Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1566964 > unrolled thread

[PATCHv3 0/7] Refactor macvtap to re-use tap functionality by other virtual intefaces

Back to article view | Back to linux.kernel

Contents

  [PATCHv3 0/7] Refactor macvtap to re-use tap functionality by other virtual intefaces Sainath Grandhi <sainath.grandhi@intel.com> - 2017-01-25 22:30 +0100
    [PATCHv3 7/7] IPVTAP: IP-VLAN based tap driver Sainath Grandhi <sainath.grandhi@intel.com> - 2017-01-25 22:30 +0100
    [PATCHv3 5/7] TAP: Extending tap device create/destroy APIs Sainath Grandhi <sainath.grandhi@intel.com> - 2017-01-25 22:30 +0100
    [PATCHv3 4/7] TAP: Abstract type of virtual interface from tap  implementation Sainath Grandhi <sainath.grandhi@intel.com> - 2017-01-25 22:30 +0100
    Re: [PATCHv3 0/7] Refactor macvtap to re-use tap functionality by  other virtual intefaces David Miller <davem@davemloft.net> - 2017-01-26 05:10 +0100

#1566964 — [PATCHv3 0/7] Refactor macvtap to re-use tap functionality by other virtual intefaces

Tap character devices can be implemented on other virtual interfaces like
ipvlan, similar to macvtap. Source code for tap functionality in macvtap
can be re-used for this purpose.

This patch series splits macvtap source into two modules, macvtap and tap.
This patch series also includes a patch for implementing tap character
device driver based on the IP-VLAN network interface, called ipvtap.

These patches are tested on x86 platform.

Sainath Grandhi (7):
  TAP: Refactoring macvtap.c
  TAP: Renaming tap related APIs, data structures, macros
  TAP: Tap character device creation/destroy API
  TAP: Abstract type of virtual interface from tap  implementation
  TAP: Extending tap device create/destroy APIs
  TAP: tap as an independent module
  IPVTAP: IP-VLAN based tap driver

 drivers/net/Kconfig              |   28 +
 drivers/net/Makefile             |    2 +
 drivers/net/ipvlan/Makefile      |    1 +
 drivers/net/ipvlan/ipvlan.h      |    7 +
 drivers/net/ipvlan/ipvlan_core.c |    5 +-
 drivers/net/ipvlan/ipvlan_main.c |   27 +-
 drivers/net/ipvlan/ipvtap.c      |  241 ++++++++
 drivers/net/macvlan.c            |    2 +-
 drivers/net/macvtap.c            | 1229 ++-----------------------------------
 drivers/net/tap.c                | 1261 ++++++++++++++++++++++++++++++++++++++
 drivers/vhost/Kconfig            |    2 +-
 drivers/vhost/net.c              |    3 +-
 include/linux/if_macvlan.h       |   17 +-
 include/linux/if_tap.h           |   75 +++
 14 files changed, 1691 insertions(+), 1209 deletions(-)
 create mode 100644 drivers/net/ipvlan/ipvtap.c
 create mode 100644 drivers/net/tap.c
 create mode 100644 include/linux/if_tap.h

-- 
2.7.4

[toc] | [next] | [standalone]

#1566966 — [PATCHv3 7/7] IPVTAP: IP-VLAN based tap driver

This patch adds a tap character device driver that is based on the
IP-VLAN network interface, called ipvtap. An ipvtap device can be created
in the same way as an ipvlan device, using 'type ipvtap', and then accessed
using the tap user space interface.

Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
---
 drivers/net/Kconfig              |  13 +++
 drivers/net/Makefile             |   1 +
 drivers/net/ipvlan/Makefile      |   1 +
 drivers/net/ipvlan/ipvlan.h      |   7 ++
 drivers/net/ipvlan/ipvlan_core.c |   5 +-
 drivers/net/ipvlan/ipvlan_main.c |  27 +++--
 drivers/net/ipvlan/ipvtap.c      | 241 +++++++++++++++++++++++++++++++++++++++
 7 files changed, 281 insertions(+), 14 deletions(-)
 create mode 100644 drivers/net/ipvlan/ipvtap.c

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 1c88437..d07b5f5 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -166,6 +166,19 @@ config IPVLAN
       To compile this driver as a module, choose M here: the module
       will be called ipvlan.
 
+config IPVTAP
+	tristate "IP-VLAN based tap driver"
+	depends on IPVLAN
+	depends on INET
+	depends on TAP
+	---help---
+	  This adds a specialized tap character device driver that is based
+	  on the IP-VLAN network interface, called ipvtap. An ipvtap device
+	  can be added in the same way as a ipvlan device, using 'type
+	  ipvtap', and then be accessed through the tap user space interface.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called ipvtap.
 
 config VXLAN
        tristate "Virtual eXtensible Local Area Network (VXLAN)"
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 7dd86ca..98ed4d9 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -7,6 +7,7 @@
 #
 obj-$(CONFIG_BONDING) += bonding/
 obj-$(CONFIG_IPVLAN) += ipvlan/
+obj-$(CONFIG_IPVTAP) += ipvlan/
 obj-$(CONFIG_DUMMY) += dummy.o
 obj-$(CONFIG_EQUALIZER) += eql.o
 obj-$(CONFIG_IFB) += ifb.o
diff --git a/drivers/net/ipvlan/Makefile b/drivers/net/ipvlan/Makefile
index df79910..8a2c64d 100644
--- a/drivers/net/ipvlan/Makefile
+++ b/drivers/net/ipvlan/Makefile
@@ -3,5 +3,6 @@
 #
 
 obj-$(CONFIG_IPVLAN) += ipvlan.o
+obj-$(CONFIG_IPVTAP) += ipvtap.o
 
 ipvlan-objs := ipvlan_core.o ipvlan_main.o
diff --git a/drivers/net/ipvlan/ipvlan.h b/drivers/net/ipvlan/ipvlan.h
index dbfbb33..4362d88 100644
--- a/drivers/net/ipvlan/ipvlan.h
+++ b/drivers/net/ipvlan/ipvlan.h
@@ -133,4 +133,11 @@ struct sk_buff *ipvlan_l3_rcv(struct net_device *dev, struct sk_buff *skb,
 			      u16 proto);
 unsigned int ipvlan_nf_input(void *priv, struct sk_buff *skb,
 			     const struct nf_hook_state *state);
+void ipvlan_count_rx(const struct ipvl_dev *ipvlan,
+		     unsigned int len, bool success, bool mcast);
+int ipvlan_link_new(struct net *src_net, struct net_device *dev,
+		    struct nlattr *tb[], struct nlattr *data[]);
+void ipvlan_link_delete(struct net_device *dev, struct list_head *head);
+void ipvlan_link_setup(struct net_device *dev);
+int ipvlan_link_register(struct rtnl_link_ops *ops);
 #endif /* __IPVLAN_H */
diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c
index 83ce74a..9af16ab 100644
--- a/drivers/net/ipvlan/ipvlan_core.c
+++ b/drivers/net/ipvlan/ipvlan_core.c
@@ -16,8 +16,8 @@ void ipvlan_init_secret(void)
 	net_get_random_once(&ipvlan_jhash_secret, sizeof(ipvlan_jhash_secret));
 }
 
-static void ipvlan_count_rx(const struct ipvl_dev *ipvlan,
-			    unsigned int len, bool success, bool mcast)
+void ipvlan_count_rx(const struct ipvl_dev *ipvlan,
+		     unsigned int len, bool success, bool mcast)
 {
 	if (!ipvlan)
 		return;
@@ -36,6 +36,7 @@ static void ipvlan_count_rx(const struct ipvl_dev *ipvlan,
 		this_cpu_inc(ipvlan->pcpu_stats->rx_errs);
 	}
 }
+EXPORT_SYMBOL_GPL(ipvlan_count_rx);
 
 static u8 ipvlan_get_v6_hash(const void *iaddr)
 {
diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index 8b0f993..ed750e2 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -494,8 +494,8 @@ static int ipvlan_nl_fillinfo(struct sk_buff *skb,
 	return ret;
 }
 
