+{
+	struct mp_file *mfile = file->private_data;
+
+	/*
+	 * Ignore return value since an error only means there was nothing to
+	 * do
+	 */
+	rtnl_lock();
+	do_unbind(mfile);
+	rtnl_unlock();
+	put_net(mfile->net);
+	kfree(mfile);
+
+	return 0;
+}
+
+#ifdef CONFIG_COMPAT
+static long mp_chr_compat_ioctl(struct file *f, unsigned int ioctl,
+				unsigned long arg)
+{
+	return mp_chr_ioctl(f, ioctl, (unsigned long)compat_ptr(arg));
+}
+#endif
+
+static const struct file_operations mp_fops = {
+	.owner  = THIS_MODULE,
+	.llseek = no_llseek,
+	.write  = do_sync_write,
+	.aio_write = mp_chr_aio_write,
+	.poll   = mp_chr_poll,
+	.unlocked_ioctl = mp_chr_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl = mp_chr_compat_ioctl,
+#endif
+	.open   = mp_chr_open,
+	.release = mp_chr_close,
+};
+
+static struct miscdevice mp_miscdev = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "mp",
+	.nodename = "net/mp",
+	.fops = &mp_fops,
+};
+
+static int mp_device_event(struct notifier_block *unused,
+		unsigned long event, void *ptr)
+{
+	struct net_device *dev = ptr;
+	struct mp_port *port;
+	struct mp_struct *mp = NULL;
+	struct socket *sock = NULL;
+	struct sock *sk;
+
+	port = dev->mp_port;
+	if (port == NULL)
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case NETDEV_UNREGISTER:
+		sock = dev->mp_port->sock;
+		mp = container_of(sock->sk, struct mp_sock, sk)->mp;
+		do_unbind(mp->mfile);
+		break;
+	case NETDEV_CHANGE:
+		sk = dev->mp_port->sock->sk;
+		sk->sk_state_change(sk);
+		break;
+	}
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block mp_notifier_block __read_mostly = {
+	.notifier_call  = mp_device_event,
+};
+
+static int mp_init(void)
+{
+	int err = 0;
+
+	ext_page_info_cache = kmem_cache_create("skb_page_info",
+						sizeof(struct page_info),
+						0, SLAB_HWCACHE_ALIGN, NULL);
+	if (!ext_page_info_cache)
+		return -ENOMEM;
+
+	err = misc_register(&mp_miscdev);
+	if (err) {
+		printk(KERN_ERR "mp: Can't register misc device\n");
+		kmem_cache_destroy(ext_page_info_cache);
+	} else {
+		printk(KERN_INFO "Registering mp misc device - minor = %d\n",
+				mp_miscdev.minor);
+		register_netdevice_notifier(&mp_notifier_block);
+	}
+	return err;
+}
+
+void mp_exit(void)
+{
+	unregister_netdevice_notifier(&mp_notifier_block);
+	misc_deregister(&mp_miscdev);
+	kmem_cache_destroy(ext_page_info_cache);
+}
+
+/* Get an underlying socket object from mp file.  Returns error unless file is
+ * attached to a device.  The returned object works like a packet socket, it
+ * can be used for sock_sendmsg/sock_recvmsg.  The caller is responsible for
+ * holding a reference to the file for as long as the socket is in use. */
+struct socket *mp_get_socket(struct file *file)
+{
+	struct mp_file *mfile = file->private_data;
+	struct mp_struct *mp;
+
+	if (file->f_op != &mp_fops)
+		return ERR_PTR(-EINVAL);
+	mp = mp_get(mfile);
+	if (!mp)
+		return ERR_PTR(-EBADFD);
+	mp_put(mfile);
+	return &mp->socket;
+}
+EXPORT_SYMBOL_GPL(mp_get_socket);
+
+module_init(mp_init);
+module_exit(mp_exit);
+MODULE_AUTHOR(DRV_COPYRIGHT);
+MODULE_DESCRIPTION(DRV_DESCRIPTION);
+MODULE_LICENSE("GPL v2");
-- 
1.7.3

[PATCH v16 14/17]Add a kconfig entry and make entry for mp device.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
Reviewed-by: Jeff Dike <jdike@linux.intel.com>
---
 drivers/vhost/Kconfig  |   10 ++++++++++
 drivers/vhost/Makefile |    2 ++
 2 files changed, 12 insertions(+), 0 deletions(-)

diff --git a/drivers/vhost/Kconfig b/drivers/vhost/Kconfig
index e4e2fd1..a6b8cbf 100644
--- a/drivers/vhost/Kconfig
+++ b/drivers/vhost/Kconfig
@@ -9,3 +9,13 @@ config VHOST_NET
 	  To compile this driver as a module, choose M here: the module will
 	  be called vhost_net.
 
+config MEDIATE_PASSTHRU
+	tristate "mediate passthru network driver (EXPERIMENTAL)"
+	depends on VHOST_NET
+	---help---
+	  zerocopy network I/O support, we call it as mediate passthru to
+	  be distiguish with hardare passthru.
+
+	  To compile this driver as a module, choose M here: the module will
+	  be called mpassthru.
+
diff --git a/drivers/vhost/Makefile b/drivers/vhost/Makefile
index 72dd020..c18b9fc 100644
--- a/drivers/vhost/Makefile
+++ b/drivers/vhost/Makefile
@@ -1,2 +1,4 @@
 obj-$(CONFIG_VHOST_NET) += vhost_net.o
 vhost_net-y := vhost.o net.o
+
+obj-$(CONFIG_MEDIATE_PASSTHRU) += mpassthru.o
-- 
1.7.3

Re: [PATCH v16 14/17]Add a kconfig entry and make entry for mp device.

[Randy Dunlap]

On Wed,  1 Dec 2010 16:08:25 +0800 xiaohui.xin@intel.com wrote:

> From: Xin Xiaohui <xiaohui.xin@intel.com>
> 
> Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
> Reviewed-by: Jeff Dike <jdike@linux.intel.com>
> ---
>  drivers/vhost/Kconfig  |   10 ++++++++++
>  drivers/vhost/Makefile |    2 ++
>  2 files changed, 12 insertions(+), 0 deletions(-)
> 
> diff --git a/drivers/vhost/Kconfig b/drivers/vhost/Kconfig
> index e4e2fd1..a6b8cbf 100644
> --- a/drivers/vhost/Kconfig
> +++ b/drivers/vhost/Kconfig
> @@ -9,3 +9,13 @@ config VHOST_NET
>  	  To compile this driver as a module, choose M here: the module will
>  	  be called vhost_net.
>  
> +config MEDIATE_PASSTHRU
> +	tristate "mediate passthru network driver (EXPERIMENTAL)"
> +	depends on VHOST_NET
> +	---help---
> +	  zerocopy network I/O support, we call it as mediate passthru to

	                       support; we call it "mediate passthru" to

> +	  be distiguish with hardare passthru.

	  distinguish it from hardware passthru.

> +
> +	  To compile this driver as a module, choose M here: the module will
> +	  be called mpassthru.
> +


---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

RE: [PATCH v16 14/17]Add a kconfig entry and make entry for mp device.

