[PATCH 0/2] Display adjacent switch port's attributes

[PATCH 0/2] Display adjacent switch port's attributes

[Stefan Raspl]

This patch series adds a new callback for ethtool to display the adjacent switch
port's attributes, as perceived by the NIC, e.g. through respective LLDP message
exchanges.
A patch for the qeth device driver provides a sample exploiter for the new
functionality.
A new option for ethtool to display the settings will be posted in a separate
patch.

Best regards,
Stefan Raspl

Stefan Raspl (2):
  ethtool: Add callback to indicate adjacent switch port attributes
  qeth: Display adjacent switch port attributes in ethtool

 drivers/s390/net/qeth_core.h      |  7 ++++
 drivers/s390/net/qeth_core_main.c | 79 +++++++++++++++++++++++++++++++++++++++
 drivers/s390/net/qeth_core_mpc.h  | 17 +++++++++
 drivers/s390/net/qeth_l2_main.c   |  1 +
 drivers/s390/net/qeth_l3_main.c   |  1 +
 include/linux/ethtool.h           |  3 ++
 include/uapi/linux/ethtool.h      | 35 +++++++++++++++++
 net/core/ethtool.c                | 22 +++++++++++
 8 files changed, 165 insertions(+)

-- 
1.8.3.4

[PATCH 1/2] ethtool: Add callback to indicate adjacent switch port attributes

[Stefan Raspl]

Switches supporting LLDP can communicate port attributes to connected devices.
Device drivers capable of accessing this information from the devices can use
the new callback get_switch_port_attrs() to report supported and enabled
settings in the card's adjacent switch port for display in ethtool.
Implementors have to use the respective SUPPORTED_SP_* and ENABLED_SP_* defines
to indicate the current settings.

Signed-off-by: Stefan Raspl <raspl@linux.vnet.ibm.com>
---
 include/linux/ethtool.h      |  3 +++
 include/uapi/linux/ethtool.h | 35 +++++++++++++++++++++++++++++++++++
 net/core/ethtool.c           | 22 ++++++++++++++++++++++
 3 files changed, 60 insertions(+)

diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index c8e3e7e3..940c7b1 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -177,6 +177,7 @@ static inline u32 ethtool_rxfh_indir_default(u32 index, u32 n_rx_rings)
  * @get_module_eeprom: Get the eeprom information from the plug-in module
  * @get_eee: Get Energy-Efficient (EEE) supported and status.
  * @set_eee: Set EEE status (enable/disable) as well as LPI timers.
+ * @get_switch_port_attrs: Get adjacent switch port attributes.
  *
  * All operations are optional (i.e. the function pointer may be set
  * to %NULL) and callers must take this into account.  Callers must
@@ -245,6 +246,8 @@ struct ethtool_ops {
 				     struct ethtool_eeprom *, u8 *);
 	int	(*get_eee)(struct net_device *, struct ethtool_eee *);
 	int	(*set_eee)(struct net_device *, struct ethtool_eee *);
+	int     (*get_switch_port_attrs)(struct net_device *,
+					 struct ethtool_swport_attrs *);
 
 
 };
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 38dbafa..f5843ac 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -136,6 +136,22 @@ struct ethtool_eeprom {
 };
 
 /**
+ * struct ethtool_swport_attrs - query adjacent switch port attributes
+ * @cmd: ETHTOOL_GPORT
+ * @port_rc: Use GPORT_RC_* as appropriate.
+ * @supported: Forwarding modes and capabilities supported by the switch port,
+ *	see SUPPORTED_SP_* flags.
+ * @enabled: Forwarding modes and capabilities currently activated at the
+ *	adjacent switch port, see ENABLED_SP_* flags.
+ */
+struct ethtool_swport_attrs {
+	__u32	cmd;
+	__u32	port_rc;
+	__u32	supported;
+	__u32	enabled;
+};
+
+/**
  * struct ethtool_eee - Energy Efficient Ethernet information
  * @cmd: ETHTOOL_{G,S}EEE
  * @supported: Mask of %SUPPORTED_* flags for the speed/duplex combinations
@@ -900,6 +916,7 @@ enum ethtool_sfeatures_retval_bits {
 #define ETHTOOL_GMODULEEEPROM	0x00000043 /* Get plug-in module eeprom */
 #define ETHTOOL_GEEE		0x00000044 /* Get EEE settings */
 #define ETHTOOL_SEEE		0x00000045 /* Set EEE settings */
+#define ETHTOOL_GPORT		0x00000046 /* Get switch port attributes */
 
 /* compatibility with older code */
 #define SPARC_ETH_GSET		ETHTOOL_GSET
@@ -1067,6 +1084,24 @@ enum ethtool_sfeatures_retval_bits {
 #define ETH_MODULE_SFF_8472		0x2
 #define ETH_MODULE_SFF_8472_LEN		512
 
