[PATCH net-next 0/2] break include loop and dsa devlink support

[PATCH net-next 0/2] break include loop and dsa devlink support

[Andrew Lunn]

These two patches add very basic support for devlink to DSA, in
preparation for playing with dpipe.

The first patch is needed to break an include loop between
netdevice.h, dsa.h and devlink.h. We need to remove dsa.h from
netdevice.h. As a result, some files fail to compile, because they
require includes pulled in via dsa.h. So this patch adds a number of
includes in various places. The majority is within the network
subsystem, but cifs also needs a few fixes.

0-day has been chewing on this for over a day now, and not found any
breakage. But Arnd's randconfig might uncover something.

Andrew Lunn (2):
  net: break include loop netdevice.h, dsa.h, devlink.h
  net: dsa: dsa2: Add basic support of devlink

 drivers/net/dsa/bcm_sf2_cfp.c                    |  3 +-
 drivers/net/dsa/mv88e6xxx/mv88e6xxx.h            |  1 +
 drivers/net/ethernet/broadcom/bcmsysport.c       |  1 +
 drivers/net/ethernet/freescale/fman/fman_memac.h |  1 +
 drivers/net/ethernet/hisilicon/hns/hnae.c        |  2 +-
 drivers/net/ieee802154/mrf24j40.c                |  1 +
 drivers/net/phy/smsc.c                           |  1 +
 drivers/net/usb/lan78xx.c                        |  1 +
 fs/cifs/cifsfs.c                                 |  1 +
 fs/cifs/connect.c                                |  1 +
 fs/cifs/smb2pdu.c                                |  1 +
 include/linux/netdevice.h                        | 12 +-----
 include/net/dsa.h                                | 14 +++++++
 net/bridge/br_if.c                               |  1 +
 net/core/netprio_cgroup.c                        |  1 +
 net/dsa/dsa.c                                    |  3 +-
 net/dsa/dsa2.c                                   | 50 ++++++++++++++++++++++--
 net/dsa/slave.c                                  |  1 +
 net/dsa/tag_brcm.c                               |  1 +
 net/dsa/tag_dsa.c                                |  1 +
 net/dsa/tag_edsa.c                               |  1 +
 net/dsa/tag_qca.c                                |  1 +
 net/dsa/tag_trailer.c                            |  1 +
 net/ipv4/ipconfig.c                              |  1 +
 24 files changed, 86 insertions(+), 16 deletions(-)

-- 
2.11.0

[PATCH net-next 1/2] net: break include loop netdevice.h, dsa.h, devlink.h

[Andrew Lunn]

There is an include loop between netdevice.h, dsa.h, devlink.h because
of NETDEV_ALIGN, making it impossible to use devlink structures in
dsa.h.

Break this loop by taking dsa.h out of netdevice.h, add a forward
declaration of dsa_switch_tree and netdev_set_default_ethtool_ops()
function, which is what netdevice.h requires.

No longer having dsa.h in netdevice.h means the includes in dsa.h no
longer get included. This breaks a few other files which depend on
these includes. Add these directly in the affected file.

Signed-off-by: Andrew Lunn <andrew-g2DYL2Zd6BY@public.gmane.org>
---
 drivers/net/dsa/bcm_sf2_cfp.c                    |  3 ++-
 drivers/net/dsa/mv88e6xxx/mv88e6xxx.h            |  1 +
 drivers/net/ethernet/broadcom/bcmsysport.c       |  1 +
 drivers/net/ethernet/freescale/fman/fman_memac.h |  1 +
 drivers/net/ethernet/hisilicon/hns/hnae.c        |  2 +-
 drivers/net/ieee802154/mrf24j40.c                |  1 +
 drivers/net/phy/smsc.c                           |  1 +
 drivers/net/usb/lan78xx.c                        |  1 +
 fs/cifs/cifsfs.c                                 |  1 +
 fs/cifs/connect.c                                |  1 +
 fs/cifs/smb2pdu.c                                |  1 +
 include/linux/netdevice.h                        | 12 ++----------
 include/net/dsa.h                                |  9 +++++++++
 net/bridge/br_if.c                               |  1 +
 net/core/netprio_cgroup.c                        |  1 +
 net/dsa/dsa.c                                    |  3 ++-
 net/dsa/dsa2.c                                   |  3 ++-
 net/dsa/slave.c                                  |  1 +
 net/dsa/tag_brcm.c                               |  1 +
 net/dsa/tag_dsa.c                                |  1 +
 net/dsa/tag_edsa.c                               |  1 +
 net/dsa/tag_qca.c                                |  1 +
 net/dsa/tag_trailer.c                            |  1 +
 net/ipv4/ipconfig.c                              |  1 +
 24 files changed, 36 insertions(+), 14 deletions(-)