[Xin, Xiaohui]

>-----Original Message-----
>From: Randy Dunlap [mailto:randy.dunlap@oracle.com]
>Sent: Thursday, December 02, 2010 1:54 AM
>To: Xin, Xiaohui
>Cc: netdev@vger.kernel.org; kvm@vger.kernel.org; linux-kernel@vger.kernel.org;
>mst@redhat.com; mingo@elte.hu; davem@davemloft.net; herbert@gondor.hengli.com.au;
>jdike@linux.intel.com
>Subject: Re: [PATCH v16 14/17]Add a kconfig entry and make entry for mp device.
>
>On Wed,  1 Dec 2010 16:08:25 +0800 xiaohui.xin@intel.com wrote:
>
>> From: Xin Xiaohui <xiaohui.xin@intel.com>
>>
>> Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
>> Reviewed-by: Jeff Dike <jdike@linux.intel.com>
>> ---
>>  drivers/vhost/Kconfig  |   10 ++++++++++
>>  drivers/vhost/Makefile |    2 ++
>>  2 files changed, 12 insertions(+), 0 deletions(-)
>>
>> diff --git a/drivers/vhost/Kconfig b/drivers/vhost/Kconfig
>> index e4e2fd1..a6b8cbf 100644
>> --- a/drivers/vhost/Kconfig
>> +++ b/drivers/vhost/Kconfig
>> @@ -9,3 +9,13 @@ config VHOST_NET
>>  	  To compile this driver as a module, choose M here: the module will
>>  	  be called vhost_net.
>>
>> +config MEDIATE_PASSTHRU
>> +	tristate "mediate passthru network driver (EXPERIMENTAL)"
>> +	depends on VHOST_NET
>> +	---help---
>> +	  zerocopy network I/O support, we call it as mediate passthru to
>
>	                       support; we call it "mediate passthru" to
>
>> +	  be distiguish with hardare passthru.
>
>	  distinguish it from hardware passthru.
>

Thanks. I will modify that.

Thanks
Xiaohui

>> +
>> +	  To compile this driver as a module, choose M here: the module will
>> +	  be called mpassthru.
>> +
>
>
>---
>~Randy
>*** Remember to use Documentation/SubmitChecklist when testing your code ***

[PATCH v16 15/17]Provides multiple submits and asynchronous notifications.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

    The vhost-net backend now only supports synchronous send/recv
    operations. The patch provides multiple submits and asynchronous
    notifications. This is needed for zero-copy case.

    Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
---
 drivers/vhost/net.c   |  361 +++++++++++++++++++++++++++++++++++++++++++++----
 drivers/vhost/vhost.c |   78 +++++++++++
 drivers/vhost/vhost.h |   15 ++-
 3 files changed, 429 insertions(+), 25 deletions(-)

diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 7c80082..8ec4edf 100644
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -24,6 +24,8 @@
 #include <linux/if_arp.h>
 #include <linux/if_tun.h>
 #include <linux/if_macvlan.h>
+#include <linux/mpassthru.h>
+#include <linux/aio.h>
 
 #include <net/sock.h>
 
@@ -32,6 +34,7 @@
 /* Max number of bytes transferred before requeueing the job.
  * Using this limit prevents one virtqueue from starving others. */
 #define VHOST_NET_WEIGHT 0x80000
+static struct kmem_cache *notify_cache;
 
 enum {
 	VHOST_NET_VQ_RX = 0,
@@ -49,6 +52,7 @@ struct vhost_net {
 	struct vhost_dev dev;
 	struct vhost_virtqueue vqs[VHOST_NET_VQ_MAX];
 	struct vhost_poll poll[VHOST_NET_VQ_MAX];
+	struct kmem_cache *cache;
 	/* Tells us whether we are polling a socket for TX.
 	 * We only do this when socket buffer fills up.
 	 * Protected by tx vq lock. */
@@ -109,11 +113,184 @@ static void tx_poll_start(struct vhost_net *net, struct socket *sock)
 	net->tx_poll_state = VHOST_NET_POLL_STARTED;
 }
 
+struct kiocb *notify_dequeue(struct vhost_virtqueue *vq)
+{
+	struct kiocb *iocb = NULL;
+	unsigned long flags;
+
+	spin_lock_irqsave(&vq->notify_lock, flags);
+	if (!list_empty(&vq->notifier)) {
+		iocb = list_first_entry(&vq->notifier,
+				struct kiocb, ki_list);
+		list_del(&iocb->ki_list);
+	}
+	spin_unlock_irqrestore(&vq->notify_lock, flags);
+	return iocb;
+}
+
+static void handle_iocb(struct kiocb *iocb)
+{
+	struct vhost_virtqueue *vq = iocb->private;
+	unsigned long flags;
+
+	spin_lock_irqsave(&vq->notify_lock, flags);
+	list_add_tail(&iocb->ki_list, &vq->notifier);
+	spin_unlock_irqrestore(&vq->notify_lock, flags);
+}
+
+static int is_async_vq(struct vhost_virtqueue *vq)
+{
+	return (vq->link_state == VHOST_VQ_LINK_ASYNC);
+}
+
+static void handle_async_rx_events_notify(struct vhost_net *net,
+		struct vhost_virtqueue *vq,
+		struct socket *sock)
+{
+	struct kiocb *iocb = NULL;
+	struct vhost_log *vq_log = NULL;
+	int rx_total_len = 0;
+	unsigned int head, log, in, out;
+	int size;
+
+	if (!is_async_vq(vq))
+		return;
+
+	if (sock->sk->sk_data_ready)
+		sock->sk->sk_data_ready(sock->sk, 0);
+
+	vq_log = unlikely(vhost_has_feature(&net->dev, VHOST_F_LOG_ALL)) ?
+		vq->log : NULL;
+
+	while ((iocb = notify_dequeue(vq)) != NULL) {
+		if (!iocb->ki_left) {
+			vhost_add_used_and_signal(&net->dev, vq,
+					iocb->ki_pos, iocb->ki_nbytes);
+			size = iocb->ki_nbytes;
+			head = iocb->ki_pos;
+			rx_total_len += iocb->ki_nbytes;
+
+			if (iocb->ki_dtor)
+				iocb->ki_dtor(iocb);
+			kmem_cache_free(net->cache, iocb);
+
+			/* when log is enabled, recomputing the log is needed,
+			 * since these buffers are in async queue, may not get
+			 * the log info before.
+			 */
+			if (unlikely(vq_log)) {
+				if (!log)
+					__vhost_get_vq_desc(&net->dev, vq,
+							vq->iov,
+							ARRAY_SIZE(vq->iov),
+							&out, &in, vq_log,
+							&log, head);
+				vhost_log_write(vq, vq_log, log, size);
+			}
+			if (unlikely(rx_total_len >= VHOST_NET_WEIGHT)) {
+				vhost_poll_queue(&vq->poll);
+				break;
+			}
+		} else {
+			int i = 0;
+			int count = iocb->ki_left;
+			int hc = count;
+			while (count--) {
+				if (iocb) {
+					vq->heads[i].id = iocb->ki_pos;
+					vq->heads[i].len = iocb->ki_nbytes;
+					size = iocb->ki_nbytes;
+					head = iocb->ki_pos;
+					rx_total_len += iocb->ki_nbytes;
+
+					if (iocb->ki_dtor)
+						iocb->ki_dtor(iocb);
+					kmem_cache_free(net->cache, iocb);
+
+					if (unlikely(vq_log)) {
+						if (!log)
+							__vhost_get_vq_desc(
+							&net->dev, vq, vq->iov,
+							ARRAY_SIZE(vq->iov),
+							&out, &in, vq_log,
+							&log, head);
+						vhost_log_write(
+							vq, vq_log, log, size);
+					}
+				} else
+					break;
+
+				i++;
+				if (count)
+					iocb = notify_dequeue(vq);
+			}
+			vhost_add_used_and_signal_n(
+					&net->dev, vq, vq->heads, hc);
+		}
+	}
+}
+
+static void handle_async_tx_events_notify(struct vhost_net *net,
+		struct vhost_virtqueue *vq)
+{
+	struct kiocb *iocb = NULL;
+	struct list_head *entry, *tmp;
+	unsigned long flags;
+	int tx_total_len = 0;
+
+	if (!is_async_vq(vq))
+		return;
+
+	spin_lock_irqsave(&vq->notify_lock, flags);
+	list_for_each_safe(entry, tmp, &vq->notifier) {
+		iocb = list_entry(entry,
+				struct kiocb, ki_list);
+		if (!iocb->ki_flags)
+			continue;
+		list_del(&iocb->ki_list);
+		vhost_add_used_and_signal(&net->dev, vq,
+				iocb->ki_pos, 0);
+		tx_total_len += iocb->ki_nbytes;
+
+		if (iocb->ki_dtor)
+			iocb->ki_dtor(iocb);
+
+		kmem_cache_free(net->cache, iocb);
+		if (unlikely(tx_total_len >= VHOST_NET_WEIGHT)) {
+			vhost_poll_queue(&vq->poll);
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&vq->notify_lock, flags);
+}
+
+static struct kiocb *create_iocb(struct vhost_net *net,
+		struct vhost_virtqueue *vq,
+		unsigned head)
+{
+	struct kiocb *iocb = NULL;
+
+	if (!is_async_vq(vq))
+		return NULL;
+
+	iocb = kmem_cache_zalloc(net->cache, GFP_KERNEL);
+	if (!iocb)
+		return NULL;
+	iocb->private = vq;
+	iocb->ki_pos = head;
+	iocb->ki_dtor = handle_iocb;
+	if (vq == &net->dev.vqs[VHOST_NET_VQ_RX])
+		iocb->ki_user_data = vq->num;
+	iocb->ki_iovec = vq->hdr;
+	return iocb;
+}
+
 /* Expects to be always run from workqueue - which acts as
  * read-size critical section for our kind of RCU. */
 static void handle_tx(struct vhost_net *net)
 {
 	struct vhost_virtqueue *vq = &net->dev.vqs[VHOST_NET_VQ_TX];
+	struct kiocb *iocb = NULL;
 	unsigned out, in, s;
 	int head;
 	struct msghdr msg = {
@@ -146,6 +323,10 @@ static void handle_tx(struct vhost_net *net)
 	if (wmem < sock->sk->sk_sndbuf / 2)
 		tx_poll_stop(net);
 	hdr_size = vq->vhost_hlen;
+	if (!vq->vhost_hlen && is_async_vq(vq))
+		hdr_size = vq->sock_hlen;
+
+	handle_async_tx_events_notify(net, vq);
 