-static int ipvlan_link_new(struct net *src_net, struct net_device *dev,
-			   struct nlattr *tb[], struct nlattr *data[])
+int ipvlan_link_new(struct net *src_net, struct net_device *dev,
+		    struct nlattr *tb[], struct nlattr *data[])
 {
 	struct ipvl_dev *ipvlan = netdev_priv(dev);
 	struct ipvl_port *port;
@@ -567,8 +567,9 @@ static int ipvlan_link_new(struct net *src_net, struct net_device *dev,
 		ipvlan_port_destroy(phy_dev);
 	return err;
 }
+EXPORT_SYMBOL_GPL(ipvlan_link_new);
 
-static void ipvlan_link_delete(struct net_device *dev, struct list_head *head)
+void ipvlan_link_delete(struct net_device *dev, struct list_head *head)
 {
 	struct ipvl_dev *ipvlan = netdev_priv(dev);
 	struct ipvl_addr *addr, *next;
@@ -583,8 +584,9 @@ static void ipvlan_link_delete(struct net_device *dev, struct list_head *head)
 	unregister_netdevice_queue(dev, head);
 	netdev_upper_dev_unlink(ipvlan->phy_dev, dev);
 }
+EXPORT_SYMBOL_GPL(ipvlan_link_delete);
 
-static void ipvlan_link_setup(struct net_device *dev)
+void ipvlan_link_setup(struct net_device *dev)
 {
 	ether_setup(dev);
 
@@ -595,6 +597,7 @@ static void ipvlan_link_setup(struct net_device *dev)
 	dev->header_ops = &ipvlan_header_ops;
 	dev->ethtool_ops = &ipvlan_ethtool_ops;
 }
+EXPORT_SYMBOL_GPL(ipvlan_link_setup);
 
 static const struct nla_policy ipvlan_nl_policy[IFLA_IPVLAN_MAX + 1] =
 {
@@ -605,22 +608,22 @@ static struct rtnl_link_ops ipvlan_link_ops = {
 	.kind		= "ipvlan",
 	.priv_size	= sizeof(struct ipvl_dev),
 
-	.get_size	= ipvlan_nl_getsize,
-	.policy		= ipvlan_nl_policy,
-	.validate	= ipvlan_nl_validate,
-	.fill_info	= ipvlan_nl_fillinfo,
-	.changelink	= ipvlan_nl_changelink,
-	.maxtype	= IFLA_IPVLAN_MAX,
-
 	.setup		= ipvlan_link_setup,
 	.newlink	= ipvlan_link_new,
 	.dellink	= ipvlan_link_delete,
 };
 
-static int ipvlan_link_register(struct rtnl_link_ops *ops)
+int ipvlan_link_register(struct rtnl_link_ops *ops)
 {
+	ops->get_size	= ipvlan_nl_getsize;
+	ops->policy	= ipvlan_nl_policy;
+	ops->validate	= ipvlan_nl_validate;
+	ops->fill_info	= ipvlan_nl_fillinfo;
+	ops->changelink = ipvlan_nl_changelink;
+	ops->maxtype	= IFLA_IPVLAN_MAX;
 	return rtnl_link_register(ops);
 }
+EXPORT_SYMBOL_GPL(ipvlan_link_register);
 
 static int ipvlan_device_event(struct notifier_block *unused,
 			       unsigned long event, void *ptr)
diff --git a/drivers/net/ipvlan/ipvtap.c b/drivers/net/ipvlan/ipvtap.c
new file mode 100644
index 0000000..22e2317
--- /dev/null
+++ b/drivers/net/ipvlan/ipvtap.c
@@ -0,0 +1,241 @@
+#include <linux/etherdevice.h>
+#include "ipvlan.h"
+#include <linux/if_vlan.h>
+#include <linux/if_tap.h>
+#include <linux/interrupt.h>
+#include <linux/nsproxy.h>
+#include <linux/compat.h>
+#include <linux/if_tun.h>
+#include <linux/module.h>
+#include <linux/skbuff.h>
+#include <linux/cache.h>
+#include <linux/sched.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/wait.h>
+#include <linux/cdev.h>
+#include <linux/idr.h>
+#include <linux/fs.h>
+#include <linux/uio.h>
+
+#include <net/net_namespace.h>
+#include <net/rtnetlink.h>
+#include <net/sock.h>
+#include <linux/virtio_net.h>
+
+#define TUN_OFFLOADS (NETIF_F_HW_CSUM | NETIF_F_TSO_ECN | NETIF_F_TSO | \
+		      NETIF_F_TSO6 | NETIF_F_UFO)
+
+static dev_t ipvtap_major;
+static struct cdev ipvtap_cdev;
+
+static const void *ipvtap_net_namespace(struct device *d)
+{
+	struct net_device *dev = to_net_dev(d->parent);
+	return dev_net(dev);
+}
+
+static struct class ipvtap_class = {
+	 .name = "ipvtap",
+	 .owner = THIS_MODULE,
+	 .ns_type = &net_ns_type_operations,
+	 .namespace = ipvtap_net_namespace,
+};
+
+struct ipvtap_dev {
+	struct ipvl_dev vlan;
+	struct tap_dev	  tap;
+};
+
+static void ipvtap_count_tx_dropped(struct tap_dev *tap)
+{
+	struct ipvtap_dev *vlantap = container_of(tap, struct ipvtap_dev, tap);
+	struct ipvl_dev *vlan = &vlantap->vlan;
+
+	this_cpu_inc(vlan->pcpu_stats->tx_drps);
+}
+
+static void ipvtap_count_rx_dropped(struct tap_dev *tap)
+{
+	struct ipvtap_dev *vlantap = container_of(tap, struct ipvtap_dev, tap);
+	struct ipvl_dev *vlan = &vlantap->vlan;
+
+	ipvlan_count_rx(vlan, 0, 0, 0);
+}
+
+static void ipvtap_update_features(struct tap_dev *tap,
+				   netdev_features_t features)
+{
+	struct ipvtap_dev *vlantap = container_of(tap, struct ipvtap_dev, tap);
+	struct ipvl_dev *vlan = &vlantap->vlan;
+
+	vlan->sfeatures = features;
+	netdev_update_features(vlan->dev);
+}
+
+static int ipvtap_newlink(struct net *src_net,
+			  struct net_device *dev,
+			  struct nlattr *tb[],
+			  struct nlattr *data[])
+{
+	struct ipvtap_dev *vlantap = netdev_priv(dev);
+	int err;
+
+	INIT_LIST_HEAD(&vlantap->tap.queue_list);
+
+	/* Since macvlan supports all offloads by default, make
+	 * tap support all offloads also.
+	 */
+	vlantap->tap.tap_features = TUN_OFFLOADS;
+	vlantap->tap.count_tx_dropped = ipvtap_count_tx_dropped;
+	vlantap->tap.update_features =	ipvtap_update_features;
+	vlantap->tap.count_rx_dropped = ipvtap_count_rx_dropped;
+
+	err = netdev_rx_handler_register(dev, tap_handle_frame, &vlantap->tap);
+	if (err)
+		return err;
+
+	/* Don't put anything that may fail after macvlan_common_newlink
+	 * because we can't undo what it does.
+	 */
+	err =  ipvlan_link_new(src_net, dev, tb, data);
+	if (err) {
+		netdev_rx_handler_unregister(dev);
+		return err;
+	}
+
+	vlantap->tap.dev = vlantap->vlan.dev;
+
+	return err;
+}
+
+static void ipvtap_dellink(struct net_device *dev,
+			   struct list_head *head)
+{
+	struct ipvtap_dev *vlan = netdev_priv(dev);
+
+	netdev_rx_handler_unregister(dev);
+	tap_del_queues(&vlan->tap);
+	ipvlan_link_delete(dev, head);
+}
+
+static void ipvtap_setup(struct net_device *dev)
+{
+	ipvlan_link_setup(dev);
+	dev->tx_queue_len = TUN_READQ_SIZE;
+	dev->priv_flags &= ~IFF_NO_QUEUE;
+}
+
+static struct rtnl_link_ops ipvtap_link_ops __read_mostly = {
+	.kind		= "ipvtap",
+	.setup		= ipvtap_setup,
+	.newlink	= ipvtap_newlink,
+	.dellink	= ipvtap_dellink,
+	.priv_size	= sizeof(struct ipvtap_dev),
+};
+
+static int ipvtap_device_event(struct notifier_block *unused,
+			       unsigned long event, void *ptr)
+{
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+	struct ipvtap_dev *vlantap;
+	struct device *classdev;
+	dev_t devt;
+	int err;
+	char tap_name[IFNAMSIZ];
+
+	if (dev->rtnl_link_ops != &ipvtap_link_ops)
+		return NOTIFY_DONE;
+
+	snprintf(tap_name, IFNAMSIZ, "tap%d", dev->ifindex);
+	vlantap = netdev_priv(dev);
+
+	switch (event) {
+	case NETDEV_REGISTER:
+		/* Create the device node here after the network device has
+		 * been registered but before register_netdevice has
+		 * finished running.
+		 */
+		err = tap_get_minor(ipvtap_major, &vlantap->tap);
+		if (err)
+			return notifier_from_errno(err);
+
+		devt = MKDEV(MAJOR(ipvtap_major), vlantap->tap.minor);
+		classdev = device_create(&ipvtap_class, &dev->dev, devt,
+					 dev, tap_name);
+		if (IS_ERR(classdev)) {
+			tap_free_minor(ipvtap_major, &vlantap->tap);
+			return notifier_from_errno(PTR_ERR(classdev));
+		}
+		err = sysfs_create_link(&dev->dev.kobj, &classdev->kobj,
+					tap_name);
+		if (err)
+			return notifier_from_errno(err);
+		break;
+	case NETDEV_UNREGISTER:
+		/* vlan->minor == 0 if NETDEV_REGISTER above failed */
+		if (vlantap->tap.minor == 0)
+			break;
+		sysfs_remove_link(&dev->dev.kobj, tap_name);
+		devt = MKDEV(MAJOR(ipvtap_major), vlantap->tap.minor);
+		device_destroy(&ipvtap_class, devt);
+		tap_free_minor(ipvtap_major, &vlantap->tap);
+		break;
+	case NETDEV_CHANGE_TX_QUEUE_LEN:
+		if (tap_queue_resize(&vlantap->tap))
+			return NOTIFY_BAD;
+		break;
+	}
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block ipvtap_notifier_block __read_mostly = {
+	.notifier_call	= ipvtap_device_event,
+};
+
+static int ipvtap_init(void)
+{
+	int err;
+
+	err = tap_create_cdev(&ipvtap_cdev, &ipvtap_major, "ipvtap");
+
+	if (err)
+		goto out1;
+
+	err = class_register(&ipvtap_class);
+	if (err)
+		goto out2;
+
+	err = register_netdevice_notifier(&ipvtap_notifier_block);
+	if (err)
+		goto out3;
+
+	err = ipvlan_link_register(&ipvtap_link_ops);
+	if (err)
+		goto out4;
+
+	return 0;
+
+out4:
+	unregister_netdevice_notifier(&ipvtap_notifier_block);
+out3:
+	class_unregister(&ipvtap_class);
+out2:
+	cdev_del(&ipvtap_cdev);
+out1:
+	return err;
+}
+module_init(ipvtap_init);
+
+static void ipvtap_exit(void)
+{
+	rtnl_link_unregister(&ipvtap_link_ops);
+	unregister_netdevice_notifier(&ipvtap_notifier_block);
+	class_unregister(&ipvtap_class);
+	tap_destroy_cdev(ipvtap_major, &ipvtap_cdev);
+}
+module_exit(ipvtap_exit);
+MODULE_ALIAS_RTNL_LINK("ipvtap");
+MODULE_AUTHOR("Sainath Grandhi <sainath.grandhi@intel.com>");
+MODULE_LICENSE("GPL");
-- 
2.7.4