diff --git a/drivers/net/dsa/bcm_sf2_cfp.c b/drivers/net/dsa/bcm_sf2_cfp.c
index 346dd9a1232d..2fb32d67065f 100644
--- a/drivers/net/dsa/bcm_sf2_cfp.c
+++ b/drivers/net/dsa/bcm_sf2_cfp.c
@@ -10,10 +10,11 @@
  */
 
 #include <linux/list.h>
-#include <net/dsa.h>
 #include <linux/ethtool.h>
 #include <linux/if_ether.h>
 #include <linux/in.h>
+#include <linux/netdevice.h>
+#include <net/dsa.h>
 #include <linux/bitmap.h>
 
 #include "bcm_sf2.h"
diff --git a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h
index 75be2c339a49..55367d05374e 100644
--- a/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx/mv88e6xxx.h
@@ -16,6 +16,7 @@
 #include <linux/irq.h>
 #include <linux/gpio/consumer.h>
 #include <linux/phy.h>
+#include <net/dsa.h>
 
 #ifndef UINT64_MAX
 #define UINT64_MAX		(u64)(~((u64)0))
diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c
index 61e26c6b26ab..099b374c1b17 100644
--- a/drivers/net/ethernet/broadcom/bcmsysport.c
+++ b/drivers/net/ethernet/broadcom/bcmsysport.c
@@ -22,6 +22,7 @@
 #include <linux/of_mdio.h>
 #include <linux/phy.h>
 #include <linux/phy_fixed.h>
+#include <net/dsa.h>
 #include <net/ip.h>
 #include <net/ipv6.h>
 
diff --git a/drivers/net/ethernet/freescale/fman/fman_memac.h b/drivers/net/ethernet/freescale/fman/fman_memac.h
index 173d8e0fd716..c4a66469a907 100644
--- a/drivers/net/ethernet/freescale/fman/fman_memac.h
+++ b/drivers/net/ethernet/freescale/fman/fman_memac.h
@@ -36,6 +36,7 @@
 #include "fman_mac.h"
 
 #include <linux/netdevice.h>
+#include <linux/phy_fixed.h>
 
 struct fman_mac *memac_config(struct fman_mac_params *params);
 int memac_set_promiscuous(struct fman_mac *memac, bool new_val);
diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.c b/drivers/net/ethernet/hisilicon/hns/hnae.c
index b6ed818f78ff..120427a40883 100644
--- a/drivers/net/ethernet/hisilicon/hns/hnae.c
+++ b/drivers/net/ethernet/hisilicon/hns/hnae.c
@@ -9,9 +9,9 @@
 
 #include <linux/dma-mapping.h>
 #include <linux/interrupt.h>
+#include <linux/of.h>
 #include <linux/skbuff.h>
 #include <linux/slab.h>
-
 #include "hnae.h"
 
