add devlink tool

Add new tool called devlink which is userspace counterpart of devlink
Netlink socket.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>

Makefile

@@ -41,7 +41,7 @@ WFLAGS += -Wmissing-declarations -Wold-style-definition -Wformat=2
 CFLAGS := $(WFLAGS) $(CCOPTS) -I../include $(DEFINES) $(CFLAGS)
 YACCFLAGS = -d -t -v
 
-SUBDIRS=lib ip tc bridge misc netem genl tipc man
+SUBDIRS=lib ip tc bridge misc netem genl tipc devlink man
 
 LIBNETLINK=../lib/libnetlink.a ../lib/libutil.a
 LDLIBS += $(LIBNETLINK)

devlink/Makefile

+include ../Config
+ifeq ($(HAVE_MNL),y)
+
+DEVLINKOBJ = devlink.o mnlg.o
+TARGETS=devlink
+
+CFLAGS += $(shell $(PKG_CONFIG) libmnl --cflags)
+LDLIBS += $(shell $(PKG_CONFIG) libmnl --libs)
+
+endif
+
+all: $(TARGETS) $(LIBS)
+
+devlink: $(DEVLINKOBJ)
+
+install: all
+	install -m 0755 $(TARGETS) $(DESTDIR)$(SBINDIR)
+
+clean:
+	rm -f $(DEVLINKOBJ) $(TARGETS)