+/* Bad return codes for switch ports */
+#define GPORT_RC_LLDP_UNSUP	1	/* switch port doesn't support */
+					/* required LLDP EVB TLV       */
+
+/* Indicates what features the adjacent switch port supports. */
+#define SUPPORTED_SP_FWD_802_1	(1 << 0)
+#define SUPPORTED_SP_FWD_RR	(1 << 1)
+#define SUPPORTED_SP_CAP_RTE	(1 << 9)
+#define SUPPORTED_SP_CAP_ECP	(1 << 10)
+#define SUPPORTED_SP_CAP_VDP	(1 << 11)
+
+/* Indicates what features the adjacent switch port has enabled. */
+#define ENABLED_SP_FWD_802_1	(1 << 0)
+#define ENABLED_SP_FWD_RR	(1 << 1)
+#define ENABLED_SP_CAP_RTE	(1 << 9)
+#define ENABLED_SP_CAP_ECP	(1 << 10)
+#define ENABLED_SP_CAP_VDP	(1 << 11)
+
 /* Reset flags */
 /* The reset() operation must clear the flags for the components which
  * were actually reset.  On successful return, the flags indicate the
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index 30071de..84f69f1 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -1446,6 +1446,25 @@ static int ethtool_get_module_eeprom(struct net_device *dev,
 				      modinfo.eeprom_len);
 }
 
+static int ethtool_get_switch_port_attrs(struct net_device *dev,
+							void __user *useraddr)
+{
+	struct ethtool_swport_attrs attrs = { ETHTOOL_GPORT };
+	int rc;
+
+	if (!dev->ethtool_ops->get_switch_port_attrs)
+		return -EOPNOTSUPP;
+
+	rc = dev->ethtool_ops->get_switch_port_attrs(dev, &attrs);
+	if (rc)
+		return rc;
+
+	if (copy_to_user(useraddr, &attrs, sizeof(attrs)))
+		return -EFAULT;
+
+	return 0;
+}
+
 /* The main entry point in this file.  Called from net/core/dev_ioctl.c */
 
 int dev_ethtool(struct net *net, struct ifreq *ifr)
@@ -1675,6 +1694,9 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
 	case ETHTOOL_GMODULEEEPROM:
 		rc = ethtool_get_module_eeprom(dev, useraddr);
 		break;
+	case ETHTOOL_GPORT:
+		rc = ethtool_get_switch_port_attrs(dev, useraddr);
+		break;
 	default:
 		rc = -EOPNOTSUPP;
 	}
-- 
1.8.3.4

[PATCH 2/2] qeth: Display adjacent switch port attributes in ethtool

[Stefan Raspl]

Add support for get_switch_port_attrs() callback in ethtool to display
adjacent switch port's attributes.

Signed-off-by: Stefan Raspl <raspl@linux.vnet.ibm.com>
Acked-by: Frank Blaschka <blaschka@linux.vnet.ibm.com>
---
 drivers/s390/net/qeth_core.h      |  7 ++++
 drivers/s390/net/qeth_core_main.c | 79 +++++++++++++++++++++++++++++++++++++++
 drivers/s390/net/qeth_core_mpc.h  | 17 +++++++++
 drivers/s390/net/qeth_l2_main.c   |  1 +
 drivers/s390/net/qeth_l3_main.c   |  1 +
 5 files changed, 105 insertions(+)

diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 41ef943..bc30a2a 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -738,6 +738,11 @@ struct qeth_rx {
 	int qdio_err;
 };
 
+struct qeth_switch_info {
+        __u32 capabilities;
+        __u32 settings;
+};
+
 #define QETH_NAPI_WEIGHT NAPI_POLL_WEIGHT
 
 struct qeth_card {
@@ -914,6 +919,8 @@ struct qeth_cmd_buffer *qeth_wait_for_buffer(struct qeth_channel *);
 int qeth_mdio_read(struct net_device *, int, int);
 int qeth_snmp_command(struct qeth_card *, char __user *);
 int qeth_query_oat_command(struct qeth_card *, char __user *);
+int qeth_core_ethtool_get_switch_port_attrs(struct net_device *,
+						struct ethtool_swport_attrs *);
 int qeth_send_control_data(struct qeth_card *, int, struct qeth_cmd_buffer *,
 	int (*reply_cb)(struct qeth_card *, struct qeth_reply*, unsigned long),
 	void *reply_param);
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index eb4e1f8..b8827bd 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -3036,6 +3036,42 @@ int qeth_query_ipassists(struct qeth_card *card, enum qeth_prot_versions prot)
 }
 EXPORT_SYMBOL_GPL(qeth_query_ipassists);
 