 #define cls_to_ae_dev(dev) container_of(dev, struct hnae_ae_dev, cls_dev)
diff --git a/drivers/net/ieee802154/mrf24j40.c b/drivers/net/ieee802154/mrf24j40.c
index 7b131f8e4093..bd63289c55e8 100644
--- a/drivers/net/ieee802154/mrf24j40.c
+++ b/drivers/net/ieee802154/mrf24j40.c
@@ -18,6 +18,7 @@
 #include <linux/spi/spi.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/regmap.h>
 #include <linux/ieee802154.h>
 #include <linux/irq.h>
diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index fb32eaf2255d..cef6967b0396 100644
--- a/drivers/net/phy/smsc.c
+++ b/drivers/net/phy/smsc.c
@@ -20,6 +20,7 @@
 #include <linux/module.h>
 #include <linux/mii.h>
 #include <linux/ethtool.h>
+#include <linux/of.h>
 #include <linux/phy.h>
 #include <linux/netdevice.h>
 #include <linux/smscphy.h>
diff --git a/drivers/net/usb/lan78xx.c b/drivers/net/usb/lan78xx.c
index d885e0325422..a17e32bf5f92 100644
--- a/drivers/net/usb/lan78xx.c
+++ b/drivers/net/usb/lan78xx.c
@@ -29,6 +29,7 @@
 #include <linux/ip.h>
 #include <linux/ipv6.h>
 #include <linux/mdio.h>
+#include <linux/phy.h>
 #include <net/ip6_checksum.h>
 #include <linux/interrupt.h>
 #include <linux/irqdomain.h>
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 15e1db8738ae..8c91f37ac0eb 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -37,6 +37,7 @@
 #include <linux/freezer.h>
 #include <linux/namei.h>
 #include <linux/random.h>
+#include <linux/uuid.h>
 #include <linux/xattr.h>
 #include <net/ipv6.h>
 #include "cifsfs.h"
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 9ae695ae3ed7..858698dcde3c 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -35,6 +35,7 @@
 #include <linux/pagevec.h>
 #include <linux/freezer.h>
 #include <linux/namei.h>
+#include <linux/uuid.h>
 #include <linux/uaccess.h>
 #include <asm/processor.h>
 #include <linux/inet.h>
diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c
index 7446496850a3..fb75fe908225 100644
--- a/fs/cifs/smb2pdu.c
+++ b/fs/cifs/smb2pdu.c
@@ -33,6 +33,7 @@
 #include <linux/vfs.h>
 #include <linux/task_io_accounting_ops.h>
 #include <linux/uaccess.h>
+#include <linux/uuid.h>
 #include <linux/pagemap.h>
 #include <linux/xattr.h>
 #include "smb2pdu.h"
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index b7365b587818..cc07c3be2705 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -41,7 +41,6 @@
 
 #include <linux/ethtool.h>
 #include <net/net_namespace.h>
-#include <net/dsa.h>
 #ifdef CONFIG_DCB
 #include <net/dcbnl.h>
 #endif
@@ -57,6 +56,8 @@
 struct netpoll_info;
 struct device;
 struct phy_device;
+struct dsa_switch_tree;
+
 /* 802.11 specific */
 struct wireless_dev;
 /* 802.15.4 specific */
@@ -2004,15 +2005,6 @@ void dev_net_set(struct net_device *dev, struct net *net)
 	write_pnet(&dev->nd_net, net);
 }
 
-static inline bool netdev_uses_dsa(struct net_device *dev)
-{
-#if IS_ENABLED(CONFIG_NET_DSA)
-	if (dev->dsa_ptr != NULL)
-		return dsa_uses_tagged_protocol(dev->dsa_ptr);
-#endif
-	return false;
-}
-
 /**
  *	netdev_priv - access network device private data
  *	@dev: network device
diff --git a/include/net/dsa.h b/include/net/dsa.h
index e42897fd7a96..f80e81912b83 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -468,6 +468,15 @@ static inline bool dsa_uses_tagged_protocol(struct dsa_switch_tree *dst)
 	return dst->rcv != NULL;
 }
 
+static inline bool netdev_uses_dsa(struct net_device *dev)
+{
+#if IS_ENABLED(CONFIG_NET_DSA)
+	if (dev->dsa_ptr != NULL)
+		return dsa_uses_tagged_protocol(dev->dsa_ptr);
+#endif
+	return false;
+}
+
 struct dsa_switch *dsa_switch_alloc(struct device *dev, size_t n);
 void dsa_unregister_switch(struct dsa_switch *ds);
 int dsa_register_switch(struct dsa_switch *ds, struct device *dev);
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 8ac1770aa222..6eb52d422dd9 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -22,6 +22,7 @@
 #include <linux/rtnetlink.h>
 #include <linux/if_ether.h>
 #include <linux/slab.h>
+#include <net/dsa.h>
 #include <net/sock.h>
 #include <linux/if_vlan.h>
 #include <net/switchdev.h>
diff --git a/net/core/netprio_cgroup.c b/net/core/netprio_cgroup.c
index 0f9275ee5595..1c4810919a0a 100644
--- a/net/core/netprio_cgroup.c
+++ b/net/core/netprio_cgroup.c
@@ -11,6 +11,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/types.h>
 #include <linux/module.h>
diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index b6d4f6a23f06..95d1a756202c 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -14,15 +14,16 @@
 #include <linux/platform_device.h>
 #include <linux/slab.h>
 #include <linux/module.h>
-#include <net/dsa.h>
 #include <linux/of.h>
 #include <linux/of_mdio.h>
 #include <linux/of_platform.h>
 #include <linux/of_net.h>
 #include <linux/of_gpio.h>
+#include <linux/netdevice.h>
 #include <linux/sysfs.h>
 #include <linux/phy_fixed.h>
 #include <linux/gpio/consumer.h>
+#include <net/dsa.h>
 #include "dsa_priv.h"
 