devlink/mnlg.c

+/*
+ *   mnlg.c	Generic Netlink helpers for libmnl
+ *
+ *              This program is free software; you can redistribute it and/or
+ *              modify it under the terms of the GNU General Public License
+ *              as published by the Free Software Foundation; either version
+ *              2 of the License, or (at your option) any later version.
+ *
+ * Authors:     Jiri Pirko <jiri@mellanox.com>
+ */
+
+#include <stdlib.h>
+#include <stdbool.h>
+#include <string.h>
+#include <errno.h>
+#include <unistd.h>
+#include <time.h>
+#include <libmnl/libmnl.h>
+#include <linux/genetlink.h>
+
+#include "mnlg.h"
+
+struct mnlg_socket {
+	struct mnl_socket *nl;
+	char *buf;
+	uint32_t id;
+	uint8_t version;
+	unsigned int seq;
+	unsigned int portid;
+};
+
+static struct nlmsghdr *__mnlg_msg_prepare(struct mnlg_socket *nlg, uint8_t cmd,
+					   uint16_t flags, uint32_t id,
+					   uint8_t version)
+{
+	struct nlmsghdr *nlh;
+	struct genlmsghdr *genl;
+
+	nlh = mnl_nlmsg_put_header(nlg->buf);
+	nlh->nlmsg_type	= id;
+	nlh->nlmsg_flags = flags;
+	nlg->seq = time(NULL);
+	nlh->nlmsg_seq = nlg->seq;
+
+	genl = mnl_nlmsg_put_extra_header(nlh, sizeof(struct genlmsghdr));
+	genl->cmd = cmd;
+	genl->version = version;
+
+	return nlh;
+}
+
+struct nlmsghdr *mnlg_msg_prepare(struct mnlg_socket *nlg, uint8_t cmd,
+				  uint16_t flags)
+{
+	return __mnlg_msg_prepare(nlg, cmd, flags, nlg->id, nlg->version);
+}
+
+int mnlg_socket_send(struct mnlg_socket *nlg, const struct nlmsghdr *nlh)
+{
+	return mnl_socket_sendto(nlg->nl, nlh, nlh->nlmsg_len);
+}
+
+int mnlg_socket_recv_run(struct mnlg_socket *nlg, mnl_cb_t data_cb, void *data)
+{
+	int err;
+
+	do {
+		err = mnl_socket_recvfrom(nlg->nl, nlg->buf,
+					  MNL_SOCKET_BUFFER_SIZE);
+		if (err <= 0)
+			break;
+		err = mnl_cb_run(nlg->buf, err, nlg->seq, nlg->portid,
+				 data_cb, data);
+	} while (err > 0);
+
+	return err;
+}
+
+struct group_info {
+	bool found;
+	uint32_t id;
+	const char *name;
+};
+
+static int parse_mc_grps_cb(const struct nlattr *attr, void *data)
+{
+	const struct nlattr **tb = data;
+	int type = mnl_attr_get_type(attr);
+
+	if (mnl_attr_type_valid(attr, CTRL_ATTR_MCAST_GRP_MAX) < 0)
+		return MNL_CB_OK;
+
+	switch (type) {
+	case CTRL_ATTR_MCAST_GRP_ID:
+		if (mnl_attr_validate(attr, MNL_TYPE_U32) < 0)
+			return MNL_CB_ERROR;
+		break;
+	case CTRL_ATTR_MCAST_GRP_NAME:
+		if (mnl_attr_validate(attr, MNL_TYPE_STRING) < 0)
+			return MNL_CB_ERROR;
+		break;
+	}
+	tb[type] = attr;
+	return MNL_CB_OK;
+}
+
+static void parse_genl_mc_grps(struct nlattr *nested,
+			       struct group_info *group_info)
+{
+	struct nlattr *pos;
+	const char *name;
+
+	mnl_attr_for_each_nested(pos, nested) {
+		struct nlattr *tb[CTRL_ATTR_MCAST_GRP_MAX + 1] = {};
+
+		mnl_attr_parse_nested(pos, parse_mc_grps_cb, tb);
+		if (!tb[CTRL_ATTR_MCAST_GRP_NAME] ||
+		    !tb[CTRL_ATTR_MCAST_GRP_ID])
+			continue;
+
+		name = mnl_attr_get_str(tb[CTRL_ATTR_MCAST_GRP_NAME]);
+		if (strcmp(name, group_info->name) != 0)
+			continue;
+
+		group_info->id = mnl_attr_get_u32(tb[CTRL_ATTR_MCAST_GRP_ID]);
+		group_info->found = true;
+	}
+}
+
+static int get_group_id_attr_cb(const struct nlattr *attr, void *data)
+{
+	const struct nlattr **tb = data;
+	int type = mnl_attr_get_type(attr);
+
+	if (mnl_attr_type_valid(attr, CTRL_ATTR_MAX) < 0)
+		return MNL_CB_ERROR;
+
+	if (type == CTRL_ATTR_MCAST_GROUPS &&
+	    mnl_attr_validate(attr, MNL_TYPE_NESTED) < 0)
+		return MNL_CB_ERROR;
+	tb[type] = attr;
+	return MNL_CB_OK;
+}
+
+static int get_group_id_cb(const struct nlmsghdr *nlh, void *data)
+{
+	struct group_info *group_info = data;
+	struct nlattr *tb[CTRL_ATTR_MAX + 1] = {};
+	struct genlmsghdr *genl = mnl_nlmsg_get_payload(nlh);
+