+static int qeth_query_switch_port_attributes_cb(struct qeth_card *card,
+				struct qeth_reply *reply, unsigned long data)
+{
+	struct qeth_ipa_cmd *cmd;
+	struct qeth_switch_info *sw_info;
+	struct qeth_query_switch_attributes *attrs;
+
+	QETH_CARD_TEXT(card, 2, "qsqiatcb");
+	cmd = (struct qeth_ipa_cmd *) data;
+	sw_info = (struct qeth_switch_info *)reply->param;
+	if (cmd->data.setadapterparms.hdr.return_code == 0) {
+		attrs = &cmd->data.setadapterparms.data.query_switch_attributes;
+		sw_info->capabilities = attrs->capabilities;
+		sw_info->settings = attrs->settings;
+	}
+	qeth_default_setadapterparms_cb(card, reply, (unsigned long) cmd);
+
+	return 0;
+}
+
+static int qeth_query_switch_port_attributes(struct qeth_card *card,
+					struct qeth_switch_info *sw_info)
+{
+	struct qeth_cmd_buffer *iob;
+
+	QETH_CARD_TEXT(card, 2, "qswiattr");
+	if (!qeth_adp_supported(card, IPA_SETADP_QUERY_SWITCH_ATTRIBUTES))
+		return -EOPNOTSUPP;
+	if (!netif_carrier_ok(card->dev))
+		return -ENOMEDIUM;
+	iob = qeth_get_adapter_cmd(card, IPA_SETADP_QUERY_SWITCH_ATTRIBUTES,
+				sizeof(struct qeth_ipacmd_setadpparms_hdr));
+	return qeth_send_ipa_cmd(card, iob,
+				qeth_query_switch_port_attributes_cb, sw_info);
+}
+
 static int qeth_query_setdiagass_cb(struct qeth_card *card,
 		struct qeth_reply *reply, unsigned long data)
 {
@@ -5696,6 +5732,49 @@ int qeth_core_ethtool_get_settings(struct net_device *netdev,
 }
 EXPORT_SYMBOL_GPL(qeth_core_ethtool_get_settings);
 
+int qeth_core_ethtool_get_switch_port_attrs(struct net_device *netdev,
+					struct ethtool_swport_attrs *attrs)
+{
+	int rc, i;
+	struct qeth_card *card = netdev->ml_priv;
+	struct qeth_switch_info sw_info;
+	static const struct {
+		int qeth_cap;
+		int etht_cap_sup;
+		int etht_cap_enbl;
+	} qattrs[] = {
+		{ QETH_SWITCH_FORW_802_1, SUPPORTED_SP_FWD_802_1,
+		  ENABLED_SP_FWD_802_1 },
+		{ QETH_SWITCH_FORW_REFL_RELAY, SUPPORTED_SP_FWD_RR,
+		  ENABLED_SP_FWD_RR },
+		{ QETH_SWITCH_CAP_RTE, SUPPORTED_SP_CAP_RTE,
+		  ENABLED_SP_CAP_RTE },
+		{ QETH_SWITCH_CAP_ECP, SUPPORTED_SP_CAP_ECP,
+		  ENABLED_SP_CAP_ECP },
+		{ QETH_SWITCH_CAP_VDP, SUPPORTED_SP_CAP_VDP,
+		  ENABLED_SP_CAP_VDP }
+	};
+
+	rc = qeth_query_switch_port_attributes(card, &sw_info);
+	if (rc)
+		return rc;
+
+	if (!sw_info.capabilities) {
+		attrs->port_rc = GPORT_RC_LLDP_UNSUP;
+		return 0;
+	}
+	for (i = 0; i < ARRAY_SIZE(qattrs); i++) {
+		if (sw_info.capabilities & qattrs[i].qeth_cap) {
+			attrs->supported |= qattrs[i].etht_cap_sup;
+			if (sw_info.settings & qattrs[i].qeth_cap)
+				attrs->enabled |= qattrs[i].etht_cap_enbl;
+		}
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(qeth_core_ethtool_get_switch_port_attrs);
+
 static int __init qeth_core_init(void)
 {
 	int rc;
diff --git a/drivers/s390/net/qeth_core_mpc.h b/drivers/s390/net/qeth_core_mpc.h
index 07085d5..9f9b034 100644
--- a/drivers/s390/net/qeth_core_mpc.h
+++ b/drivers/s390/net/qeth_core_mpc.h
@@ -240,6 +240,7 @@ enum qeth_ipa_setadp_cmd {
 	IPA_SETADP_SET_DIAG_ASSIST		= 0x00002000L,
 	IPA_SETADP_SET_ACCESS_CONTROL		= 0x00010000L,
 	IPA_SETADP_QUERY_OAT			= 0x00080000L,
+	IPA_SETADP_QUERY_SWITCH_ATTRIBUTES	= 0x00100000L,
 };
 enum qeth_ipa_mac_ops {
 	CHANGE_ADDR_READ_MAC		= 0,
@@ -404,6 +405,21 @@ struct qeth_qoat_priv {
 	char *buffer;
 };
 