[toc] | [prev] | [next] | [standalone]

#1566967 — [PATCHv3 5/7] TAP: Extending tap device create/destroy APIs

Extending tap APIs get/free_minor and create/destroy_cdev to handle more than one
type of virtual interface.

Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
---
 drivers/net/macvtap_main.c |  6 +--
 drivers/net/tap.c          | 98 +++++++++++++++++++++++++++++++++++-----------
 include/linux/if_tap.h     |  4 +-
 3 files changed, 80 insertions(+), 28 deletions(-)

diff --git a/drivers/net/macvtap_main.c b/drivers/net/macvtap_main.c
index dd6f4e4..50fe993 100644
--- a/drivers/net/macvtap_main.c
+++ b/drivers/net/macvtap_main.c
@@ -163,7 +163,7 @@ static int macvtap_device_event(struct notifier_block *unused,
 		 * been registered but before register_netdevice has
 		 * finished running.
 		 */
-		err = tap_get_minor(&vlantap->tap);
+		err = tap_get_minor(macvtap_major, &vlantap->tap);
 		if (err)
 			return notifier_from_errno(err);
 
@@ -171,7 +171,7 @@ static int macvtap_device_event(struct notifier_block *unused,
 		classdev = device_create(&macvtap_class, &dev->dev, devt,
 					 dev, tap_name);
 		if (IS_ERR(classdev)) {
-			tap_free_minor(&vlantap->tap);
+			tap_free_minor(macvtap_major, &vlantap->tap);
 			return notifier_from_errno(PTR_ERR(classdev));
 		}
 		err = sysfs_create_link(&dev->dev.kobj, &classdev->kobj,
@@ -186,7 +186,7 @@ static int macvtap_device_event(struct notifier_block *unused,
 		sysfs_remove_link(&dev->dev.kobj, tap_name);
 		devt = MKDEV(MAJOR(macvtap_major), vlantap->tap.minor);
 		device_destroy(&macvtap_class, devt);
-		tap_free_minor(&vlantap->tap);
+		tap_free_minor(macvtap_major, &vlantap->tap);
 		break;
 	case NETDEV_CHANGE_TX_QUEUE_LEN:
 		if (tap_queue_resize(&vlantap->tap))
diff --git a/drivers/net/tap.c b/drivers/net/tap.c
index ede436a..1219ee9 100644
--- a/drivers/net/tap.c
+++ b/drivers/net/tap.c
@@ -99,12 +99,16 @@ static struct proto tap_proto = {
 };
 
 #define TAP_NUM_DEVS (1U << MINORBITS)
+
+static LIST_HEAD(major_list);
+
 struct major_info {
 	dev_t major;
 	struct idr minor_idr;
 	struct mutex minor_lock;
 	const char *device_name;
-} macvtap_major;
+	struct list_head next;
+};
 
 #define GOODCOPY_LEN 128
 
@@ -385,44 +389,73 @@ rx_handler_result_t tap_handle_frame(struct sk_buff **pskb)
 	return RX_HANDLER_CONSUMED;
 }
 
-int tap_get_minor(struct tap_dev *tap)
+static struct major_info *tap_get_major(int major)
+{
+	struct major_info *tap_major, *tmp;
+
+	list_for_each_entry_safe(tap_major, tmp, &major_list, next) {
+		if (tap_major->major == major) {
+			return tap_major;
+		}
+	}
+
+	return NULL;
+}
+
+int tap_get_minor(dev_t major, struct tap_dev *tap)
 {
 	int retval = -ENOMEM;
+	struct major_info *tap_major;
+
+	tap_major = tap_get_major(MAJOR(major));
+	if (!tap_major)
+		return -EINVAL;
 
-	mutex_lock(&macvtap_major.minor_lock);
-	retval = idr_alloc(&macvtap_major.minor_idr, tap, 1, TAP_NUM_DEVS, GFP_KERNEL);
+	mutex_lock(&tap_major->minor_lock);
+	retval = idr_alloc(&tap_major->minor_idr, tap, 1, TAP_NUM_DEVS, GFP_KERNEL);
 	if (retval >= 0) {
 		tap->minor = retval;
 	} else if (retval == -ENOSPC) {
 		netdev_err(tap->dev, "Too many tap devices\n");
 		retval = -EINVAL;
 	}
-	mutex_unlock(&macvtap_major.minor_lock);
+	mutex_unlock(&tap_major->minor_lock);
 	return retval < 0 ? retval : 0;
 }
 
