[B.A.T.M.A.N.] [RFC next] batman-adv: ELP - use new ethtool_link_get_ksettings API

[B.A.T.M.A.N.] [RFC next] batman-adv: ELP - use new ethtool_link_get_ksettings API

[Antonio Quartulli]

The ethtool API is changing in linux-4.6 and the B.A.T.M.A.N. V
code has to be changed accordingly.

Fixes: 5c3245172c01 ("batman-adv: ELP - compute the metric based on the estimated throughput")
Signed-off-by: Antonio Quartulli <a@unstable.cc>
---

This patch contains some compat code which is an attempt to implement
backward compatibility with kernel <4.6. Unfortunately my attempt does
not work because it is not possible to redefine something like
link_settings.base.speed to something else (due to the '.' in the string).

Any suggestion? :)


 compat-include/linux/ethtool.h | 35 +++++++++++++++++++++++++++++++++++
 net/batman-adv/bat_v_elp.c     | 12 ++++++------
 2 files changed, 41 insertions(+), 6 deletions(-)
 create mode 100644 compat-include/linux/ethtool.h

diff --git a/compat-include/linux/ethtool.h b/compat-include/linux/ethtool.h
new file mode 100644
index 0000000..b11524d
--- /dev/null
+++ b/compat-include/linux/ethtool.h
@@ -0,0 +1,35 @@
+/* Copyright (C) 2016  B.A.T.M.A.N. contributors:
+ *
+ * Antonio Quartulli
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * 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.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * This file contains macros for maintaining compatibility with older versions
+ * of the Linux kernel.
+ */
+
+#ifndef _NET_BATMAN_ADV_COMPAT_LINUX_ETHTOOL_H_
+#define _NET_BATMAN_ADV_COMPAT_LINUX_ETHTOOL_H_
+
+#include <linux/version.h>
+#include_next <linux/ethtool.h>
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 6, 0)
+
+#define ethtool_link_ksettings ethtool_cmd
+#define __ethtool_get_link_ksettings __ethtool_get_settings
+#define link_settings.base.speed ethtool_cmd_speed(&link_settings)
+
+#endif /* < KERNEL_VERSION(4, 6, 0) */
+
+#endif	/* _NET_BATMAN_ADV_COMPAT_LINUX_ETHTOOL_H_ */
diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c
index e94b4a0..3844e7e 100644
--- a/net/batman-adv/bat_v_elp.c
+++ b/net/batman-adv/bat_v_elp.c
@@ -73,8 +73,8 @@ static void batadv_v_elp_start_timer(struct batadv_hard_iface *hard_iface)
 static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh)
 {
 	struct batadv_hard_iface *hard_iface = neigh->if_incoming;
+	struct ethtool_link_ksettings link_settings;
 	struct station_info sinfo;
-	struct ethtool_cmd cmd;
 	u32 throughput;
 	int ret;
 
@@ -110,19 +110,19 @@ static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh)
 	/* if not a wifi interface, check if this device provides data via
 	 * ethtool (e.g. an Ethernet adapter)
 	 */
-	memset(&cmd, 0, sizeof(cmd));
+	memset(&link_settings, 0, sizeof(link_settings));
 	rtnl_lock();
-	ret = __ethtool_get_settings(hard_iface->net_dev, &cmd);
+	ret = __ethtool_get_link_ksettings(hard_iface->net_dev, &link_settings);
 	rtnl_unlock();
 	if (ret == 0) {
 		/* link characteristics might change over time */
-		if (cmd.duplex == DUPLEX_FULL)
+		if (link_settings.base.duplex == DUPLEX_FULL)
 			hard_iface->bat_v.flags |= BATADV_FULL_DUPLEX;
 		else
 			hard_iface->bat_v.flags &= ~BATADV_FULL_DUPLEX;
 
-		throughput = ethtool_cmd_speed(&cmd);
-		if (throughput && throughput != SPEED_UNKNOWN)
+		throughput = link_settings.base.speed;
+		if (throughput && (throughput != SPEED_UNKNOWN))
 			return throughput * 10;
 	}
 
-- 
2.7.2

[B.A.T.M.A.N.] [RFC next v2] batman-adv: ELP - use new ethtool_link_get_ksettings API

[Sven Eckelmann]

From: Antonio Quartulli <a@unstable.cc>

The ethtool API is changing in linux-4.6 and the B.A.T.M.A.N. V
code has to be changed accordingly.

Fixes: 5c3245172c01 ("batman-adv: ELP - compute the metric based on the estimated throughput")
Signed-off-by: Antonio Quartulli <a@unstable.cc>
---
v2:
 - Add pseudo-struct and wrapper function which only provides speed + duplex

Antonio, this may be a way to add some compat code for this problem. I've
only did a fast compile check. So I would recommend to really test it :)

 compat-include/linux/ethtool.h | 62 ++++++++++++++++++++++++++++++++++++++++++
 net/batman-adv/bat_v_elp.c     | 12 ++++----
 2 files changed, 68 insertions(+), 6 deletions(-)
 create mode 100644 compat-include/linux/ethtool.h