 	for (;;) {
 		head = vhost_get_vq_desc(&net->dev, vq, vq->iov,
@@ -157,11 +338,14 @@ static void handle_tx(struct vhost_net *net)
 			break;
 		/* Nothing new?  Wait for eventfd to tell us they refilled. */
 		if (head == vq->num) {
-			wmem = atomic_read(&sock->sk->sk_wmem_alloc);
-			if (wmem >= sock->sk->sk_sndbuf * 3 / 4) {
-				tx_poll_start(net, sock);
-				set_bit(SOCK_ASYNC_NOSPACE, &sock->flags);
-				break;
+			if (!is_async_vq(vq)) {
+				wmem = atomic_read(&sock->sk->sk_wmem_alloc);
+				if (wmem >= sock->sk->sk_sndbuf * 3 / 4) {
+					tx_poll_start(net, sock);
+					set_bit(SOCK_ASYNC_NOSPACE,
+					&sock->flags);
+					break;
+				}
 			}
 			if (unlikely(vhost_enable_notify(vq))) {
 				vhost_disable_notify(vq);
@@ -178,6 +362,13 @@ static void handle_tx(struct vhost_net *net)
 		s = move_iovec_hdr(vq->iov, vq->hdr, hdr_size, out);
 		msg.msg_iovlen = out;
 		len = iov_length(vq->iov, out);
+		/* if async operations supported */
+		if (is_async_vq(vq)) {
+			iocb = create_iocb(net, vq, head);
+			if (!iocb)
+				break;
+		}
+
 		/* Sanity check */
 		if (!len) {
 			vq_err(vq, "Unexpected header len for TX: "
@@ -186,12 +377,18 @@ static void handle_tx(struct vhost_net *net)
 			break;
 		}
 		/* TODO: Check specific error and bomb out unless ENOBUFS? */
-		err = sock->ops->sendmsg(NULL, sock, &msg, len);
+		err = sock->ops->sendmsg(iocb, sock, &msg, len);
 		if (unlikely(err < 0)) {
+			if (is_async_vq(vq))
+				kmem_cache_free(net->cache, iocb);
 			vhost_discard_vq_desc(vq, 1);
 			tx_poll_start(net, sock);
 			break;
 		}
+
+		if (is_async_vq(vq))
+			continue;
+
 		if (err != len)
 			pr_debug("Truncated TX packet: "
 				 " len %d != %zd\n", err, len);
@@ -203,6 +400,8 @@ static void handle_tx(struct vhost_net *net)
 		}
 	}
 
+	handle_async_tx_events_notify(net, vq);
+
 	mutex_unlock(&vq->mutex);
 	unuse_mm(net->dev.mm);
 }
@@ -396,7 +595,8 @@ static void handle_rx_big(struct vhost_net *net)
 static void handle_rx_mergeable(struct vhost_net *net)
 {
 	struct vhost_virtqueue *vq = &net->dev.vqs[VHOST_NET_VQ_RX];
-	unsigned uninitialized_var(in), log;
+	unsigned uninitialized_var(in), log, out;
+	struct kiocb *iocb;
 	struct vhost_log *vq_log;
 	struct msghdr msg = {
 		.msg_name = NULL,
@@ -417,28 +617,44 @@ static void handle_rx_mergeable(struct vhost_net *net)
 	size_t vhost_hlen, sock_hlen;
 	size_t vhost_len, sock_len;
 	struct socket *sock = rcu_dereference(vq->private_data);
-	if (!sock || skb_queue_empty(&sock->sk->sk_receive_queue))
+	if (!sock || (skb_queue_empty(&sock->sk->sk_receive_queue) &&
+		      !is_async_vq(vq)))
 		return;
-
 	use_mm(net->dev.mm);
 	mutex_lock(&vq->mutex);
 	vhost_disable_notify(vq);
 	vhost_hlen = vq->vhost_hlen;
 	sock_hlen = vq->sock_hlen;
 
+	/* In async cases, when write log is enabled, in case the submitted
+	 * buffers did not get log info before the log enabling, so we'd
+	 * better recompute the log info when needed. We do this in
+	 * handle_async_rx_events_notify().
+	 */
+
 	vq_log = unlikely(vhost_has_feature(&net->dev, VHOST_F_LOG_ALL)) ?
 		vq->log : NULL;
 