-void tap_free_minor(struct tap_dev *tap)
+void tap_free_minor(dev_t major, struct tap_dev *tap)
 {
-	mutex_lock(&macvtap_major.minor_lock);
+	struct major_info *tap_major;
+
+	tap_major = tap_get_major(MAJOR(major));
+	if (!tap_major)
+		return;
+
+	mutex_lock(&tap_major->minor_lock);
 	if (tap->minor) {
-		idr_remove(&macvtap_major.minor_idr, tap->minor);
+		idr_remove(&tap_major->minor_idr, tap->minor);
 		tap->minor = 0;
 	}
-	mutex_unlock(&macvtap_major.minor_lock);
+	mutex_unlock(&tap_major->minor_lock);
 }
 
-static struct tap_dev *dev_get_by_tap_minor(int minor)
+static struct tap_dev *dev_get_by_tap_file(int major, int minor)
 {
 	struct net_device *dev = NULL;
 	struct tap_dev *tap;
+	struct major_info *tap_major;
+
+	tap_major = tap_get_major(major);
+	if (!tap_major)
+		return NULL;
 
-	mutex_lock(&macvtap_major.minor_lock);
-	tap = idr_find(&macvtap_major.minor_idr, minor);
+	mutex_lock(&tap_major->minor_lock);
+	tap = idr_find(&tap_major->minor_idr, minor);
 	if (tap) {
 		dev = tap->dev;
 		dev_hold(dev);
 	}
-	mutex_unlock(&macvtap_major.minor_lock);
+	mutex_unlock(&tap_major->minor_lock);
 	return tap;
 }
 
@@ -454,7 +487,7 @@ static int tap_open(struct inode *inode, struct file *file)
 	int err = -ENODEV;
 
 	rtnl_lock();
-	tap = dev_get_by_tap_minor(iminor(inode));
+	tap = dev_get_by_tap_file(imajor(inode), iminor(inode));
 	if (!tap)
 		goto err;
 
@@ -1161,6 +1194,26 @@ int tap_queue_resize(struct tap_dev *tap)
 	return ret;
 }
 
+static int tap_list_add(dev_t major, const char *device_name)
+{
+	struct major_info *tap_major;
+	int err = 0;
+
+	tap_major = kzalloc(sizeof(*tap_major), GFP_ATOMIC);
+	if (!tap_major)
+		return -ENOMEM;
+
+	tap_major->major = MAJOR(major);
+
+	idr_init(&tap_major->minor_idr);
+	mutex_init(&tap_major->minor_lock);
+
+	tap_major->device_name = device_name;
+
+	list_add_tail(&tap_major->next, &major_list);
+	return err;
+}
+
 int tap_create_cdev(struct cdev *tap_cdev,
 		    dev_t *tap_major, const char *device_name)
 {
@@ -1175,14 +1228,7 @@ int tap_create_cdev(struct cdev *tap_cdev,
 	if (err)
 		goto out2;
 
-	macvtap_major.major = MAJOR(*tap_major);
-
-	idr_init(&macvtap_major.minor_idr);
-	mutex_init(&macvtap_major.minor_lock);
-
-	macvtap_major.device_name = device_name;
-
-	return err;
+	return tap_list_add(*tap_major, device_name);
 
 out2:
 	unregister_chrdev_region(*tap_major, TAP_NUM_DEVS);
@@ -1192,7 +1238,13 @@ int tap_create_cdev(struct cdev *tap_cdev,
 
 void tap_destroy_cdev(dev_t major, struct cdev *tap_cdev)
 {
+	struct major_info *tap_major;
+
+	tap_major = tap_get_major(MAJOR(major));
+	if (!tap_major)
+		return;
+
 	cdev_del(tap_cdev);
 	unregister_chrdev_region(major, TAP_NUM_DEVS);
-	idr_destroy(&macvtap_major.minor_idr);
+	idr_destroy(&tap_major->minor_idr);
 }
diff --git a/include/linux/if_tap.h b/include/linux/if_tap.h
index 75031e5..362e71c 100644
--- a/include/linux/if_tap.h
+++ b/include/linux/if_tap.h
@@ -65,8 +65,8 @@ struct tap_queue {
 
 rx_handler_result_t tap_handle_frame(struct sk_buff **pskb);
 void tap_del_queues(struct tap_dev *tap);
-int tap_get_minor(struct tap_dev *tap);
-void tap_free_minor(struct tap_dev *tap);
+int tap_get_minor(dev_t major, struct tap_dev *tap);
+void tap_free_minor(dev_t major, struct tap_dev *tap);
 int tap_queue_resize(struct tap_dev *tap);
 int tap_create_cdev(struct cdev *tap_cdev,
 		    dev_t *tap_major, const char *device_name);
-- 
2.7.4

[toc] | [prev] | [next] | [standalone]

#1566972 — [PATCHv3 4/7] TAP: Abstract type of virtual interface from tap implementation

macvlan object is re-structured to hold tap related elements in a separate
entity, tap_dev. Upon NETDEV_REGISTER device_event, tap_dev is registered with
idr and fetched again on tap_open. Few of the tap functions are modified to
accepted tap_dev as argument. tap_dev object includes callbacks to be used by
underlying virtual interface to take care of tx and rx accounting.

Signed-off-by: Sainath Grandhi <sainath.grandhi@intel.com>
---
 drivers/net/macvlan.c      |   2 +-
 drivers/net/macvtap_main.c |  71 +++++++++---
 drivers/net/tap.c          | 264 ++++++++++++++++++++-------------------------
 include/linux/if_tap.h     |  57 +++++++++-
 4 files changed, 229 insertions(+), 165 deletions(-)

diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 20b3fdf2..79383f9 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -1526,7 +1526,6 @@ static const struct nla_policy macvlan_policy[IFLA_MACVLAN_MAX + 1] = {
 int macvlan_link_register(struct rtnl_link_ops *ops)
 {
 	/* common fields */
-	ops->priv_size		= sizeof(struct macvlan_dev);
 	ops->validate		= macvlan_validate;
 	ops->maxtype		= IFLA_MACVLAN_MAX;
 	ops->policy		= macvlan_policy;
@@ -1549,6 +1548,7 @@ static struct rtnl_link_ops macvlan_link_ops = {
 	.newlink	= macvlan_newlink,
 	.dellink	= macvlan_dellink,
 	.get_link_net	= macvlan_get_link_net,
+	.priv_size      = sizeof(struct macvlan_dev),
 };
 
 static int macvlan_device_event(struct notifier_block *unused,
diff --git a/drivers/net/macvtap_main.c b/drivers/net/macvtap_main.c
index 694e385..dd6f4e4 100644
--- a/drivers/net/macvtap_main.c
+++ b/drivers/net/macvtap_main.c
@@ -24,6 +24,11 @@
 #include <linux/virtio_net.h>
 #include <linux/skb_array.h>
 
+struct macvtap_dev {
+	struct macvlan_dev vlan;
+	struct tap_dev    tap;
+};
+
 /*
  * Variables for dealing with macvtaps device numbers.
  */
@@ -46,22 +51,55 @@ static struct cdev macvtap_cdev;
 #define TUN_OFFLOADS (NETIF_F_HW_CSUM | NETIF_F_TSO_ECN | NETIF_F_TSO | \
 		      NETIF_F_TSO6 | NETIF_F_UFO)
 
+static void macvtap_count_tx_dropped(struct tap_dev *tap)
+{
+	struct macvtap_dev *vlantap = container_of(tap, struct macvtap_dev, tap);
+	struct macvlan_dev *vlan = &vlantap->vlan;
+
+	this_cpu_inc(vlan->pcpu_stats->tx_dropped);
+}
+
+static void macvtap_count_rx_dropped(struct tap_dev *tap)
+{
+	struct macvtap_dev *vlantap = container_of(tap, struct macvtap_dev, tap);
+	struct macvlan_dev *vlan = &vlantap->vlan;
+
+	macvlan_count_rx(vlan, 0, 0, 0);
+}
+
+static void macvtap_update_features(struct tap_dev *tap,
+				    netdev_features_t features)
+{
+	struct macvtap_dev *vlantap = container_of(tap, struct macvtap_dev, tap);
+	struct macvlan_dev *vlan = &vlantap->vlan;
+
+	vlan->set_features = features;
+	netdev_update_features(vlan->dev);
+}
+
 static int macvtap_newlink(struct net *src_net,
 			   struct net_device *dev,
 			   struct nlattr *tb[],
 			   struct nlattr *data[])
 {
-	struct macvlan_dev *vlan = netdev_priv(dev);
+	struct macvtap_dev *vlantap = netdev_priv(dev);
 	int err;
 