+#define QETH_SWITCH_FORW_802_1		0x00000001
+#define QETH_SWITCH_FORW_REFL_RELAY	0x00000002
+#define QETH_SWITCH_CAP_RTE		0x00000004
+#define QETH_SWITCH_CAP_ECP		0x00000008
+#define QETH_SWITCH_CAP_VDP		0x00000010
+
+struct qeth_query_switch_attributes {
+	__u8  version;
+	__u8  reserved1;
+	__u16 reserved2;
+	__u32 capabilities;
+	__u32 settings;
+	__u8  reserved3[8];
+};
+
 struct qeth_ipacmd_setadpparms_hdr {
 	__u32 supp_hw_cmds;
 	__u32 reserved1;
@@ -424,6 +440,7 @@ struct qeth_ipacmd_setadpparms {
 		struct qeth_snmp_cmd snmp;
 		struct qeth_set_access_ctrl set_access_ctrl;
 		struct qeth_query_oat query_oat;
+		struct qeth_query_switch_attributes query_switch_attributes;
 		__u32 mode;
 	} data;
 } __attribute__ ((packed));
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index ec8ccda..ae4de33 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -911,6 +911,7 @@ static const struct ethtool_ops qeth_l2_ethtool_ops = {
 	.get_sset_count = qeth_core_get_sset_count,
 	.get_drvinfo = qeth_core_get_drvinfo,
 	.get_settings = qeth_core_ethtool_get_settings,
+	.get_switch_port_attrs = qeth_core_ethtool_get_switch_port_attrs,
 };
 
 static const struct ethtool_ops qeth_l2_osn_ops = {
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index c1b0b27..356076a 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -3198,6 +3198,7 @@ static const struct ethtool_ops qeth_l3_ethtool_ops = {
 	.get_sset_count = qeth_core_get_sset_count,
 	.get_drvinfo = qeth_core_get_drvinfo,
 	.get_settings = qeth_core_ethtool_get_settings,
+	.get_switch_port_attrs = qeth_core_ethtool_get_switch_port_attrs,
 };
 
 /*
-- 
1.8.3.4

Re: [PATCH 0/2] Display adjacent switch port's attributes

[Stephen Hemminger]

On Wed, 11 Dec 2013 14:28:58 +0100
Stefan Raspl <raspl@linux.vnet.ibm.com> wrote:

> This patch series adds a new callback for ethtool to display the adjacent switch
> port's attributes, as perceived by the NIC, e.g. through respective LLDP message
> exchanges.
> A patch for the qeth device driver provides a sample exploiter for the new
> functionality.
> A new option for ethtool to display the settings will be posted in a separate
> patch.
> 
> Best regards,
> Stefan Raspl
> 
> Stefan Raspl (2):
>   ethtool: Add callback to indicate adjacent switch port attributes
>   qeth: Display adjacent switch port attributes in ethtool
> 
>  drivers/s390/net/qeth_core.h      |  7 ++++
>  drivers/s390/net/qeth_core_main.c | 79 +++++++++++++++++++++++++++++++++++++++
>  drivers/s390/net/qeth_core_mpc.h  | 17 +++++++++
>  drivers/s390/net/qeth_l2_main.c   |  1 +
>  drivers/s390/net/qeth_l3_main.c   |  1 +
>  include/linux/ethtool.h           |  3 ++
>  include/uapi/linux/ethtool.h      | 35 +++++++++++++++++
>  net/core/ethtool.c                | 22 +++++++++++
>  8 files changed, 165 insertions(+)
> 

Why more ethtool and not netlink? ethtool is a brittle interface, non-extensible
and does not support notifications of changes.

Re: [PATCH 0/2] Display adjacent switch port's attributes

[Nicolas Dichtel]

Le 11/12/2013 21:13, Stephen Hemminger a écrit :
> On Wed, 11 Dec 2013 14:28:58 +0100
> Stefan Raspl <raspl@linux.vnet.ibm.com> wrote:
>
>> This patch series adds a new callback for ethtool to display the adjacent switch
>> port's attributes, as perceived by the NIC, e.g. through respective LLDP message
>> exchanges.
>> A patch for the qeth device driver provides a sample exploiter for the new
>> functionality.
>> A new option for ethtool to display the settings will be posted in a separate
>> patch.
>>
>> Best regards,
>> Stefan Raspl
>>
>> Stefan Raspl (2):
>>    ethtool: Add callback to indicate adjacent switch port attributes
>>    qeth: Display adjacent switch port attributes in ethtool
>>
>>   drivers/s390/net/qeth_core.h      |  7 ++++
>>   drivers/s390/net/qeth_core_main.c | 79 +++++++++++++++++++++++++++++++++++++++
>>   drivers/s390/net/qeth_core_mpc.h  | 17 +++++++++
>>   drivers/s390/net/qeth_l2_main.c   |  1 +
>>   drivers/s390/net/qeth_l3_main.c   |  1 +
>>   include/linux/ethtool.h           |  3 ++
>>   include/uapi/linux/ethtool.h      | 35 +++++++++++++++++
>>   net/core/ethtool.c                | 22 +++++++++++
>>   8 files changed, 165 insertions(+)
>>
>
> Why more ethtool and not netlink? ethtool is a brittle interface, non-extensible
> and does not support notifications of changes.
I agree with Stephen, netlink has more pros.