-	while ((sock_len = peek_head_len(sock->sk))) {
-		sock_len += sock_hlen;
-		vhost_len = sock_len + vhost_hlen;
-		headcount = get_rx_bufs(vq, vq->heads, vhost_len,
+	handle_async_rx_events_notify(net, vq, sock);
+
+	while (is_async_vq(vq) || (sock_len = peek_head_len(sock->sk))) {
+		if (is_async_vq(vq))
+			headcount = vhost_get_vq_desc(&net->dev, vq, vq->iov,
+						      ARRAY_SIZE(vq->iov),
+						      &out, &in,
+						      vq->log, &log);
+		else {
+			sock_len += sock_hlen;
+			vhost_len = sock_len + vhost_hlen;
+			headcount = get_rx_bufs(vq, vq->heads, vhost_len,
 					&in, vq_log, &log);
+		}
 		/* On error, stop handling until the next kick. */
 		if (unlikely(headcount < 0))
 			break;
 		/* OK, now we need to know about added descriptors. */
-		if (!headcount) {
+		if ((!headcount && !is_async_vq(vq)) ||
+			(headcount == vq->num && is_async_vq(vq))) {
 			if (unlikely(vhost_enable_notify(vq))) {
 				/* They have slipped one in as we were
 				 * doing that: check again. */
@@ -450,16 +666,41 @@ static void handle_rx_mergeable(struct vhost_net *net)
 			break;
 		}
 		/* We don't need to be notified again. */
-		if (unlikely((vhost_hlen)))
-			/* Skip header. TODO: support TSO. */
-			move_iovec_hdr(vq->iov, vq->hdr, vhost_hlen, in);
-		else
-			/* Copy the header for use in VIRTIO_NET_F_MRG_RXBUF:
-			 * needed because sendmsg can modify msg_iov. */
-			copy_iovec_hdr(vq->iov, vq->hdr, sock_hlen, in);
+		if (unlikely((vhost_hlen))) {
+			if (is_async_vq(vq))
+				vq->hdr[0].iov_len = vhost_hlen;
+			else
+				/* Skip header. TODO: support TSO. */
+				move_iovec_hdr(vq->iov, vq->hdr,
+						vhost_hlen, in);
+		} else {
+			if (is_async_vq(vq))
+				vq->hdr[0].iov_len = sock_hlen;
+			else
+				/* Copy the header for use in
+				 * VIRTIO_NET_F_MRG_RXBUF:
+				 * needed because sendmsg can
+				 * modify msg_iov. */
+				copy_iovec_hdr(vq->iov, vq->hdr,
+						sock_hlen, in);
+		}
 		msg.msg_iovlen = in;
-		err = sock->ops->recvmsg(NULL, sock, &msg,
+		if (is_async_vq(vq)) {
+			iocb = create_iocb(net, vq, headcount);
+			if (!iocb)
+				break;
+		}
+		err = sock->ops->recvmsg(iocb, sock, &msg,
 					 sock_len, MSG_DONTWAIT | MSG_TRUNC);
+		if (is_async_vq(vq)) {
+			if (err < 0) {
+				kmem_cache_free(net->cache, iocb);
+				vhost_discard_vq_desc(vq, headcount);
+				break;
+			}
+			continue;
+		}
+
 		/* Userspace might have consumed the packet meanwhile:
 		 * it's not supposed to do this usually, but might be hard
 		 * to prevent. Discard data we got (if any) and keep going. */
@@ -496,6 +737,8 @@ static void handle_rx_mergeable(struct vhost_net *net)
 		}
 	}
 
+	handle_async_rx_events_notify(net, vq, sock);
+
 	mutex_unlock(&vq->mutex);
 	unuse_mm(net->dev.mm);
 }
@@ -561,6 +804,7 @@ static int vhost_net_open(struct inode *inode, struct file *f)
 	vhost_poll_init(n->poll + VHOST_NET_VQ_TX, handle_tx_net, POLLOUT, dev);
 	vhost_poll_init(n->poll + VHOST_NET_VQ_RX, handle_rx_net, POLLIN, dev);
 	n->tx_poll_state = VHOST_NET_POLL_DISABLED;
+	n->cache = NULL;
 
 	f->private_data = n;
 
@@ -624,6 +868,25 @@ static void vhost_net_flush(struct vhost_net *n)
 	vhost_net_flush_vq(n, VHOST_NET_VQ_RX);
 }
 
+static void vhost_async_cleanup(struct vhost_net *n)
+{
+	/* clean the notifier */
+	struct vhost_virtqueue *vq;
+	struct kiocb *iocb = NULL;
+	if (n->cache) {
+		vq = &n->dev.vqs[VHOST_NET_VQ_RX];
+		if (vq->link_state == VHOST_VQ_LINK_ASYNC) { 
+			while ((iocb = notify_dequeue(vq)) != NULL)
+				kmem_cache_free(n->cache, iocb);
+		}
+		vq = &n->dev.vqs[VHOST_NET_VQ_TX];
+		if (vq->link_state == VHOST_VQ_LINK_ASYNC) {
+			while ((iocb = notify_dequeue(vq)) != NULL)
+				kmem_cache_free(n->cache, iocb);
+		}
+	}
+}
+
 static int vhost_net_release(struct inode *inode, struct file *f)
 {
 	struct vhost_net *n = f->private_data;
@@ -640,6 +903,7 @@ static int vhost_net_release(struct inode *inode, struct file *f)
 	/* We do an extra flush before freeing memory,
 	 * since jobs can re-queue themselves. */
 	vhost_net_flush(n);
+	vhost_async_cleanup(n);
 	kfree(n);
 	return 0;
 }
@@ -691,21 +955,62 @@ static struct socket *get_tap_socket(int fd)
 	return sock;
 }
 
-static struct socket *get_socket(int fd)
+static struct socket *get_mp_socket(int fd)
+{
+	struct file *file = fget(fd);
+	struct socket *sock;
+	if (!file)
+		return ERR_PTR(-EBADF);
+	sock = mp_get_socket(file);
+	if (IS_ERR(sock))
+		fput(file);
+	return sock;
+}
+
+static struct socket *get_socket(struct vhost_virtqueue *vq, int fd,
+				 enum vhost_vq_link_state *state)
 {
 	struct socket *sock;
 	/* special case to disable backend */
 	if (fd == -1)
 		return NULL;
+
+	*state = VHOST_VQ_LINK_SYNC;
+
 	sock = get_raw_socket(fd);
 	if (!IS_ERR(sock))
 		return sock;
 	sock = get_tap_socket(fd);
 	if (!IS_ERR(sock))
 		return sock;
+	/* If we dont' have notify_cache, then dont do mpassthru */
+	if (!notify_cache)
+		return ERR_PTR(-ENOTSOCK);
+	/* If we don't have mergeable buffer then dont do mpassthru */
+	if (vhost_has_feature(vq->dev, VIRTIO_NET_F_MRG_RXBUF)) {
+		sock = get_mp_socket(fd);
+		if (!IS_ERR(sock)) {
+			*state = VHOST_VQ_LINK_ASYNC;
+			return sock;
+		}
+	}
 	return ERR_PTR(-ENOTSOCK);
 }
 