-	INIT_LIST_HEAD(&vlan->queue_list);
+	INIT_LIST_HEAD(&vlantap->tap.queue_list);
 
 	/* Since macvlan supports all offloads by default, make
 	 * tap support all offloads also.
 	 */
-	vlan->tap_features = TUN_OFFLOADS;
+	vlantap->tap.tap_features = TUN_OFFLOADS;
 
-	err = netdev_rx_handler_register(dev, tap_handle_frame, vlan);
+	/* Register callbacks for rx/tx drops accounting and updating
+	 * net_device features
+	 */
+	vlantap->tap.count_tx_dropped = macvtap_count_tx_dropped;
+	vlantap->tap.count_rx_dropped = macvtap_count_rx_dropped;
+	vlantap->tap.update_features  = macvtap_update_features;
+
+	err = netdev_rx_handler_register(dev, tap_handle_frame, &vlantap->tap);
 	if (err)
 		return err;
 
@@ -74,14 +112,18 @@ static int macvtap_newlink(struct net *src_net,
 		return err;
 	}
 
+	vlantap->tap.dev = vlantap->vlan.dev;
+
 	return 0;
 }
 
 static void macvtap_dellink(struct net_device *dev,
 			    struct list_head *head)
 {
+	struct macvtap_dev *vlantap = netdev_priv(dev);
+
 	netdev_rx_handler_unregister(dev);
-	tap_del_queues(dev);
+	tap_del_queues(&vlantap->tap);
 	macvlan_dellink(dev, head);
 }
 
@@ -96,13 +138,14 @@ static struct rtnl_link_ops macvtap_link_ops __read_mostly = {
 	.setup		= macvtap_setup,
 	.newlink	= macvtap_newlink,
 	.dellink	= macvtap_dellink,
+	.priv_size      = sizeof(struct macvtap_dev),
 };
 
 static int macvtap_device_event(struct notifier_block *unused,
 				unsigned long event, void *ptr)
 {
 	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
-	struct macvlan_dev *vlan;
+	struct macvtap_dev *vlantap;
 	struct device *classdev;
 	dev_t devt;
 	int err;
@@ -112,7 +155,7 @@ static int macvtap_device_event(struct notifier_block *unused,
 		return NOTIFY_DONE;
 
 	snprintf(tap_name, IFNAMSIZ, "tap%d", dev->ifindex);
-	vlan = netdev_priv(dev);
+	vlantap = netdev_priv(dev);
 
 	switch (event) {
 	case NETDEV_REGISTER:
@@ -120,15 +163,15 @@ static int macvtap_device_event(struct notifier_block *unused,
 		 * been registered but before register_netdevice has
 		 * finished running.
 		 */
-		err = tap_get_minor(vlan);
+		err = tap_get_minor(&vlantap->tap);
 		if (err)
 			return notifier_from_errno(err);
 
-		devt = MKDEV(MAJOR(macvtap_major), vlan->minor);
+		devt = MKDEV(MAJOR(macvtap_major), vlantap->tap.minor);
 		classdev = device_create(&macvtap_class, &dev->dev, devt,
 					 dev, tap_name);
 		if (IS_ERR(classdev)) {
-			tap_free_minor(vlan);
+			tap_free_minor(&vlantap->tap);
 			return notifier_from_errno(PTR_ERR(classdev));
 		}
 		err = sysfs_create_link(&dev->dev.kobj, &classdev->kobj,
@@ -138,15 +181,15 @@ static int macvtap_device_event(struct notifier_block *unused,
 		break;
 	case NETDEV_UNREGISTER:
 		/* vlan->minor == 0 if NETDEV_REGISTER above failed */
-		if (vlan->minor == 0)
+		if (vlantap->tap.minor == 0)
 			break;
 		sysfs_remove_link(&dev->dev.kobj, tap_name);
-		devt = MKDEV(MAJOR(macvtap_major), vlan->minor);
+		devt = MKDEV(MAJOR(macvtap_major), vlantap->tap.minor);
 		device_destroy(&macvtap_class, devt);
-		tap_free_minor(vlan);
+		tap_free_minor(&vlantap->tap);
 		break;
 	case NETDEV_CHANGE_TX_QUEUE_LEN:
-		if (tap_queue_resize(vlan))
+		if (tap_queue_resize(&vlantap->tap))
 			return NOTIFY_BAD;
 		break;
 	}
diff --git a/drivers/net/tap.c b/drivers/net/tap.c
index ec7ebed..ede436a 100644
--- a/drivers/net/tap.c
+++ b/drivers/net/tap.c
@@ -1,5 +1,5 @@
 #include <linux/etherdevice.h>
-#include <linux/if_macvlan.h>
+#include <linux/if_tap.h>
 #include <linux/if_vlan.h>
 #include <linux/interrupt.h>
 #include <linux/nsproxy.h>
@@ -23,30 +23,6 @@
 #include <linux/virtio_net.h>
 #include <linux/skb_array.h>
 
-/*
- * A tap queue is the central object of this driver, it connects
- * an open character device to a macvlan interface. There can be
- * multiple queues on one interface, which map back to queues
- * implemented in hardware on the underlying device.
- *
- * tap_proto is used to allocate queues through the sock allocation
- * mechanism.
- *
- */
-struct tap_queue {
-	struct sock sk;
-	struct socket sock;
-	struct socket_wq wq;
-	int vnet_hdr_sz;
-	struct macvlan_dev __rcu *vlan;
-	struct file *file;
-	unsigned int flags;
-	u16 queue_index;
-	bool enabled;
-	struct list_head next;
-	struct skb_array skb_array;
-};
-
 #define TAP_IFFEATURES (IFF_VNET_HDR | IFF_MULTI_QUEUE)
 
 #define TAP_VNET_LE 0x80000000
@@ -137,7 +113,7 @@ static const struct proto_ops tap_socket_ops;
 #define RX_OFFLOADS (NETIF_F_GRO | NETIF_F_LRO)
 #define TAP_FEATURES (NETIF_F_GSO | NETIF_F_SG | NETIF_F_FRAGLIST)
 
-static struct macvlan_dev *tap_get_vlan_rcu(const struct net_device *dev)
+static struct tap_dev *tap_dev_get_rcu(const struct net_device *dev)
 {
 	return rcu_dereference(dev->rx_handler_data);
 }
@@ -159,10 +135,9 @@ static struct macvlan_dev *tap_get_vlan_rcu(const struct net_device *dev)
  * when both our references and any pending SKBs are gone.
  */
 
-static int tap_enable_queue(struct net_device *dev, struct file *file,
+static int tap_enable_queue(struct tap_dev *tap, struct file *file,
 			    struct tap_queue *q)
 {
-	struct macvlan_dev *vlan = netdev_priv(dev);
 	int err = -EINVAL;
 
 	ASSERT_RTNL();
@@ -171,62 +146,60 @@ static int tap_enable_queue(struct net_device *dev, struct file *file,
 		goto out;
 
 	err = 0;
-	rcu_assign_pointer(vlan->taps[vlan->numvtaps], q);
-	q->queue_index = vlan->numvtaps;
+	rcu_assign_pointer(tap->taps[tap->numvtaps], q);
+	q->queue_index = tap->numvtaps;
 	q->enabled = true;
 
-	vlan->numvtaps++;
+	tap->numvtaps++;
 out:
 	return err;
 }
 
 /* Requires RTNL */
-static int tap_set_queue(struct net_device *dev, struct file *file,
+static int tap_set_queue(struct tap_dev *tap, struct file *file,
 			 struct tap_queue *q)
 {
-	struct macvlan_dev *vlan = netdev_priv(dev);
-
-	if (vlan->numqueues == MAX_TAP_QUEUES)
+	if (tap->numqueues == MAX_TAP_QUEUES)
 		return -EBUSY;
 