Re: [PATCH 0/2] Display adjacent switch port's attributes

[Stefan Raspl]

Am 11.12.2013 21:13, schrieb Stephen Hemminger:
> On Wed, 11 Dec 2013 14:28:58 +0100
> Stefan Raspl <raspl@linux.vnet.ibm.com> wrote:
> 
>> This patch series adds a new callback for ethtool to display the adjacent switch
>> port's attributes, as perceived by the NIC, e.g. through respective LLDP message
>> exchanges.
>> A patch for the qeth device driver provides a sample exploiter for the new
>> functionality.
>> A new option for ethtool to display the settings will be posted in a separate
>> patch.
>>
>> Best regards,
>> Stefan Raspl
>>
>> Stefan Raspl (2):
>>   ethtool: Add callback to indicate adjacent switch port attributes
>>   qeth: Display adjacent switch port attributes in ethtool
>>
>>  drivers/s390/net/qeth_core.h      |  7 ++++
>>  drivers/s390/net/qeth_core_main.c | 79 +++++++++++++++++++++++++++++++++++++++
>>  drivers/s390/net/qeth_core_mpc.h  | 17 +++++++++
>>  drivers/s390/net/qeth_l2_main.c   |  1 +
>>  drivers/s390/net/qeth_l3_main.c   |  1 +
>>  include/linux/ethtool.h           |  3 ++
>>  include/uapi/linux/ethtool.h      | 35 +++++++++++++++++
>>  net/core/ethtool.c                | 22 +++++++++++
>>  8 files changed, 165 insertions(+)
>>
> 
> Why more ethtool and not netlink? ethtool is a brittle interface, non-extensible
> and does not support notifications of changes.
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

I agree, netlink would certainly be nice. But ethtool is the
de-facto standard for now, and there doesn't seem to be a
netlink-based alternative in sight - or is there?
Offering a netlink based tool seems to be a different discussion,
and if that takes shape, existing ethtool functionality (including
this series) can and should be migrated.

Re: [PATCH 0/2] Display adjacent switch port's attributes

[David Miller]

From: Stefan Raspl <raspl@linux.vnet.ibm.com>
Date: Thu, 12 Dec 2013 14:57:18 +0100

> I agree, netlink would certainly be nice. But ethtool is the
> de-facto standard for now, and there doesn't seem to be a
> netlink-based alternative in sight - or is there?
> Offering a netlink based tool seems to be a different discussion,
> and if that takes shape, existing ethtool functionality (including
> this series) can and should be migrated.

I completely disagree, ethtool is not de-facto for anything in
particular.  It's suitable for some things, not suitable for
others.

And just because it's used for other aspects of a device's
configuration doesn't mean that netlink isn't appropriate for other
apsects.

Re: [PATCH 0/2] Display adjacent switch port's attributes

[John Fastabend]

On 12/12/2013 9:28 AM, David Miller wrote:
> From: Stefan Raspl <raspl@linux.vnet.ibm.com>
> Date: Thu, 12 Dec 2013 14:57:18 +0100
>
>> I agree, netlink would certainly be nice. But ethtool is the
>> de-facto standard for now, and there doesn't seem to be a
>> netlink-based alternative in sight - or is there?
>> Offering a netlink based tool seems to be a different discussion,
>> and if that takes shape, existing ethtool functionality (including
>> this series) can and should be migrated.
>
> I completely disagree, ethtool is not de-facto for anything in
> particular.  It's suitable for some things, not suitable for
> others.
>
> And just because it's used for other aspects of a device's
> configuration doesn't mean that netlink isn't appropriate for other
> apsects.

Just to elaborate...

Any application using lldp information will want to get events when
TLVs change. Maybe you can contrive ethtool to do this but its going to
be ugly. Netlink can support multicast events and applications can
register for them. Also netlink's TLV format matches nicely with LLDPs
TLV format.

If you push an ethtool interface now we get stuck with an interface
that is only useful in a very narrow use case. And netlink interfaces
are relatively easy to construct so lets do this right the first time.
I also happen to maintain lldpad and would be happy to plug in support
for this via netlink so we can push firmware and software based agent
LLDP info up to libvirt/network manager whatever in a consistent way.
Similarly other existing lldp agents could do the same.