+static void vhost_init_link_state(struct vhost_net *n, int index)
+{
+	struct vhost_virtqueue *vq = n->vqs + index;
+
+	WARN_ON(!mutex_is_locked(&vq->mutex));
+	if (vq->link_state == VHOST_VQ_LINK_ASYNC &&
+		vq == &n->dev.vqs[VHOST_NET_VQ_RX]) {
+		INIT_LIST_HEAD(&vq->notifier);
+		spin_lock_init(&vq->notify_lock);
+		if (!n->cache)
+			n->cache = notify_cache;
+	}
+}
+
 static long vhost_net_set_backend(struct vhost_net *n, unsigned index, int fd)
 {
 	struct socket *sock, *oldsock;
@@ -729,11 +1034,14 @@ static long vhost_net_set_backend(struct vhost_net *n, unsigned index, int fd)
 		r = -EFAULT;
 		goto err_vq;
 	}
-	sock = get_socket(fd);
+	sock = get_socket(vq, fd, &vq->link_state);
 	if (IS_ERR(sock)) {
 		r = PTR_ERR(sock);
 		goto err_vq;
 	}
+	if (vq == &n->dev.vqs[VHOST_NET_VQ_TX])
+		vq->link_state = VHOST_VQ_LINK_SYNC;
+	vhost_init_link_state(n, index);
 
 	/* start polling new socket */
 	oldsock = vq->private_data;
@@ -879,6 +1187,9 @@ static struct miscdevice vhost_net_misc = {
 
 static int vhost_net_init(void)
 {
+	notify_cache = kmem_cache_create("vhost_kiocb",
+					sizeof(struct kiocb), 0,
+					SLAB_HWCACHE_ALIGN, NULL);
 	return misc_register(&vhost_net_misc);
 }
 module_init(vhost_net_init);
@@ -886,6 +1197,8 @@ module_init(vhost_net_init);
 static void vhost_net_exit(void)
 {
 	misc_deregister(&vhost_net_misc);
+	if (notify_cache)
+		kmem_cache_destroy(notify_cache);
 }
 module_exit(vhost_net_exit);
 
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index dd3d6f7..295d9ab 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -1015,6 +1015,84 @@ static int get_indirect(struct vhost_dev *dev, struct vhost_virtqueue *vq,
 	return 0;
 }
 
+/* To recompute the log */
+int __vhost_get_vq_desc(struct vhost_dev *dev, struct vhost_virtqueue *vq,
+			struct iovec iov[], unsigned int iov_size,
+			unsigned int *out_num, unsigned int *in_num,
+			struct vhost_log *log, unsigned int *log_num,
+			unsigned int head)
+{
+	struct vring_desc desc;
+	unsigned int i, found = 0;
+	int ret;
+
+	/* When we start there are none of either input nor output. */
+	*out_num = *in_num = 0;
+	if (unlikely(log))
+		*log_num = 0;
+
+	i = head;
+	do {
+		unsigned iov_count = *in_num + *out_num;
+		if (unlikely(i >= vq->num)) {
+			vq_err(vq, "Desc index is %u > %u, head = %u",
+					i, vq->num, head);
+			return -EINVAL;
+		}
+		if (unlikely(++found > vq->num)) {
+			vq_err(vq, "Loop detected: last one at %u "
+					"vq size %u head %u\n",
+					i, vq->num, head);
+			return -EINVAL;
+		}
+		ret = copy_from_user(&desc, vq->desc + i, sizeof desc);
+		if (unlikely(ret)) {
+			vq_err(vq, "Failed to get descriptor: idx %d addr %p\n",
+					i, vq->desc + i);
+			return -EFAULT;
+		}
+		if (desc.flags & VRING_DESC_F_INDIRECT) {
+			ret = get_indirect(dev, vq, iov, iov_size,
+					out_num, in_num,
+					log, log_num, &desc);
+			if (unlikely(ret < 0)) {
+				vq_err(vq, "Failure detected "
+				       "in indirect descriptor at idx %d\n", i);
+				return ret;
+			}
+			continue;
+		}
+
+		ret = translate_desc(dev, desc.addr, desc.len, iov + iov_count,
+				iov_size - iov_count);
+		if (unlikely(ret < 0)) {
+			vq_err(vq, "Translation failure %d descriptor idx %d\n",
+					ret, i);
+			return ret;
+		}
+		if (desc.flags & VRING_DESC_F_WRITE) {
+			/* If this is an input descriptor,
+			 * increment that count. */
+			*in_num += ret;
+			if (unlikely(log)) {
+				log[*log_num].addr = desc.addr;
+				log[*log_num].len = desc.len;
+				++*log_num;
+			}
+		} else {
+			/* If it's an output descriptor, they're all supposed
+			 * to come before any input descriptors. */
+			if (unlikely(*in_num)) {
+				vq_err(vq, "Descriptor has out after in: "
+						"idx %d\n", i);
+				return -EINVAL;
+			}
+			*out_num += ret;
+		}
+	} while ((i = next_desc(&desc)) != -1);
+
+	return head;
+}
 /* This looks in the virtqueue and for the first available buffer, and converts
  * it to an iovec for convenient access.  Since descriptors consist of some
  * number of output then some number of input descriptors, it's actually two
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index afd7729..915336d 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -55,6 +55,11 @@ struct vhost_log {
 	u64 len;
 };
 
+enum vhost_vq_link_state {
+	VHOST_VQ_LINK_SYNC = 0,
+	VHOST_VQ_LINK_ASYNC = 1,
+};
+
 /* The virtqueue structure describes a queue attached to a device. */
 struct vhost_virtqueue {
 	struct vhost_dev *dev;
@@ -110,6 +115,10 @@ struct vhost_virtqueue {
 	/* Log write descriptors */
 	void __user *log_base;
 	struct vhost_log log[VHOST_NET_MAX_SG];
+	/* Differiate async socket for 0-copy from normal */
+	enum vhost_vq_link_state link_state;
+	struct list_head notifier;
+	spinlock_t notify_lock;
 };
 
 struct vhost_dev {
@@ -136,7 +145,11 @@ void vhost_dev_cleanup(struct vhost_dev *);
 long vhost_dev_ioctl(struct vhost_dev *, unsigned int ioctl, unsigned long arg);
 int vhost_vq_access_ok(struct vhost_virtqueue *vq);
 int vhost_log_access_ok(struct vhost_dev *);
-
+int __vhost_get_vq_desc(struct vhost_dev *, struct vhost_virtqueue *,
+			  struct iovec iov[], unsigned int iov_count,
+			  unsigned int *out_num, unsigned int *in_num,
+			  struct vhost_log *log, unsigned int *log_num,
+			  unsigned int head);
 int vhost_get_vq_desc(struct vhost_dev *, struct vhost_virtqueue *,
 		      struct iovec iov[], unsigned int iov_count,
 		      unsigned int *out_num, unsigned int *in_num,
-- 
1.7.3

[PATCH v16 16/17]An example how to modifiy NIC driver to use napi_gro_frags() interface

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

This example is made on ixgbe driver.
It provides API is_rx_buffer_mapped_as_page() to indicate
if the driver use napi_gro_frags() interface or not.
The example allocates 2 pages for DMA for one ring descriptor
using netdev_alloc_page(). When packets is coming, using
napi_gro_frags() to allocate skb and to receive the packets.