 static struct sk_buff *dsa_slave_notag_xmit(struct sk_buff *skb,
diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c
index 737be6470c7f..d039c8d7adfd 100644
--- a/net/dsa/dsa2.c
+++ b/net/dsa/dsa2.c
@@ -13,11 +13,12 @@
 #include <linux/device.h>
 #include <linux/err.h>
 #include <linux/list.h>
+#include <linux/netdevice.h>
 #include <linux/slab.h>
 #include <linux/rtnetlink.h>
-#include <net/dsa.h>
 #include <linux/of.h>
 #include <linux/of_net.h>
+#include <net/dsa.h>
 #include "dsa_priv.h"
 
 static LIST_HEAD(dsa_switch_trees);
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 78128acfbf63..7693182df81e 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -17,6 +17,7 @@
 #include <linux/of_mdio.h>
 #include <linux/mdio.h>
 #include <linux/list.h>
+#include <net/dsa.h>
 #include <net/rtnetlink.h>
 #include <net/switchdev.h>
 #include <net/pkt_cls.h>
diff --git a/net/dsa/tag_brcm.c b/net/dsa/tag_brcm.c
index 5d925b6b2bb1..e2ed6cf68261 100644
--- a/net/dsa/tag_brcm.c
+++ b/net/dsa/tag_brcm.c
@@ -12,6 +12,7 @@
 #include <linux/etherdevice.h>
 #include <linux/list.h>
 #include <linux/slab.h>
+#include <net/dsa.h>
 #include "dsa_priv.h"
 
 /* This tag length is 4 bytes, older ones were 6 bytes, we do not
diff --git a/net/dsa/tag_dsa.c b/net/dsa/tag_dsa.c
index 72579ceea381..e42ba906100c 100644
--- a/net/dsa/tag_dsa.c
+++ b/net/dsa/tag_dsa.c
@@ -11,6 +11,7 @@
 #include <linux/etherdevice.h>
 #include <linux/list.h>
 #include <linux/slab.h>
+#include <net/dsa.h>
 #include "dsa_priv.h"
 
 #define DSA_HLEN	4
diff --git a/net/dsa/tag_edsa.c b/net/dsa/tag_edsa.c
index 648c051817a1..6a9b7a9e4e15 100644
--- a/net/dsa/tag_edsa.c
+++ b/net/dsa/tag_edsa.c
@@ -11,6 +11,7 @@
 #include <linux/etherdevice.h>
 #include <linux/list.h>
 #include <linux/slab.h>
+#include <net/dsa.h>
 #include "dsa_priv.h"
 
 #define DSA_HLEN	4
diff --git a/net/dsa/tag_qca.c b/net/dsa/tag_qca.c
index 30240f343aea..4e0dad759d04 100644
--- a/net/dsa/tag_qca.c
+++ b/net/dsa/tag_qca.c
@@ -12,6 +12,7 @@
  */
 
 #include <linux/etherdevice.h>
+#include <net/dsa.h>
 #include "dsa_priv.h"
 
 #define QCA_HDR_LEN	2
diff --git a/net/dsa/tag_trailer.c b/net/dsa/tag_trailer.c
index 26f977176978..74c948512550 100644
--- a/net/dsa/tag_trailer.c
+++ b/net/dsa/tag_trailer.c
@@ -11,6 +11,7 @@
 #include <linux/etherdevice.h>
 #include <linux/list.h>
 #include <linux/slab.h>
+#include <net/dsa.h>
 #include "dsa_priv.h"
 
 static struct sk_buff *trailer_xmit(struct sk_buff *skb, struct net_device *dev)
diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
index fd9f34bbd740..9def8ed31c76 100644
--- a/net/ipv4/ipconfig.c
+++ b/net/ipv4/ipconfig.c
@@ -57,6 +57,7 @@
 #include <linux/export.h>
 #include <net/net_namespace.h>
 #include <net/arp.h>
+#include <net/dsa.h>
 #include <net/ip.h>
 #include <net/ipconfig.h>
 #include <net/route.h>
-- 
2.11.0