Thanks,
.John

Re: [PATCH 0/2] Display adjacent switch port's attributes

[Florian Fainelli]

2013/12/12 John Fastabend <john.r.fastabend@intel.com>:
> On 12/12/2013 9:28 AM, David Miller wrote:
>>
>> From: Stefan Raspl <raspl@linux.vnet.ibm.com>
>> Date: Thu, 12 Dec 2013 14:57:18 +0100
>>
>>> I agree, netlink would certainly be nice. But ethtool is the
>>> de-facto standard for now, and there doesn't seem to be a
>>> netlink-based alternative in sight - or is there?
>>> Offering a netlink based tool seems to be a different discussion,
>>> and if that takes shape, existing ethtool functionality (including
>>> this series) can and should be migrated.
>>
>>
>> I completely disagree, ethtool is not de-facto for anything in
>> particular.  It's suitable for some things, not suitable for
>> others.
>>
>> And just because it's used for other aspects of a device's
>> configuration doesn't mean that netlink isn't appropriate for other
>> apsects.
>
>
> Just to elaborate...
>
> Any application using lldp information will want to get events when
> TLVs change. Maybe you can contrive ethtool to do this but its going to
> be ugly. Netlink can support multicast events and applications can
> register for them. Also netlink's TLV format matches nicely with LLDPs
> TLV format.

ethtool and netlink usually intersect for a few bits of information
such as link status for instance. It is useful to have this
information twice, with ethtool as a debugging aid, and via netlink to
take appropriate actions.

Maybe we just need to be clear on what needs to be present in ethtool
only (configuration, static information) and see on a case-by-case
what needs to be present in both ethtool and netlink?

>
> If you push an ethtool interface now we get stuck with an interface
> that is only useful in a very narrow use case. And netlink interfaces
> are relatively easy to construct so lets do this right the first time.
> I also happen to maintain lldpad and would be happy to plug in support
> for this via netlink so we can push firmware and software based agent
> LLDP info up to libvirt/network manager whatever in a consistent way.
> Similarly other existing lldp agents could do the same.
>
> Thanks,
> .John
-- 
Florian

Re: [PATCH 0/2] Display adjacent switch port's attributes

[John Fastabend]

[...]

>> Just to elaborate...
>>
>> Any application using lldp information will want to get events when
>> TLVs change. Maybe you can contrive ethtool to do this but its going to
>> be ugly. Netlink can support multicast events and applications can
>> register for them. Also netlink's TLV format matches nicely with LLDPs
>> TLV format.
>
> ethtool and netlink usually intersect for a few bits of information
> such as link status for instance. It is useful to have this
> information twice, with ethtool as a debugging aid, and via netlink to
> take appropriate actions.
>
> Maybe we just need to be clear on what needs to be present in ethtool
> only (configuration, static information) and see on a case-by-case
> what needs to be present in both ethtool and netlink?
>

OK if there is an enable/disable bit in ethtool that might make some
sense. Or an error flag that is helpful to have.

In this case we are dealing with peer attributes which are dynamic and
in my opinion should go into netlink and duplicating them in ethtool
although possible doesn't seem very useful to me.

.John

Re: [PATCH 0/2] Display adjacent switch port's attributes

[Ben Hutchings]

On Thu, 2013-12-12 at 13:47 -0800, John Fastabend wrote:
> [...]
> 
> >> Just to elaborate...
> >>
> >> Any application using lldp information will want to get events when
> >> TLVs change. Maybe you can contrive ethtool to do this but its going to
> >> be ugly. Netlink can support multicast events and applications can
> >> register for them. Also netlink's TLV format matches nicely with LLDPs
> >> TLV format.
> >
> > ethtool and netlink usually intersect for a few bits of information
> > such as link status for instance. It is useful to have this
> > information twice, with ethtool as a debugging aid, and via netlink to
> > take appropriate actions.
> >
> > Maybe we just need to be clear on what needs to be present in ethtool
> > only (configuration, static information) and see on a case-by-case
> > what needs to be present in both ethtool and netlink?
> >
> 
> OK if there is an enable/disable bit in ethtool that might make some
> sense. Or an error flag that is helpful to have.
> 
> In this case we are dealing with peer attributes which are dynamic and
> in my opinion should go into netlink and duplicating them in ethtool
> although possible doesn't seem very useful to me.

I agree with this - rtnetlink seems to make more sense.

Ben.

-- 
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.