-	rcu_assign_pointer(q->vlan, vlan);
-	rcu_assign_pointer(vlan->taps[vlan->numvtaps], q);
+	rcu_assign_pointer(q->tap, tap);
+	rcu_assign_pointer(tap->taps[tap->numvtaps], q);
 	sock_hold(&q->sk);
 
 	q->file = file;
-	q->queue_index = vlan->numvtaps;
+	q->queue_index = tap->numvtaps;
 	q->enabled = true;
 	file->private_data = q;
-	list_add_tail(&q->next, &vlan->queue_list);
+	list_add_tail(&q->next, &tap->queue_list);
 
-	vlan->numvtaps++;
-	vlan->numqueues++;
+	tap->numvtaps++;
+	tap->numqueues++;
 
 	return 0;
 }
 
 static int tap_disable_queue(struct tap_queue *q)
 {
-	struct macvlan_dev *vlan;
+	struct tap_dev *tap;
 	struct tap_queue *nq;
 
 	ASSERT_RTNL();
 	if (!q->enabled)
 		return -EINVAL;
 
-	vlan = rtnl_dereference(q->vlan);
+	tap = rtnl_dereference(q->tap);
 
-	if (vlan) {
+	if (tap) {
 		int index = q->queue_index;
-		BUG_ON(index >= vlan->numvtaps);
-		nq = rtnl_dereference(vlan->taps[vlan->numvtaps - 1]);
+		BUG_ON(index >= tap->numvtaps);
+		nq = rtnl_dereference(tap->taps[tap->numvtaps - 1]);
 		nq->queue_index = index;
 
-		rcu_assign_pointer(vlan->taps[index], nq);
-		RCU_INIT_POINTER(vlan->taps[vlan->numvtaps - 1], NULL);
+		rcu_assign_pointer(tap->taps[index], nq);
+		RCU_INIT_POINTER(tap->taps[tap->numvtaps - 1], NULL);
 		q->enabled = false;
 
-		vlan->numvtaps--;
+		tap->numvtaps--;
 	}
 
 	return 0;
@@ -242,17 +215,17 @@ static int tap_disable_queue(struct tap_queue *q)
  */
 static void tap_put_queue(struct tap_queue *q)
 {
-	struct macvlan_dev *vlan;
+	struct tap_dev *tap;
 
 	rtnl_lock();
-	vlan = rtnl_dereference(q->vlan);
+	tap = rtnl_dereference(q->tap);
 
-	if (vlan) {
+	if (tap) {
 		if (q->enabled)
 			BUG_ON(tap_disable_queue(q));
 
-		vlan->numqueues--;
-		RCU_INIT_POINTER(q->vlan, NULL);
+		tap->numqueues--;
+		RCU_INIT_POINTER(q->tap, NULL);
 		sock_put(&q->sk);
 		list_del_init(&q->next);
 	}
@@ -270,17 +243,16 @@ static void tap_put_queue(struct tap_queue *q)
  * Cache vlan->numvtaps since it can become zero during the execution
  * of this function.
  */
-static struct tap_queue *tap_get_queue(struct net_device *dev,
+static struct tap_queue *tap_get_queue(struct tap_dev *tap,
 				       struct sk_buff *skb)
 {
-	struct macvlan_dev *vlan = netdev_priv(dev);
-	struct tap_queue *tap = NULL;
+	struct tap_queue *queue = NULL;
 	/* Access to taps array is protected by rcu, but access to numvtaps
 	 * isn't. Below we use it to lookup a queue, but treat it as a hint
 	 * and validate that the result isn't NULL - in case we are
 	 * racing against queue removal.
 	 */
-	int numvtaps = ACCESS_ONCE(vlan->numvtaps);
+	int numvtaps = ACCESS_ONCE(tap->numvtaps);
 	__u32 rxq;
 
 	if (!numvtaps)
@@ -292,7 +264,7 @@ static struct tap_queue *tap_get_queue(struct net_device *dev,
 	/* Check if we can use flow to select a queue */
 	rxq = skb_get_hash(skb);
 	if (rxq) {
-		tap = rcu_dereference(vlan->taps[rxq % numvtaps]);
+		queue = rcu_dereference(tap->taps[rxq % numvtaps]);
 		goto out;
 	}
 
@@ -302,14 +274,14 @@ static struct tap_queue *tap_get_queue(struct net_device *dev,
 		while (unlikely(rxq >= numvtaps))
 			rxq -= numvtaps;
 
-		tap = rcu_dereference(vlan->taps[rxq]);
+		queue = rcu_dereference(tap->taps[rxq]);
 		goto out;
 	}
 
 single:
-	tap = rcu_dereference(vlan->taps[0]);
+	queue = rcu_dereference(tap->taps[0]);
 out:
-	return tap;
+	return queue;
 }
 
 /*
@@ -317,39 +289,38 @@ static struct tap_queue *tap_get_queue(struct net_device *dev,
  * that it holds on all queues and safely set the pointer
  * from the queues to NULL.
  */
-void tap_del_queues(struct net_device *dev)
+void tap_del_queues(struct tap_dev *tap)
 {
-	struct macvlan_dev *vlan = netdev_priv(dev);
 	struct tap_queue *q, *tmp;
 
 	ASSERT_RTNL();
-	list_for_each_entry_safe(q, tmp, &vlan->queue_list, next) {
+	list_for_each_entry_safe(q, tmp, &tap->queue_list, next) {
 		list_del_init(&q->next);
-		RCU_INIT_POINTER(q->vlan, NULL);
+		RCU_INIT_POINTER(q->tap, NULL);
 		if (q->enabled)
-			vlan->numvtaps--;
-		vlan->numqueues--;
+			tap->numvtaps--;
+		tap->numqueues--;
 		sock_put(&q->sk);
 	}
-	BUG_ON(vlan->numvtaps);
-	BUG_ON(vlan->numqueues);
+	BUG_ON(tap->numvtaps);
+	BUG_ON(tap->numqueues);
 	/* guarantee that any future tap_set_queue will fail */
-	vlan->numvtaps = MAX_TAP_QUEUES;
+	tap->numvtaps = MAX_TAP_QUEUES;
 }
 
 rx_handler_result_t tap_handle_frame(struct sk_buff **pskb)
 {
 	struct sk_buff *skb = *pskb;
 	struct net_device *dev = skb->dev;
-	struct macvlan_dev *vlan;
+	struct tap_dev *tap;
 	struct tap_queue *q;
 	netdev_features_t features = TAP_FEATURES;
 
-	vlan = tap_get_vlan_rcu(dev);
-	if (!vlan)
+	tap = tap_dev_get_rcu(dev);
+	if (!tap)
 		return RX_HANDLER_PASS;
 
-	q = tap_get_queue(dev, skb);
+	q = tap_get_queue(tap, skb);
 	if (!q)
 		return RX_HANDLER_PASS;
 
@@ -363,7 +334,7 @@ rx_handler_result_t tap_handle_frame(struct sk_buff **pskb)
 	 * enabled.
 	 */
 	if (q->flags & IFF_VNET_HDR)
-		features |= vlan->tap_features;
+		features |= tap->tap_features;
 	if (netif_needs_gso(skb, features)) {
 		struct sk_buff *segs = __skb_gso_segment(skb, features, false);
 
@@ -408,50 +379,51 @@ rx_handler_result_t tap_handle_frame(struct sk_buff **pskb)
 
 drop:
 	/* Count errors/drops only here, thus don't care about args. */
-	macvlan_count_rx(vlan, 0, 0, 0);
+	if (tap->count_rx_dropped)
+		tap->count_rx_dropped(tap);
 	kfree_skb(skb);
 	return RX_HANDLER_CONSUMED;
 }
 