[PATCH net-next 2/2] net: dsa: dsa2: Add basic support of devlink

[Andrew Lunn]

Register the switch and its ports with devlink.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
---
 include/net/dsa.h |  5 +++++
 net/dsa/dsa2.c    | 47 +++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 50 insertions(+), 2 deletions(-)

diff --git a/include/net/dsa.h b/include/net/dsa.h
index f80e81912b83..951b5e49e899 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -19,6 +19,7 @@
 #include <linux/workqueue.h>
 #include <linux/of.h>
 #include <linux/ethtool.h>
+#include <net/devlink.h>
 
 struct tc_action;
 struct phy_device;
@@ -182,6 +183,7 @@ struct dsa_port {
 	unsigned int		ageing_time;
 	u8			stp_state;
 	struct net_device	*bridge_dev;
+	struct devlink_port	devlink_port;
 };
 
 struct dsa_switch {
@@ -237,6 +239,9 @@ struct dsa_switch {
 	unsigned int ageing_time_min;
 	unsigned int ageing_time_max;
 
+	/* devlink used to represent this switch device */
+	struct devlink		*devlink;
+
 	/* Dynamically allocated ports, keep last */
 	size_t num_ports;
 	struct dsa_port ports[];
diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c
index d039c8d7adfd..033b3bfb63dc 100644
--- a/net/dsa/dsa2.c
+++ b/net/dsa/dsa2.c
@@ -24,6 +24,9 @@
 static LIST_HEAD(dsa_switch_trees);
 static DEFINE_MUTEX(dsa2_mutex);
 
+static const struct devlink_ops dsa_devlink_ops = {
+};
+
 static struct dsa_switch_tree *dsa_get_dst(u32 tree)
 {
 	struct dsa_switch_tree *dst;
@@ -223,12 +226,18 @@ static int dsa_dsa_port_apply(struct dsa_port *port, u32 index,
 		return err;
 	}
 
-	return 0;
+	memset(&ds->ports[index].devlink_port, 0,
+	       sizeof(ds->ports[index].devlink_port));
+
+	return devlink_port_register(ds->devlink,
+				     &ds->ports[index].devlink_port,
+				     index);
 }
 
 static void dsa_dsa_port_unapply(struct dsa_port *port, u32 index,
 				 struct dsa_switch *ds)
 {
+	devlink_port_unregister(&ds->ports[index].devlink_port);
 	dsa_cpu_dsa_destroy(port);
 }
 
@@ -246,12 +255,17 @@ static int dsa_cpu_port_apply(struct dsa_port *port, u32 index,
 
 	ds->cpu_port_mask |= BIT(index);
 
-	return 0;
+	memset(&ds->ports[index].devlink_port, 0,
+	       sizeof(ds->ports[index].devlink_port));
+	err = devlink_port_register(ds->devlink, &ds->ports[index].devlink_port,
+				    index);
+	return err;
 }
 
 static void dsa_cpu_port_unapply(struct dsa_port *port, u32 index,
 				 struct dsa_switch *ds)
 {
+	devlink_port_unregister(&ds->ports[index].devlink_port);
 	dsa_cpu_dsa_destroy(port);
 	ds->cpu_port_mask &= ~BIT(index);
 
@@ -276,12 +290,23 @@ static int dsa_user_port_apply(struct dsa_port *port, u32 index,
 		return err;
 	}
 
+	memset(&ds->ports[index].devlink_port, 0,
+	       sizeof(ds->ports[index].devlink_port));
+	err = devlink_port_register(ds->devlink, &ds->ports[index].devlink_port,
+				    index);
+	if (err)
+		return err;
+
+	devlink_port_type_eth_set(&ds->ports[index].devlink_port,
+				  ds->ports[index].netdev);
+
 	return 0;
 }
 
 static void dsa_user_port_unapply(struct dsa_port *port, u32 index,
 				  struct dsa_switch *ds)
 {
+	devlink_port_unregister(&ds->ports[index].devlink_port);
 	if (ds->ports[index].netdev) {
 		dsa_slave_destroy(ds->ports[index].netdev);
 		ds->ports[index].netdev = NULL;
@@ -302,6 +327,17 @@ static int dsa_ds_apply(struct dsa_switch_tree *dst, struct dsa_switch *ds)
 	 */
 	ds->phys_mii_mask = ds->enabled_port_mask;
 