Re: [PATCH 0/2] Display adjacent switch port's attributes

[Ben Hutchings]

On Thu, 2013-12-12 at 13:47 -0800, John Fastabend wrote:
> [...]
> 
> >> Just to elaborate...
> >>
> >> Any application using lldp information will want to get events when
> >> TLVs change. Maybe you can contrive ethtool to do this but its going to
> >> be ugly. Netlink can support multicast events and applications can
> >> register for them. Also netlink's TLV format matches nicely with LLDPs
> >> TLV format.
> >
> > ethtool and netlink usually intersect for a few bits of information
> > such as link status for instance. It is useful to have this
> > information twice, with ethtool as a debugging aid, and via netlink to
> > take appropriate actions.
> >
> > Maybe we just need to be clear on what needs to be present in ethtool
> > only (configuration, static information) and see on a case-by-case
> > what needs to be present in both ethtool and netlink?
> >
> 
> OK if there is an enable/disable bit in ethtool that might make some
> sense. Or an error flag that is helpful to have.
> 
> In this case we are dealing with peer attributes which are dynamic and
> in my opinion should go into netlink and duplicating them in ethtool
> although possible doesn't seem very useful to me.

I agree with this - rtnetlink seems to make more sense.

Ben.

-- 
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.

Re: [PATCH 0/2] Display adjacent switch port's attributes

[Stefan Raspl]

Am 12.12.2013 22:47, schrieb John Fastabend:
> [...]
> 
>>> Just to elaborate...
>>>
>>> Any application using lldp information will want to get events when
>>> TLVs change. Maybe you can contrive ethtool to do this but its
>>> going to
>>> be ugly. Netlink can support multicast events and applications can
>>> register for them. Also netlink's TLV format matches nicely with
>>> LLDPs
>>> TLV format.
>>
>> ethtool and netlink usually intersect for a few bits of information
>> such as link status for instance. It is useful to have this
>> information twice, with ethtool as a debugging aid, and via
>> netlink to
>> take appropriate actions.
>>
>> Maybe we just need to be clear on what needs to be present in ethtool
>> only (configuration, static information) and see on a case-by-case
>> what needs to be present in both ethtool and netlink?
>>
> 
> OK if there is an enable/disable bit in ethtool that might make some
> sense. Or an error flag that is helpful to have.
> 
> In this case we are dealing with peer attributes which are dynamic and
> in my opinion should go into netlink and duplicating them in ethtool
> although possible doesn't seem very useful to me.

I think what most folks in the discussion so far assume is that we
have a full LLDP implementation with respective hooks and events.
This is not true for our device: We can only poll it for the
adjacent link port's state, and that's it. I.e. we don't receive any
events for changes on the port's state.
lldpad seems to handle the entire LLDP layer in software. And I'm
not sure if our device integrates with that so well, as it handles
LLDP on its own. We'd have to disable pretty much all functionality
that lldpad offers, and limit support to display of a few parameters
which we would have to poll on demand.
Likewise with netlink: If one of the arguments for netlink is that
it supports notifications, then we can't take any advantage of that
either, for the reasons stated above.
By its nature, what we can offer and support with our device is
simple debugging functionality, displaying the current state of the
switch port at a given moment - just like ethtool will display the
current port speed and media type. Hence the original idea to add
respective functionality to ethtool in a generic manner, so others
with similar constraints could use it as well.
If ethtool is not acceptable, and if lldpad and netlink are no good
fits either, would respective sysfs attributes for our device type work?

Stefan

Re: [PATCH 0/2] Display adjacent switch port's attributes

[Stefan Raspl]

On 12/16/2013 04:32 PM, Stefan Raspl wrote:
> Am 12.12.2013 22:47, schrieb John Fastabend:
>> [...]
>>
>>>> Just to elaborate...
>>>>
>>>> Any application using lldp information will want to get events when
>>>> TLVs change. Maybe you can contrive ethtool to do this but its
>>>> going to
>>>> be ugly. Netlink can support multicast events and applications can
>>>> register for them. Also netlink's TLV format matches nicely with
>>>> LLDPs
>>>> TLV format.
>>>
>>> ethtool and netlink usually intersect for a few bits of information
>>> such as link status for instance. It is useful to have this
>>> information twice, with ethtool as a debugging aid, and via
>>> netlink to
>>> take appropriate actions.
>>>
>>> Maybe we just need to be clear on what needs to be present in ethtool
>>> only (configuration, static information) and see on a case-by-case
>>> what needs to be present in both ethtool and netlink?
>>>
>>
>> OK if there is an enable/disable bit in ethtool that might make some
>> sense. Or an error flag that is helpful to have.
>>
>> In this case we are dealing with peer attributes which are dynamic and
>> in my opinion should go into netlink and duplicating them in ethtool
>> although possible doesn't seem very useful to me.
> 
> I think what most folks in the discussion so far assume is that we
> have a full LLDP implementation with respective hooks and events.
> This is not true for our device: We can only poll it for the
> adjacent link port's state, and that's it. I.e. we don't receive any
> events for changes on the port's state.
> lldpad seems to handle the entire LLDP layer in software. And I'm
> not sure if our device integrates with that so well, as it handles
> LLDP on its own. We'd have to disable pretty much all functionality
> that lldpad offers, and limit support to display of a few parameters
> which we would have to poll on demand.
> Likewise with netlink: If one of the arguments for netlink is that
> it supports notifications, then we can't take any advantage of that
> either, for the reasons stated above.
> By its nature, what we can offer and support with our device is
> simple debugging functionality, displaying the current state of the
> switch port at a given moment - just like ethtool will display the
> current port speed and media type. Hence the original idea to add
> respective functionality to ethtool in a generic manner, so others
> with similar constraints could use it as well.
> If ethtool is not acceptable, and if lldpad and netlink are no good
> fits either, would respective sysfs attributes for our device type work?