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
---
 drivers/net/ixgbe/ixgbe.h      |    3 +
 drivers/net/ixgbe/ixgbe_main.c |  169 +++++++++++++++++++++++++++++++--------
 2 files changed, 137 insertions(+), 35 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe.h b/drivers/net/ixgbe/ixgbe.h
index 9e15eb9..89367ca 100644
--- a/drivers/net/ixgbe/ixgbe.h
+++ b/drivers/net/ixgbe/ixgbe.h
@@ -131,6 +131,9 @@ struct ixgbe_rx_buffer {
 	struct page *page;
 	dma_addr_t page_dma;
 	unsigned int page_offset;
+	u16 mapped_as_page;
+	struct page *page_skb;
+	unsigned int page_skb_offset;
 };
 
 struct ixgbe_queue_stats {
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index e32af43..cd69080 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -1029,6 +1029,12 @@ static inline void ixgbe_release_rx_desc(struct ixgbe_hw *hw,
 	IXGBE_WRITE_REG(hw, IXGBE_RDT(rx_ring->reg_idx), val);
 }
 
+static bool is_rx_buffer_mapped_as_page(struct ixgbe_rx_buffer *bi,
+					struct net_device *dev)
+{
+	return true;
+}
+
 /**
  * ixgbe_alloc_rx_buffers - Replace used receive buffers; packet split
  * @adapter: address of board private structure
@@ -1045,13 +1051,17 @@ static void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
 	i = rx_ring->next_to_use;
 	bi = &rx_ring->rx_buffer_info[i];
 
+
 	while (cleaned_count--) {
 		rx_desc = IXGBE_RX_DESC_ADV(*rx_ring, i);
 
+		bi->mapped_as_page =
+			is_rx_buffer_mapped_as_page(bi, adapter->netdev);
+
 		if (!bi->page_dma &&
 		    (rx_ring->flags & IXGBE_RING_RX_PS_ENABLED)) {
 			if (!bi->page) {
-				bi->page = alloc_page(GFP_ATOMIC);
+				bi->page = netdev_alloc_page(adapter->netdev);
 				if (!bi->page) {
 					adapter->alloc_rx_page_failed++;
 					goto no_buffers;
@@ -1068,7 +1078,7 @@ static void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
 						    DMA_FROM_DEVICE);
 		}
 
-		if (!bi->skb) {
+		if (!bi->mapped_as_page && !bi->skb) {
 			struct sk_buff *skb;
 			/* netdev_alloc_skb reserves 32 bytes up front!! */
 			uint bufsz = rx_ring->rx_buf_len + SMP_CACHE_BYTES;
@@ -1088,6 +1098,19 @@ static void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
 			                         rx_ring->rx_buf_len,
 						 DMA_FROM_DEVICE);
 		}
+
+		if (bi->mapped_as_page && !bi->page_skb) {
+			bi->page_skb = netdev_alloc_page(adapter->netdev);
+			if (!bi->page_skb) {
+				adapter->alloc_rx_page_failed++;
+				goto no_buffers;
+			}
+			bi->page_skb_offset = 0;
+			bi->dma = dma_map_page(&pdev->dev, bi->page_skb,
+					bi->page_skb_offset,
+					(PAGE_SIZE / 2),
+					PCI_DMA_FROMDEVICE);
+		}
 		/* Refresh the desc even if buffer_addrs didn't change because
 		 * each write-back erases this info. */
 		if (rx_ring->flags & IXGBE_RING_RX_PS_ENABLED) {
@@ -1165,6 +1188,13 @@ struct ixgbe_rsc_cb {
 	bool delay_unmap;
 };
 
+static bool is_no_buffer(struct ixgbe_rx_buffer *rx_buffer_info)
+{
+	return (!rx_buffer_info->skb ||
+		!rx_buffer_info->page_skb) &&
+		!rx_buffer_info->page;
+}
+
 #define IXGBE_RSC_CB(skb) ((struct ixgbe_rsc_cb *)(skb)->cb)
 
 static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
@@ -1174,6 +1204,7 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 	struct ixgbe_adapter *adapter = q_vector->adapter;
 	struct net_device *netdev = adapter->netdev;
 	struct pci_dev *pdev = adapter->pdev;
+	struct napi_struct *napi = &q_vector->napi;
 	union ixgbe_adv_rx_desc *rx_desc, *next_rxd;
 	struct ixgbe_rx_buffer *rx_buffer_info, *next_buffer;
 	struct sk_buff *skb;
@@ -1211,32 +1242,74 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 			len = le16_to_cpu(rx_desc->wb.upper.length);
 		}
 
+		if (is_no_buffer(rx_buffer_info))
+			break;
 		cleaned = true;
-		skb = rx_buffer_info->skb;
-		prefetch(skb->data);
-		rx_buffer_info->skb = NULL;
 
-		if (rx_buffer_info->dma) {
-			if ((adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED) &&
-			    (!(staterr & IXGBE_RXD_STAT_EOP)) &&
-				 (!(skb->prev))) {
-				/*
-				 * When HWRSC is enabled, delay unmapping
-				 * of the first packet. It carries the
-				 * header information, HW may still
-				 * access the header after the writeback.
-				 * Only unmap it when EOP is reached
-				 */
-				IXGBE_RSC_CB(skb)->delay_unmap = true;
-				IXGBE_RSC_CB(skb)->dma = rx_buffer_info->dma;
-			} else {
-				dma_unmap_single(&pdev->dev,
-				                 rx_buffer_info->dma,
-				                 rx_ring->rx_buf_len,
-				                 DMA_FROM_DEVICE);
+		if (!rx_buffer_info->mapped_as_page) {
+			skb = rx_buffer_info->skb;
+			prefetch(skb->data);
+			rx_buffer_info->skb = NULL;
+
+			if (rx_buffer_info->dma) {
+				if ((adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED) &&
+						(!(staterr & IXGBE_RXD_STAT_EOP)) &&
+						(!(skb->prev))) {
+					/*
+					 * When HWRSC is enabled, delay unmapping
+					 * of the first packet. It carries the
+					 * header information, HW may still
+					 * access the header after the writeback.
+					 * Only unmap it when EOP is reached
+					 */
+					IXGBE_RSC_CB(skb)->delay_unmap = true;
+					IXGBE_RSC_CB(skb)->dma = rx_buffer_info->dma;
+				} else
+					dma_unmap_single(&pdev->dev,
+							rx_buffer_info->dma,
+							rx_ring->rx_buf_len,
+							DMA_FROM_DEVICE);
+				rx_buffer_info->dma = 0;
+				skb_put(skb, len);
+			}
+		} else {
+                        if (!rx_buffer_info->skb && rx_buffer_info->page_skb) {
+                                napi->skb = NULL;
+                                skb = napi_get_frags(napi);
+                        } else {
+                                skb = rx_buffer_info->skb;
+                                napi->skb = skb;
+                        }
+			prefetch(rx_buffer_info->page_skb_offset);
+			rx_buffer_info->skb = NULL;
+			if (rx_buffer_info->dma) {
+				if ((adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED) &&
+						(!(staterr & IXGBE_RXD_STAT_EOP)) &&
+						(!(skb->prev))) {
+					/*
+					 * When HWRSC is enabled, delay unmapping
+					 * of the first packet. It carries the
+					 * header information, HW may still
+					 * access the header after the writeback.
+					 * Only unmap it when EOP is reached
+					 */
+					IXGBE_RSC_CB(skb)->delay_unmap = true;
+					IXGBE_RSC_CB(skb)->dma = rx_buffer_info->dma;
+				} else
+					dma_unmap_page(&pdev->dev, rx_buffer_info->dma,
+							PAGE_SIZE / 2,
+							PCI_DMA_FROMDEVICE);
+				rx_buffer_info->dma = 0;
+				skb_fill_page_desc(skb,
+						skb_shinfo(skb)->nr_frags,
+						rx_buffer_info->page_skb,
+						rx_buffer_info->page_skb_offset,
+						len);
+				rx_buffer_info->page_skb = NULL;
+				skb->len += len;
+				skb->data_len += len;
+				skb->truesize += len;
 			}
-			rx_buffer_info->dma = 0;
-			skb_put(skb, len);
 		}
 