+	/* Add the switch to devlink before calling setup, so that setup can
+	 * add dpipe tables
+	 */
+	ds->devlink = devlink_alloc(&dsa_devlink_ops, 0);
+	if (!ds->devlink)
+		return -ENOMEM;
+
+	err = devlink_register(ds->devlink, ds->dev);
+	if (err)
+		return err;
+
 	err = ds->ops->setup(ds);
 	if (err < 0)
 		return err;
@@ -382,6 +418,13 @@ static void dsa_ds_unapply(struct dsa_switch_tree *dst, struct dsa_switch *ds)
 		mdiobus_unregister(ds->slave_mii_bus);
 
 	dsa_switch_unregister_notifier(ds);
+
+	if (ds->devlink) {
+		devlink_unregister(ds->devlink);
+		devlink_free(ds->devlink);
+		ds->devlink = NULL;
+	}
+
 }
 
 static int dsa_dst_apply(struct dsa_switch_tree *dst)
-- 
2.11.0

Re: [PATCH net-next 1/2] net: break include loop netdevice.h, dsa.h, devlink.h

[Florian Fainelli]

On 03/28/2017 02:45 PM, Andrew Lunn wrote:
> There is an include loop between netdevice.h, dsa.h, devlink.h because
> of NETDEV_ALIGN, making it impossible to use devlink structures in
> dsa.h.
> 
> Break this loop by taking dsa.h out of netdevice.h, add a forward
> declaration of dsa_switch_tree and netdev_set_default_ethtool_ops()
> function, which is what netdevice.h requires.
> 
> No longer having dsa.h in netdevice.h means the includes in dsa.h no
> longer get included. This breaks a few other files which depend on
> these includes. Add these directly in the affected file.
> 
> Signed-off-by: Andrew Lunn <andrew-g2DYL2Zd6BY@public.gmane.org>

Woah, thanks for doing this!

Reviewed-by: Florian Fainelli <f.fainelli-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
--
Florian

Re: [PATCH net-next 2/2] net: dsa: dsa2: Add basic support of devlink

[Florian Fainelli]

On 03/28/2017 02:45 PM, Andrew Lunn wrote:
> Register the switch and its ports with devlink.
> 
> Signed-off-by: Andrew Lunn <andrew@lunn.ch>

Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Tested-by: Florian Fainelli <f.fainelli@gmail.com>

root@net-vm:~# devlink -p port
mdio_bus/fixed-0:1f/0: type eth netdev lan1
mdio_bus/fixed-0:1f/1: type eth netdev lan2
mdio_bus/fixed-0:1f/2: type eth netdev lan3
mdio_bus/fixed-0:1f/3: type eth netdev lan4
mdio_bus/fixed-0:1f/5: type notset
root@net-vm:~# devlink -p dev
mdio_bus/fixed-0:1f
-- 
Florian

Re: [PATCH net-next 2/2] net: dsa: dsa2: Add basic support of devlink

[Andrew Lunn]

On Tue, Mar 28, 2017 at 05:33:18PM -0700, Florian Fainelli wrote:
> On 03/28/2017 02:45 PM, Andrew Lunn wrote:
> > Register the switch and its ports with devlink.
> > 
> > Signed-off-by: Andrew Lunn <andrew-g2DYL2Zd6BY@public.gmane.org>
> 
> Reviewed-by: Florian Fainelli <f.fainelli-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
> Tested-by: Florian Fainelli <f.fainelli-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
> 
> root@net-vm:~# devlink -p port
> mdio_bus/fixed-0:1f/0: type eth netdev lan1
> mdio_bus/fixed-0:1f/1: type eth netdev lan2
> mdio_bus/fixed-0:1f/2: type eth netdev lan3
> mdio_bus/fixed-0:1f/3: type eth netdev lan4
> mdio_bus/fixed-0:1f/5: type notset
> root@net-vm:~# devlink -p dev
> mdio_bus/fixed-0:1f

Thanks Florian

I have another patch somewhere which adds new types for CPU and DSA
ports. Once these 2 patches have landed, and the reports of broken
builds stop, i will dig it out.