Since I didn't receive any further feedback, please let me summarize:

* As elaborated in my most recent reply (cited above), our device
  does not provide for any notifications regarding LLDP-related
  events - we can only query the current state. Hence we could not
  take advantage of a netlink interface. Plus even if we still went
  with netlink, we'd have to introduce yet another userspace tool
  just for querying the current state.
* For the same reason, lldpad would be hard to integrate with,
  since all we can do is to query a limited amount of information,
  where lldpad seems to be targeted at devices that can provide
  events.
* Integration with ethtool was our attempt at providing a common
  interface for our and other devices with similar characteristics
  regarding LLDP, since ethtool is semantically a good fit.
  However, it was indicated that this is not desirable.
* It seems that the only choice left is to implement sysfs
  attributes to query LLDP-related attributes. Any other device
  with similar characteristics would probably need to re-do the
  same functionality independently. Is this really what we want to
  do?

Please let me know what you think.

Regards,
Stefan

Re: [PATCH 0/2] Display adjacent switch port's attributes

[Stefan Raspl]

On 12/16/2013 04:32 PM, Stefan Raspl wrote:
> Am 12.12.2013 22:47, schrieb John Fastabend:
>> [...]
>>
>>>> Just to elaborate...
>>>>
>>>> Any application using lldp information will want to get events when
>>>> TLVs change. Maybe you can contrive ethtool to do this but its
>>>> going to
>>>> be ugly. Netlink can support multicast events and applications can
>>>> register for them. Also netlink's TLV format matches nicely with
>>>> LLDPs
>>>> TLV format.
>>>
>>> ethtool and netlink usually intersect for a few bits of information
>>> such as link status for instance. It is useful to have this
>>> information twice, with ethtool as a debugging aid, and via
>>> netlink to
>>> take appropriate actions.
>>>
>>> Maybe we just need to be clear on what needs to be present in ethtool
>>> only (configuration, static information) and see on a case-by-case
>>> what needs to be present in both ethtool and netlink?
>>>
>>
>> OK if there is an enable/disable bit in ethtool that might make some
>> sense. Or an error flag that is helpful to have.
>>
>> In this case we are dealing with peer attributes which are dynamic and
>> in my opinion should go into netlink and duplicating them in ethtool
>> although possible doesn't seem very useful to me.
> 
> I think what most folks in the discussion so far assume is that we
> have a full LLDP implementation with respective hooks and events.
> This is not true for our device: We can only poll it for the
> adjacent link port's state, and that's it. I.e. we don't receive any
> events for changes on the port's state.
> lldpad seems to handle the entire LLDP layer in software. And I'm
> not sure if our device integrates with that so well, as it handles
> LLDP on its own. We'd have to disable pretty much all functionality
> that lldpad offers, and limit support to display of a few parameters
> which we would have to poll on demand.
> Likewise with netlink: If one of the arguments for netlink is that
> it supports notifications, then we can't take any advantage of that
> either, for the reasons stated above.
> By its nature, what we can offer and support with our device is
> simple debugging functionality, displaying the current state of the
> switch port at a given moment - just like ethtool will display the
> current port speed and media type. Hence the original idea to add
> respective functionality to ethtool in a generic manner, so others
> with similar constraints could use it as well.
> If ethtool is not acceptable, and if lldpad and netlink are no good
> fits either, would respective sysfs attributes for our device type work?

Since I didn't receive any further feedback, please let me summarize:

* As elaborated in my most recent reply (cited above), our device
  does not provide for any notifications regarding LLDP-related
  events - we can only query the current state. Hence we could not
  take advantage of a netlink interface. Plus even if we still went
  with netlink, we'd have to introduce yet another userspace tool
  just for querying the current state.
* For the same reason, lldpad would be hard to integrate with,
  since all we can do is to query a limited amount of information,
  where lldpad seems to be targeted at devices that can provide
  events.
* Integration with ethtool was our attempt at providing a common
  interface for our and other devices with similar characteristics
  regarding LLDP, since ethtool is semantically a good fit.
  However, it was indicated that this is not desirable.
* It seems that the only choice left is to implement sysfs
  attributes to query LLDP-related attributes. Any other device
  with similar characteristics would probably need to re-do the
  same functionality independently. Is this really what we want to
  do?

Please let me know what you think.

Regards,
Stefan