 		if (upper_len) {
@@ -1283,10 +1356,16 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 				skb = ixgbe_transform_rsc_queue(skb, &(rx_ring->rsc_count));
 			if (adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED) {
 				if (IXGBE_RSC_CB(skb)->delay_unmap) {
-					dma_unmap_single(&pdev->dev,
-							 IXGBE_RSC_CB(skb)->dma,
-					                 rx_ring->rx_buf_len,
-							 DMA_FROM_DEVICE);
+					if (!rx_buffer_info->mapped_as_page)
+						dma_unmap_single(&pdev->dev,
+								IXGBE_RSC_CB(skb)->dma,
+								rx_ring->rx_buf_len,
+								DMA_FROM_DEVICE);
+					else
+						dma_unmap_page(&pdev->dev,
+								IXGBE_RSC_CB(skb)->dma,
+								PAGE_SIZE / 2,
+								DMA_FROM_DEVICE);
 					IXGBE_RSC_CB(skb)->dma = 0;
 					IXGBE_RSC_CB(skb)->delay_unmap = false;
 				}
@@ -1304,6 +1383,11 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 				rx_buffer_info->dma = next_buffer->dma;
 				next_buffer->skb = skb;
 				next_buffer->dma = 0;
+				if (rx_buffer_info->mapped_as_page) {
+					rx_buffer_info->page_skb =
+							next_buffer->page_skb;
+					next_buffer->page_skb = NULL;
+				}
 			} else {
 				skb->next = next_buffer->skb;
 				skb->next->prev = skb;
@@ -1323,7 +1407,8 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 		total_rx_bytes += skb->len;
 		total_rx_packets++;
 
-		skb->protocol = eth_type_trans(skb, adapter->netdev);
+		if (!rx_buffer_info->mapped_as_page)
+			skb->protocol = eth_type_trans(skb, adapter->netdev);
 #ifdef IXGBE_FCOE
 		/* if ddp, not passing to ULD unless for FCP_RSP or error */
 		if (adapter->flags & IXGBE_FLAG_FCOE_ENABLED) {
@@ -1332,7 +1417,14 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 				goto next_desc;
 		}
 #endif /* IXGBE_FCOE */
-		ixgbe_receive_skb(q_vector, skb, staterr, rx_ring, rx_desc);
+
+		if (!rx_buffer_info->mapped_as_page)
+			ixgbe_receive_skb(q_vector, skb, staterr,
+						rx_ring, rx_desc);
+		else {
+			skb_record_rx_queue(skb, rx_ring->queue_index);
+			napi_gro_frags(napi);
+		}
 
 next_desc:
 		rx_desc->wb.upper.status_error = 0;
@@ -3622,9 +3714,16 @@ static void ixgbe_clean_rx_ring(struct ixgbe_adapter *adapter,
 
 		rx_buffer_info = &rx_ring->rx_buffer_info[i];
 		if (rx_buffer_info->dma) {
-			dma_unmap_single(&pdev->dev, rx_buffer_info->dma,
-			                 rx_ring->rx_buf_len,
-					 DMA_FROM_DEVICE);
+			if (!rx_buffer_info->mapped_as_page)
+				dma_unmap_single(&pdev->dev, rx_buffer_info->dma,
+						rx_ring->rx_buf_len,
+						PCI_DMA_FROMDEVICE);
+			else {
+				dma_unmap_page(&pdev->dev, rx_buffer_info->dma,
+						PAGE_SIZE / 2,
+						PCI_DMA_FROMDEVICE);
+				rx_buffer_info->page_skb = NULL;
+			}
 			rx_buffer_info->dma = 0;
 		}
 		if (rx_buffer_info->skb) {
@@ -3651,7 +3750,7 @@ static void ixgbe_clean_rx_ring(struct ixgbe_adapter *adapter,
 				       PAGE_SIZE / 2, DMA_FROM_DEVICE);
 			rx_buffer_info->page_dma = 0;
 		}
-		put_page(rx_buffer_info->page);
+		netdev_free_page(adapter->netdev, rx_buffer_info->page);
 		rx_buffer_info->page = NULL;
 		rx_buffer_info->page_offset = 0;
 	}
-- 
1.7.3

[PATCH v16 17/17]An example how to alloc user buffer based on napi_gro_frags() interface.

[xiaohui.xin]

From: Xin Xiaohui <xiaohui.xin@intel.com>

This example is made on ixgbe driver which using napi_gro_frags().
It can get buffers from guest side directly using netdev_alloc_page()
and release guest buffers using netdev_free_page().

Signed-off-by: Xin Xiaohui <xiaohui.xin@intel.com>
---
 drivers/net/ixgbe/ixgbe_main.c |   37 +++++++++++++++++++++++++++++++++----
 1 files changed, 33 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index cd69080..807a51e 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -1032,7 +1032,14 @@ static inline void ixgbe_release_rx_desc(struct ixgbe_hw *hw,
 static bool is_rx_buffer_mapped_as_page(struct ixgbe_rx_buffer *bi,
 					struct net_device *dev)
 {
-	return true;
+	return dev_is_mpassthru(dev);
+}
+
+static u32 get_page_skb_offset(struct net_device *dev)
+{
+	if (!dev_is_mpassthru(dev))
+		return 0;
+	return dev->mp_port->vnet_hlen;
 }
 
 /**
@@ -1105,7 +1112,8 @@ static void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
 				adapter->alloc_rx_page_failed++;
 				goto no_buffers;
 			}
-			bi->page_skb_offset = 0;
+			bi->page_skb_offset =
+				get_page_skb_offset(adapter->netdev);
 			bi->dma = dma_map_page(&pdev->dev, bi->page_skb,
 					bi->page_skb_offset,
 					(PAGE_SIZE / 2),
@@ -1242,8 +1250,10 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 			len = le16_to_cpu(rx_desc->wb.upper.length);
 		}
 
-		if (is_no_buffer(rx_buffer_info))
+		if (is_no_buffer(rx_buffer_info)) {
+			printk("no buffers\n");
 			break;
+		}
 		cleaned = true;
 
 		if (!rx_buffer_info->mapped_as_page) {
@@ -1305,6 +1315,11 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 						rx_buffer_info->page_skb,
 						rx_buffer_info->page_skb_offset,
 						len);
+				if (dev_is_mpassthru(netdev) &&
+						netdev->mp_port->hash)
+					skb_shinfo(skb)->destructor_arg =
+						netdev->mp_port->hash(netdev,
+						rx_buffer_info->page_skb);
 				rx_buffer_info->page_skb = NULL;
 				skb->len += len;
 				skb->data_len += len;
@@ -1322,7 +1337,8 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 			                   upper_len);
 
 			if ((rx_ring->rx_buf_len > (PAGE_SIZE / 2)) ||
-			    (page_count(rx_buffer_info->page) != 1))
+			    (page_count(rx_buffer_info->page) != 1) ||
+				dev_is_mpassthru(netdev))
 				rx_buffer_info->page = NULL;
 			else
 				get_page(rx_buffer_info->page);
@@ -6535,6 +6551,16 @@ static void ixgbe_netpoll(struct net_device *netdev)
 }
 #endif
 