But my main priority at the moment is to play with dpipe and make sure
it can represent tables we have in DSA switches, before the Binary API
is frozen.

       Andrew

Re: [PATCH net-next 0/2] break include loop and dsa devlink support

[David Miller]

From: Andrew Lunn <andrew-g2DYL2Zd6BY@public.gmane.org>
Date: Tue, 28 Mar 2017 23:45:05 +0200

> These two patches add very basic support for devlink to DSA, in
> preparation for playing with dpipe.
> 
> The first patch is needed to break an include loop between
> netdevice.h, dsa.h and devlink.h. We need to remove dsa.h from
> netdevice.h. As a result, some files fail to compile, because they
> require includes pulled in via dsa.h. So this patch adds a number of
> includes in various places. The majority is within the network
> subsystem, but cifs also needs a few fixes.
> 
> 0-day has been chewing on this for over a day now, and not found any
> breakage. But Arnd's randconfig might uncover something.

Looks good, series applied, thanks.

Re: [PATCH net-next 2/2] net: dsa: dsa2: Add basic support of devlink

[Florian Fainelli]

On 03/28/2017 06:31 PM, Andrew Lunn wrote:
> On Tue, Mar 28, 2017 at 05:33:18PM -0700, Florian Fainelli wrote:
>> On 03/28/2017 02:45 PM, Andrew Lunn wrote:
>>> Register the switch and its ports with devlink.
>>>
>>> Signed-off-by: Andrew Lunn <andrew@lunn.ch>
>>
>> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
>> Tested-by: Florian Fainelli <f.fainelli@gmail.com>
>>
>> root@net-vm:~# devlink -p port
>> mdio_bus/fixed-0:1f/0: type eth netdev lan1
>> mdio_bus/fixed-0:1f/1: type eth netdev lan2
>> mdio_bus/fixed-0:1f/2: type eth netdev lan3
>> mdio_bus/fixed-0:1f/3: type eth netdev lan4
>> mdio_bus/fixed-0:1f/5: type notset
>> root@net-vm:~# devlink -p dev
>> mdio_bus/fixed-0:1f
> 
> Thanks Florian
> 
> I have another patch somewhere which adds new types for CPU and DSA
> ports. Once these 2 patches have landed, and the reports of broken
> builds stop, i will dig it out.
> 
> But my main priority at the moment is to play with dpipe and make sure
> it can represent tables we have in DSA switches, before the Binary API
> is frozen.

Yes, that is a very good point actually.

Tangential question: we do not currently report to user space what kind
of switch tagging protocol is used (DSA, eDSA etc.) I was going to add a
sysfs attribute for that under the switch device's directory, but maybe
devlink would be more appropriate for that?
-- 
Florian

Re: [PATCH net-next 2/2] net: dsa: dsa2: Add basic support of devlink

[Andrew Lunn]

> Tangential question: we do not currently report to user space what kind
> of switch tagging protocol is used (DSA, eDSA etc.) I was going to add a
> sysfs attribute for that under the switch device's directory, but maybe
> devlink would be more appropriate for that?

We don't report it. What are you thinking you want to do with this
information?

I suppose it might be possible to teach libpcap/tcpdump about it, so
it can slide in an additional dissector. EDSA you can detect anyway,
since it uses a well know Ethertype. tcpdump already supports
this. But the other tag protocols cannot be detected.

devlink does seem like a good candidate for this, make it part of the
port type.

     Andrew

Re: [PATCH net-next 2/2] net: dsa: dsa2: Add basic support of devlink

[Jiri Pirko]

Wed, Mar 29, 2017 at 07:36:24PM CEST, andrew@lunn.ch wrote:
>> Tangential question: we do not currently report to user space what kind
>> of switch tagging protocol is used (DSA, eDSA etc.) I was going to add a
>> sysfs attribute for that under the switch device's directory, but maybe
>> devlink would be more appropriate for that?
>
>We don't report it. What are you thinking you want to do with this
>information?
>
>I suppose it might be possible to teach libpcap/tcpdump about it, so
>it can slide in an additional dissector. EDSA you can detect anyway,
>since it uses a well know Ethertype. tcpdump already supports
>this. But the other tag protocols cannot be detected.
>
>devlink does seem like a good candidate for this, make it part of the
>port type.

I agree this could go out via devlink. Not sure exactly what attr.
Certainly some new.


>
>     Andrew