-int tap_get_minor(struct macvlan_dev *vlan)
+int tap_get_minor(struct tap_dev *tap)
 {
 	int retval = -ENOMEM;
 
 	mutex_lock(&macvtap_major.minor_lock);
-	retval = idr_alloc(&macvtap_major.minor_idr, vlan, 1, TAP_NUM_DEVS, GFP_KERNEL);
+	retval = idr_alloc(&macvtap_major.minor_idr, tap, 1, TAP_NUM_DEVS, GFP_KERNEL);
 	if (retval >= 0) {
-		vlan->minor = retval;
+		tap->minor = retval;
 	} else if (retval == -ENOSPC) {
-		netdev_err(vlan->dev, "Too many tap devices\n");
+		netdev_err(tap->dev, "Too many tap devices\n");
 		retval = -EINVAL;
 	}
 	mutex_unlock(&macvtap_major.minor_lock);
 	return retval < 0 ? retval : 0;
 }
 
-void tap_free_minor(struct macvlan_dev *vlan)
+void tap_free_minor(struct tap_dev *tap)
 {
 	mutex_lock(&macvtap_major.minor_lock);
-	if (vlan->minor) {
-		idr_remove(&macvtap_major.minor_idr, vlan->minor);
-		vlan->minor = 0;
+	if (tap->minor) {
+		idr_remove(&macvtap_major.minor_idr, tap->minor);
+		tap->minor = 0;
 	}
 	mutex_unlock(&macvtap_major.minor_lock);
 }
 
-static struct net_device *dev_get_by_tap_minor(int minor)
+static struct tap_dev *dev_get_by_tap_minor(int minor)
 {
 	struct net_device *dev = NULL;
-	struct macvlan_dev *vlan;
+	struct tap_dev *tap;
 
 	mutex_lock(&macvtap_major.minor_lock);
-	vlan = idr_find(&macvtap_major.minor_idr, minor);
-	if (vlan) {
-		dev = vlan->dev;
+	tap = idr_find(&macvtap_major.minor_idr, minor);
+	if (tap) {
+		dev = tap->dev;
 		dev_hold(dev);
 	}
 	mutex_unlock(&macvtap_major.minor_lock);
-	return dev;
+	return tap;
 }
 
 static void tap_sock_write_space(struct sock *sk)
@@ -477,13 +449,13 @@ static void tap_sock_destruct(struct sock *sk)
 static int tap_open(struct inode *inode, struct file *file)
 {
 	struct net *net = current->nsproxy->net_ns;
-	struct net_device *dev;
+	struct tap_dev *tap;
 	struct tap_queue *q;
 	int err = -ENODEV;
 
 	rtnl_lock();
-	dev = dev_get_by_tap_minor(iminor(inode));
-	if (!dev)
+	tap = dev_get_by_tap_minor(iminor(inode));
+	if (!tap)
 		goto err;
 
 	err = -ENOMEM;
@@ -511,18 +483,18 @@ static int tap_open(struct inode *inode, struct file *file)
 	 * The macvlan supports zerocopy iff the lower device supports zero
 	 * copy so we don't have to look at the lower device directly.
 	 */
-	if ((dev->features & NETIF_F_HIGHDMA) && (dev->features & NETIF_F_SG))
+	if ((tap->dev->features & NETIF_F_HIGHDMA) && (tap->dev->features & NETIF_F_SG))
 		sock_set_flag(&q->sk, SOCK_ZEROCOPY);
 
 	err = -ENOMEM;
-	if (skb_array_init(&q->skb_array, dev->tx_queue_len, GFP_KERNEL))
+	if (skb_array_init(&q->skb_array, tap->dev->tx_queue_len, GFP_KERNEL))
 		goto err_array;
 
-	err = tap_set_queue(dev, file, q);
+	err = tap_set_queue(tap, file, q);
 	if (err)
 		goto err_queue;
 
-	dev_put(dev);
+	dev_put(tap->dev);
 
 	rtnl_unlock();
 	return err;
@@ -532,8 +504,8 @@ static int tap_open(struct inode *inode, struct file *file)
 err_array:
 	sock_put(&q->sk);
 err:
-	if (dev)
-		dev_put(dev);
+	if (tap)
+		dev_put(tap->dev);
 
 	rtnl_unlock();
 	return err;
@@ -601,7 +573,7 @@ static ssize_t tap_get_user(struct tap_queue *q, struct msghdr *m,
 {
 	int good_linear = SKB_MAX_HEAD(TAP_RESERVE);
 	struct sk_buff *skb;
-	struct macvlan_dev *vlan;
+	struct tap_dev *tap;
 	unsigned long total_len = iov_iter_count(from);
 	unsigned long len = total_len;
 	int err;
@@ -698,7 +670,7 @@ static ssize_t tap_get_user(struct tap_queue *q, struct msghdr *m,
 		skb_set_network_header(skb, depth);
 
 	rcu_read_lock();
-	vlan = rcu_dereference(q->vlan);
+	tap = rcu_dereference(q->tap);
 	/* copy skb_ubuf_info for callback when skb has no error */
 	if (zerocopy) {
 		skb_shinfo(skb)->destructor_arg = m->msg_control;
@@ -709,8 +681,8 @@ static ssize_t tap_get_user(struct tap_queue *q, struct msghdr *m,
 		uarg->callback(uarg, false);
 	}
 
-	if (vlan) {
-		skb->dev = vlan->dev;
+	if (tap) {
+		skb->dev = tap->dev;
 		dev_queue_xmit(skb);
 	} else {
 		kfree_skb(skb);
@@ -724,9 +696,9 @@ static ssize_t tap_get_user(struct tap_queue *q, struct msghdr *m,
 
 err:
 	rcu_read_lock();
-	vlan = rcu_dereference(q->vlan);
-	if (vlan)
-		this_cpu_inc(vlan->pcpu_stats->tx_dropped);
+	tap = rcu_dereference(q->tap);
+	if (tap && tap->count_tx_dropped)
+		tap->count_tx_dropped(tap);
 	rcu_read_unlock();
 
 	return err;
@@ -853,55 +825,55 @@ static ssize_t tap_read_iter(struct kiocb *iocb, struct iov_iter *to)
 	return ret;
 }
 
-static struct macvlan_dev *tap_get_vlan(struct tap_queue *q)
+static struct tap_dev *tap_get_tap_dev(struct tap_queue *q)
 {
-	struct macvlan_dev *vlan;
+	struct tap_dev *tap;
 
 	ASSERT_RTNL();
-	vlan = rtnl_dereference(q->vlan);
-	if (vlan)
-		dev_hold(vlan->dev);
+	tap = rtnl_dereference(q->tap);
+	if (tap)
+		dev_hold(tap->dev);
 
-	return vlan;
+	return tap;
 }
 
-static void tap_put_vlan(struct macvlan_dev *vlan)
+static void tap_put_tap_dev(struct tap_dev *tap)
 {
-	dev_put(vlan->dev);
+	dev_put(tap->dev);
 }
 
 static int tap_ioctl_set_queue(struct file *file, unsigned int flags)
 {
 	struct tap_queue *q = file->private_data;
-	struct macvlan_dev *vlan;
+	struct tap_dev *tap;
 	int ret;
 
-	vlan = tap_get_vlan(q);
-	if (!vlan)
+	tap = tap_get_tap_dev(q);
+	if (!tap)
 		return -EINVAL;
 
 	if (flags & IFF_ATTACH_QUEUE)
-		ret = tap_enable_queue(vlan->dev, file, q);
+		ret = tap_enable_queue(tap, file, q);
 	else if (flags & IFF_DETACH_QUEUE)
 		ret = tap_disable_queue(q);
 	else
 		ret = -EINVAL;
 
-	tap_put_vlan(vlan);
+	tap_put_tap_dev(tap);
 	return ret;
 }
 
 static int set_offload(struct tap_queue *q, unsigned long arg)
 {
-	struct macvlan_dev *vlan;
+	struct tap_dev *tap;
 	netdev_features_t features;
 	netdev_features_t feature_mask = 0;
 
-	vlan = rtnl_dereference(q->vlan);
-	if (!vlan)
+	tap = rtnl_dereference(q->tap);
+	if (!tap)
 		return -ENOLINK;
 