+#if defined(CONFIG_MEDIATE_PASSTHRU) || defined(CONFIG_MEDIATE_PASSTHRU_MODULE)
+static int ixgbe_ndo_mp_port_prep(struct net_device *dev, struct mp_port *port)
+{
+	port->hdr_len = 128;
+	port->data_len = 2048;
+	port->npages = 1;
+	return 0;
+}
+#endif
+
 static const struct net_device_ops ixgbe_netdev_ops = {
 	.ndo_open 		= ixgbe_open,
 	.ndo_stop		= ixgbe_close,
@@ -6554,6 +6580,9 @@ static const struct net_device_ops ixgbe_netdev_ops = {
 	.ndo_set_vf_vlan	= ixgbe_ndo_set_vf_vlan,
 	.ndo_set_vf_tx_rate	= ixgbe_ndo_set_vf_bw,
 	.ndo_get_vf_config	= ixgbe_ndo_get_vf_config,
+#if defined(CONFIG_MEDIATE_PASSTHRU) || defined(CONFIG_MEDIATE_PASSTHRU_MODULE)
+	.ndo_mp_port_prep	= ixgbe_ndo_mp_port_prep,
+#endif
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	.ndo_poll_controller	= ixgbe_netpoll,
 #endif
-- 
1.7.3

[PATCH v16 00/17] Provide a zero-copy method on KVM virtio-net.

[xiaohui.xin]

We provide an zero-copy method which driver side may get external
buffers to DMA. Here external means driver don't use kernel space
to allocate skb buffers. Currently the external buffer can be from
guest virtio-net driver.

The idea is simple, just to pin the guest VM user space and then
let host NIC driver has the chance to directly DMA to it. 
The patches are based on vhost-net backend driver. We add a device
which provides proto_ops as sendmsg/recvmsg to vhost-net to
send/recv directly to/from the NIC driver. KVM guest who use the
vhost-net backend may bind any ethX interface in the host side to
get copyless data transfer thru guest virtio-net frontend.

patch 01-11:  	net core and kernel changes.
patch 12-14:  	new device as interface to mantpulate external buffers.
patch 15: 	for vhost-net.
patch 16:	An example on modifying NIC driver to using napi_gro_frags().
patch 17:	An example how to get guest buffers based on driver
		who using napi_gro_frags().

The guest virtio-net driver submits multiple requests thru vhost-net
backend driver to the kernel. And the requests are queued and then
completed after corresponding actions in h/w are done.

For read, user space buffers are dispensed to NIC driver for rx when
a page constructor API is invoked. Means NICs can allocate user buffers
from a page constructor. We add a hook in netif_receive_skb() function
to intercept the incoming packets, and notify the zero-copy device.

For write, the zero-copy deivce may allocates a new host skb and puts
payload on the skb_shinfo(skb)->frags, and copied the header to skb->data.
The request remains pending until the skb is transmitted by h/w.

We provide multiple submits and asynchronous notifiicaton to 
vhost-net too.

Our goal is to improve the bandwidth and reduce the CPU usage.
Exact performance data will be provided later.

What we have not done yet:
	Performance tuning

what we have done in v1:
	polish the RCU usage
	deal with write logging in asynchroush mode in vhost
	add notifier block for mp device
	rename page_ctor to mp_port in netdevice.h to make it looks generic
	add mp_dev_change_flags() for mp device to change NIC state
	add CONIFG_VHOST_MPASSTHRU to limit the usage when module is not load
	a small fix for missing dev_put when fail
	using dynamic minor instead of static minor number
	a __KERNEL__ protect to mp_get_sock()

what we have done in v2:
	
	remove most of the RCU usage, since the ctor pointer is only
	changed by BIND/UNBIND ioctl, and during that time, NIC will be
	stopped to get good cleanup(all outstanding requests are finished),
	so the ctor pointer cannot be raced into wrong situation.

	Remove the struct vhost_notifier with struct kiocb.
	Let vhost-net backend to alloc/free the kiocb and transfer them
	via sendmsg/recvmsg.

	use get_user_pages_fast() and set_page_dirty_lock() when read.

	Add some comments for netdev_mp_port_prep() and handle_mpassthru().

what we have done in v3:
	the async write logging is rewritten 
	a drafted synchronous write function for qemu live migration
	a limit for locked pages from get_user_pages_fast() to prevent Dos
	by using RLIMIT_MEMLOCK
	

what we have done in v4:
	add iocb completion callback from vhost-net to queue iocb in mp device
	replace vq->receiver by mp_sock_data_ready()
	remove stuff in mp device which access structures from vhost-net
	modify skb_reserve() to ignore host NIC driver reserved space
	rebase to the latest vhost tree
	split large patches into small pieces, especially for net core part.
	

what we have done in v5:
	address Arnd Bergmann's comments
		-remove IFF_MPASSTHRU_EXCL flag in mp device
		-Add CONFIG_COMPAT macro
		-remove mp_release ops
	move dev_is_mpassthru() as inline func
	fix a bug in memory relinquish
	Apply to current git (2.6.34-rc6) tree.

what we have done in v6:
	move create_iocb() out of page_dtor which may happen in interrupt context
	-This remove the potential issues which lock called in interrupt context
	make the cache used by mp, vhost as static, and created/destoryed during
	modules init/exit functions.
	-This makes multiple mp guest created at the same time.

what we have done in v7:
	some cleanup prepared to suppprt PS mode

what we have done in v8:
	discarding the modifications to point skb->data to guest buffer directly.
	Add code to modify driver to support napi_gro_frags() with Herbert's comments.
	To support PS mode.
	Add mergeable buffer support in mp device.
	Add GSO/GRO support in mp deice.
	Address comments from Eric Dumazet about cache line and rcu usage.

what we have done in v9:
	v8 patch is based on a fix in dev_gro_receive().
	But Herbert did not agree with the fix we have sent out.
	And he suggest another fix. v9 is modified to base on that fix.
	

what we have done in v10:
	Fix a partial csum error.
	Cleanup some unused fields with struct page_info{} in mp device.
	Modify kmem_cache_zalloc() to kmem_cache_alloc() based on Michael S. Thirkin.

what we have done in v11:
	Address comments from Michael S. Thirkin to add two new ioctls in mp device.
	But still need to revise.

what we have done in v12:
	Address most comments from Ben Hutchings, except the compat ioctls.
	As the comments are sparse, so do not make a split patch.
	Change struct mpassthru_port to struct mp_port, and struct page_ctor
	to struct page_pool.

what we have done in v13:
	Export functions to other drivers like macvtap, in case it want to reuse it to
	get zero-copy.
	Rebase on 2.6.36-rc7.

what we have done in v14:
	Address the comments from David Miller for bonding device issue.
	Currently, we treat it in two cases. One case is that bonding is created before
	zero-copy mode is enabled for a device. The code will check if all the slaves are
	capable of zero-copy. If yes, it will force all the slaves in zero-copy mode.
	If not, fails zero-copy. The other case is that zero-copy is enabled before bonding
	is created, just fail bonding.

what we have done in v15:
	Address comments from Eric Dumazet about how to clear destructor_arg field of shinfo.

what we have done in v16:
	Remove the modification to skb_release_data(), and don't touch the function now.
	Before we think it's simple to free the guest buffer in skb_release_data() when kernel
	wants to free the skb in case something is wrong. And now we think in RX zero-copy case,
	the skb will never tour into the stack, so we can only care if the driver wants to free
	the skb, and intercept the wrong skb there and then release the guest buffer. Thus we 
	can avoid to modify skb_release_data().

Performance:
	We have seen the performance data request from mailling-list.
	And we are now looking into this.