diff --git a/compat-include/linux/ethtool.h b/compat-include/linux/ethtool.h
new file mode 100644
index 0000000..87f7577
--- /dev/null
+++ b/compat-include/linux/ethtool.h
@@ -0,0 +1,62 @@
+/* Copyright (C) 2016  B.A.T.M.A.N. contributors:
+ *
+ * Antonio Quartulli
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * 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.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ *
+ * This file contains macros for maintaining compatibility with older versions
+ * of the Linux kernel.
+ */
+
+#ifndef _NET_BATMAN_ADV_COMPAT_LINUX_ETHTOOL_H_
+#define _NET_BATMAN_ADV_COMPAT_LINUX_ETHTOOL_H_
+
+#include <linux/version.h>
+#include_next <linux/ethtool.h>
+
+#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 6, 0)
+
+#define ethtool_link_ksettings batadv_ethtool_link_ksettings
+
+struct batadv_ethtool_link_ksettings {
+	struct {
+		__u32	speed;
+		__u8	duplex;
+	} base;
+};
+
+#define __ethtool_get_link_ksettings(__dev, __link_settings) \
+	batadv_ethtool_get_link_ksettings(__dev, __link_settings)
+
+static inline int
+batadv_ethtool_get_link_ksettings(struct net_device *dev,
+				  struct ethtool_link_ksettings *link_ksettings)
+{
+	struct ethtool_cmd cmd;
+	int ret;
+
+	memset(&cmd, 0, sizeof(cmd));
+	ret = __ethtool_get_settings(dev, &cmd);
+
+	if (ret != 0)
+		return ret;
+
+	link_ksettings->base.duplex = cmd.duplex;
+	link_ksettings->base.speed = ethtool_cmd_speed(&cmd);
+
+	return 0;
+}
+
+#endif /* < KERNEL_VERSION(4, 6, 0) */
+
+#endif	/* _NET_BATMAN_ADV_COMPAT_LINUX_ETHTOOL_H_ */
diff --git a/net/batman-adv/bat_v_elp.c b/net/batman-adv/bat_v_elp.c
index e94b4a0..3844e7e 100644
--- a/net/batman-adv/bat_v_elp.c
+++ b/net/batman-adv/bat_v_elp.c
@@ -73,8 +73,8 @@ static void batadv_v_elp_start_timer(struct batadv_hard_iface *hard_iface)
 static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh)
 {
 	struct batadv_hard_iface *hard_iface = neigh->if_incoming;
+	struct ethtool_link_ksettings link_settings;
 	struct station_info sinfo;
-	struct ethtool_cmd cmd;
 	u32 throughput;
 	int ret;
 
@@ -110,19 +110,19 @@ static u32 batadv_v_elp_get_throughput(struct batadv_hardif_neigh_node *neigh)
 	/* if not a wifi interface, check if this device provides data via
 	 * ethtool (e.g. an Ethernet adapter)
 	 */
-	memset(&cmd, 0, sizeof(cmd));
+	memset(&link_settings, 0, sizeof(link_settings));
 	rtnl_lock();
-	ret = __ethtool_get_settings(hard_iface->net_dev, &cmd);
+	ret = __ethtool_get_link_ksettings(hard_iface->net_dev, &link_settings);
 	rtnl_unlock();
 	if (ret == 0) {
 		/* link characteristics might change over time */
-		if (cmd.duplex == DUPLEX_FULL)
+		if (link_settings.base.duplex == DUPLEX_FULL)
 			hard_iface->bat_v.flags |= BATADV_FULL_DUPLEX;
 		else
 			hard_iface->bat_v.flags &= ~BATADV_FULL_DUPLEX;
 
-		throughput = ethtool_cmd_speed(&cmd);
-		if (throughput && throughput != SPEED_UNKNOWN)
+		throughput = link_settings.base.speed;
+		if (throughput && (throughput != SPEED_UNKNOWN))
 			return throughput * 10;
 	}
 
-- 
2.7.0

Re: [B.A.T.M.A.N.] [RFC next v2] batman-adv: ELP - use new ethtool_link_get_ksettings API

[Antonio Quartulli]

[-- Attachment #1: Type: text/plain, Size: 1219 bytes --]

On Sun, Feb 28, 2016 at 08:31:34AM +0100, Sven Eckelmann wrote:
> +#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 6, 0)
> +
> +#define ethtool_link_ksettings batadv_ethtool_link_ksettings
> +
> +struct batadv_ethtool_link_ksettings {
> +	struct {
> +		__u32	speed;
> +		__u8	duplex;
> +	} base;
> +};
> +
> +#define __ethtool_get_link_ksettings(__dev, __link_settings) \
> +	batadv_ethtool_get_link_ksettings(__dev, __link_settings)
> +
> +static inline int
> +batadv_ethtool_get_link_ksettings(struct net_device *dev,
> +				  struct ethtool_link_ksettings *link_ksettings)
> +{
> +	struct ethtool_cmd cmd;
> +	int ret;
> +
> +	memset(&cmd, 0, sizeof(cmd));
> +	ret = __ethtool_get_settings(dev, &cmd);
> +
> +	if (ret != 0)
> +		return ret;
> +
> +	link_ksettings->base.duplex = cmd.duplex;
> +	link_ksettings->base.speed = ethtool_cmd_speed(&cmd);
> +
> +	return 0;
> +}
> +
> +#endif /* < KERNEL_VERSION(4, 6, 0) */

Hi Sven,

I really like this approach! thanks a lot!
By the way, you should have added your name to the list of contributor of this
file though :)


How about re-sending this RFC as PATCH (with your sign-off as well) ?


Cheers,

-- 
Antonio Quartulli

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

Re: [B.A.T.M.A.N.] [RFC next v2] batman-adv: ELP - use new ethtool_link_get_ksettings API

[Sven Eckelmann]

[-- Attachment #1: Type: text/plain, Size: 482 bytes --]

On Sunday 28 February 2016 17:08:38 Antonio Quartulli wrote:
[...]
> I really like this approach! thanks a lot!
> By the way, you should have added your name to the list of contributor of
> this file though :)

Maybe you never noticed but my name is nowhere in the batman-adv/batctl 
sourcecode. The only reason why it is in alfred/packet.h: because Simon added 
it.

> How about re-sending this RFC as PATCH (with your sign-off as well) ?

Yes, I will send it.

Kind regards,
	Sven

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 819 bytes --]