-	features = vlan->dev->features;
+	features = tap->dev->features;
 
 	if (arg & TUN_F_CSUM) {
 		feature_mask = NETIF_F_HW_CSUM;
@@ -935,9 +907,9 @@ static int set_offload(struct tap_queue *q, unsigned long arg)
 	/* tap_features are the same as features on tun/tap and
 	 * reflect user expectations.
 	 */
-	vlan->tap_features = feature_mask;
-	vlan->set_features = features;
-	netdev_update_features(vlan->dev);
+	tap->tap_features = feature_mask;
+	if (tap->update_features)
+		tap->update_features(tap, features);
 
 	return 0;
 }
@@ -949,7 +921,7 @@ static long tap_ioctl(struct file *file, unsigned int cmd,
 		      unsigned long arg)
 {
 	struct tap_queue *q = file->private_data;
-	struct macvlan_dev *vlan;
+	struct tap_dev *tap;
 	void __user *argp = (void __user *)arg;
 	struct ifreq __user *ifr = argp;
 	unsigned int __user *up = argp;
@@ -975,18 +947,18 @@ static long tap_ioctl(struct file *file, unsigned int cmd,
 
 	case TUNGETIFF:
 		rtnl_lock();
-		vlan = tap_get_vlan(q);
-		if (!vlan) {
+		tap = tap_get_tap_dev(q);
+		if (!tap) {
 			rtnl_unlock();
 			return -ENOLINK;
 		}
 
 		ret = 0;
 		u = q->flags;
-		if (copy_to_user(&ifr->ifr_name, vlan->dev->name, IFNAMSIZ) ||
+		if (copy_to_user(&ifr->ifr_name, tap->dev->name, IFNAMSIZ) ||
 		    put_user(u, &ifr->ifr_flags))
 			ret = -EFAULT;
-		tap_put_vlan(vlan);
+		tap_put_tap_dev(tap);
 		rtnl_unlock();
 		return ret;
 
@@ -1059,18 +1031,18 @@ static long tap_ioctl(struct file *file, unsigned int cmd,
 
 	case SIOCGIFHWADDR:
 		rtnl_lock();
-		vlan = tap_get_vlan(q);
-		if (!vlan) {
+		tap = tap_get_tap_dev(q);
+		if (!tap) {
 			rtnl_unlock();
 			return -ENOLINK;
 		}
 		ret = 0;
-		u = vlan->dev->type;
-		if (copy_to_user(&ifr->ifr_name, vlan->dev->name, IFNAMSIZ) ||
-		    copy_to_user(&ifr->ifr_hwaddr.sa_data, vlan->dev->dev_addr, ETH_ALEN) ||
+		u = tap->dev->type;
+		if (copy_to_user(&ifr->ifr_name, tap->dev->name, IFNAMSIZ) ||
+		    copy_to_user(&ifr->ifr_hwaddr.sa_data, tap->dev->dev_addr, ETH_ALEN) ||
 		    put_user(u, &ifr->ifr_hwaddr.sa_family))
 			ret = -EFAULT;
-		tap_put_vlan(vlan);
+		tap_put_tap_dev(tap);
 		rtnl_unlock();
 		return ret;
 
@@ -1078,13 +1050,13 @@ static long tap_ioctl(struct file *file, unsigned int cmd,
 		if (copy_from_user(&sa, &ifr->ifr_hwaddr, sizeof(sa)))
 			return -EFAULT;
 		rtnl_lock();
-		vlan = tap_get_vlan(q);
-		if (!vlan) {
+		tap = tap_get_tap_dev(q);
+		if (!tap) {
 			rtnl_unlock();
 			return -ENOLINK;
 		}
-		ret = dev_set_mac_address(vlan->dev, &sa);
-		tap_put_vlan(vlan);
+		ret = dev_set_mac_address(tap->dev, &sa);
+		tap_put_tap_dev(tap);
 		rtnl_unlock();
 		return ret;
 
@@ -1167,19 +1139,19 @@ struct socket *tap_get_socket(struct file *file)
 }
 EXPORT_SYMBOL_GPL(tap_get_socket);
 
-int tap_queue_resize(struct macvlan_dev *vlan)
+int tap_queue_resize(struct tap_dev *tap)
 {
-	struct net_device *dev = vlan->dev;
+	struct net_device *dev = tap->dev;
 	struct tap_queue *q;
 	struct skb_array **arrays;
-	int n = vlan->numqueues;
+	int n = tap->numqueues;
 	int ret, i = 0;
 
 	arrays = kmalloc(sizeof *arrays * n, GFP_KERNEL);
 	if (!arrays)
 		return -ENOMEM;
 
-	list_for_each_entry(q, &vlan->queue_list, next)
+	list_for_each_entry(q, &tap->queue_list, next)
 		arrays[i++] = &q->skb_array;
 
 	ret = skb_array_resize_multiple(arrays, n,
diff --git a/include/linux/if_tap.h b/include/linux/if_tap.h
index a2dfd90..75031e5 100644
--- a/include/linux/if_tap.h
+++ b/include/linux/if_tap.h
@@ -14,11 +14,60 @@ static inline struct socket *tap_get_socket(struct file *f)
 }
 #endif /* CONFIG_MACVTAP */
 
+#include <net/sock.h>
+#include <linux/skb_array.h>
+
+#define MAX_TAP_QUEUES 256
+
+struct tap_queue;
+
+struct tap_dev {
+	struct net_device	*dev;
+	u16			flags;
+	/* This array tracks active taps. */
+	struct tap_queue    __rcu *taps[MAX_TAP_QUEUES];
+	/* This list tracks all taps (both enabled and disabled) */
+	struct list_head	queue_list;
+	int			numvtaps;
+	int			numqueues;
+	netdev_features_t	tap_features;
+	int			minor;
+
+	void (*update_features)(struct tap_dev *tap, netdev_features_t features);
+	void (*count_tx_dropped)(struct tap_dev *tap);
+	void (*count_rx_dropped)(struct tap_dev *tap);
+};
+
+/*
+ * A tap queue is the central object of tap module, it connects
+ * an open character device to virtual interface. There can be
+ * multiple queues on one interface, which map back to queues
+ * implemented in hardware on the underlying device.
+ *
+ * tap_proto is used to allocate queues through the sock allocation
+ * mechanism.
+ *
+ */
+
+struct tap_queue {
+	struct sock sk;
+	struct socket sock;
+	struct socket_wq wq;
+	int vnet_hdr_sz;
+	struct tap_dev __rcu *tap;
+	struct file *file;
+	unsigned int flags;
+	u16 queue_index;
+	bool enabled;
+	struct list_head next;
+	struct skb_array skb_array;
+};
+
 rx_handler_result_t tap_handle_frame(struct sk_buff **pskb);
-void tap_del_queues(struct net_device *dev);
-int tap_get_minor(struct macvlan_dev *vlan);
-void tap_free_minor(struct macvlan_dev *vlan);
-int tap_queue_resize(struct macvlan_dev *vlan);
+void tap_del_queues(struct tap_dev *tap);
+int tap_get_minor(struct tap_dev *tap);
+void tap_free_minor(struct tap_dev *tap);
+int tap_queue_resize(struct tap_dev *tap);
 int tap_create_cdev(struct cdev *tap_cdev,
 		    dev_t *tap_major, const char *device_name);
 void tap_destroy_cdev(dev_t major, struct cdev *tap_cdev);
-- 
2.7.4

[toc] | [prev] | [next] | [standalone]

#1567110 — Re: [PATCHv3 0/7] Refactor macvtap to re-use tap functionality by other virtual intefaces

From: Sainath Grandhi <sainath.grandhi@intel.com>
Date: Wed, 25 Jan 2017 13:25:32 -0800

> Tap character devices can be implemented on other virtual interfaces like
> ipvlan, similar to macvtap. Source code for tap functionality in macvtap
> can be re-used for this purpose.
> 
> This patch series splits macvtap source into two modules, macvtap and tap.
> This patch series also includes a patch for implementing tap character
> device driver based on the IP-VLAN network interface, called ipvtap.
> 
> These patches are tested on x86 platform.

Please resubmit this series, after you have waited for other feedback,
with your Subject lines fixed.

Subsystem prefixes in the Subject lines such as "TAP" and "IPVTAP"
should be in all lowercase letters not all uppercase.

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web