+	mnl_attr_parse(nlh, sizeof(*genl), get_group_id_attr_cb, tb);
+	if (!tb[CTRL_ATTR_MCAST_GROUPS])
+		return MNL_CB_ERROR;
+	parse_genl_mc_grps(tb[CTRL_ATTR_MCAST_GROUPS], group_info);
+	return MNL_CB_OK;
+}
+
+int mnlg_socket_group_add(struct mnlg_socket *nlg, const char *group_name)
+{
+	struct nlmsghdr *nlh;
+	struct group_info group_info;
+	int err;
+
+	nlh = __mnlg_msg_prepare(nlg, CTRL_CMD_GETFAMILY,
+				 NLM_F_REQUEST | NLM_F_ACK, GENL_ID_CTRL, 1);
+	mnl_attr_put_u32(nlh, CTRL_ATTR_FAMILY_ID, nlg->id);
+
+	err = mnlg_socket_send(nlg, nlh);
+	if (err < 0)
+		return err;
+
+	group_info.found = false;
+	group_info.name = group_name;
+	err = mnlg_socket_recv_run(nlg, get_group_id_cb, &group_info);
+	if (err < 0)
+		return err;
+
+	if (!group_info.found) {
+		errno = ENOENT;
+		return -1;
+	}
+
+	err = mnl_socket_setsockopt(nlg->nl, NETLINK_ADD_MEMBERSHIP,
+				    &group_info.id, sizeof(group_info.id));
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
+static int get_family_id_attr_cb(const struct nlattr *attr, void *data)
+{
+	const struct nlattr **tb = data;
+	int type = mnl_attr_get_type(attr);
+
+	if (mnl_attr_type_valid(attr, CTRL_ATTR_MAX) < 0)
+		return MNL_CB_ERROR;
+
+	if (type == CTRL_ATTR_FAMILY_ID &&
+	    mnl_attr_validate(attr, MNL_TYPE_U16) < 0)
+		return MNL_CB_ERROR;
+	tb[type] = attr;
+	return MNL_CB_OK;
+}
+
+static int get_family_id_cb(const struct nlmsghdr *nlh, void *data)
+{
+	uint32_t *p_id = data;
+	struct nlattr *tb[CTRL_ATTR_MAX + 1] = {};
+	struct genlmsghdr *genl = mnl_nlmsg_get_payload(nlh);
+
+	mnl_attr_parse(nlh, sizeof(*genl), get_family_id_attr_cb, tb);
+	if (!tb[CTRL_ATTR_FAMILY_ID])
+		return MNL_CB_ERROR;
+	*p_id = mnl_attr_get_u16(tb[CTRL_ATTR_FAMILY_ID]);
+	return MNL_CB_OK;
+}
+
+struct mnlg_socket *mnlg_socket_open(const char *family_name, uint8_t version)
+{
+	struct mnlg_socket *nlg;
+	struct nlmsghdr *nlh;
+	int err;
+
+	nlg = malloc(sizeof(*nlg));
+	if (!nlg)
+		return NULL;
+
+	nlg->buf = malloc(MNL_SOCKET_BUFFER_SIZE);
+	if (!nlg->buf)
+		goto err_buf_alloc;
+
+	nlg->nl = mnl_socket_open(NETLINK_GENERIC);
+	if (!nlg->nl)
+		goto err_mnl_socket_open;
+
+	err = mnl_socket_bind(nlg->nl, 0, MNL_SOCKET_AUTOPID);
+	if (err < 0)
+		goto err_mnl_socket_bind;
+
+	nlg->portid = mnl_socket_get_portid(nlg->nl);
+
+	nlh = __mnlg_msg_prepare(nlg, CTRL_CMD_GETFAMILY,
+				 NLM_F_REQUEST | NLM_F_ACK, GENL_ID_CTRL, 1);
+	mnl_attr_put_strz(nlh, CTRL_ATTR_FAMILY_NAME, family_name);
+
+	err = mnlg_socket_send(nlg, nlh);
+	if (err < 0)
+		goto err_mnlg_socket_send;
+
+	err = mnlg_socket_recv_run(nlg, get_family_id_cb, &nlg->id);
+	if (err < 0)
+		goto err_mnlg_socket_recv_run;
+
+	nlg->version = version;
+	return nlg;
+
+err_mnlg_socket_recv_run:
+err_mnlg_socket_send:
+err_mnl_socket_bind:
+	mnl_socket_close(nlg->nl);
+err_mnl_socket_open:
+	free(nlg->buf);
+err_buf_alloc:
+	free(nlg);
+	return NULL;
+}
+
+void mnlg_socket_close(struct mnlg_socket *nlg)
+{
+	mnl_socket_close(nlg->nl);
+	free(nlg->buf);
+	free(nlg);
+}

devlink/mnlg.h

+/*
+ *   mnlg.h	Generic Netlink helpers for libmnl
+ *
+ *              This program is free software; you can redistribute it and/or
+ *              modify it under the terms of the GNU General Public License
+ *              as published by the Free Software Foundation; either version
+ *              2 of the License, or (at your option) any later version.
+ *
+ * Authors:     Jiri Pirko <jiri@mellanox.com>
+ */
+
+#ifndef _MNLG_H_
+#define _MNLG_H_
+
+#include <libmnl/libmnl.h>
+
+struct mnlg_socket;
+
+struct nlmsghdr *mnlg_msg_prepare(struct mnlg_socket *nlg, uint8_t cmd,
+				  uint16_t flags);
+int mnlg_socket_send(struct mnlg_socket *nlg, const struct nlmsghdr *nlh);
+int mnlg_socket_recv_run(struct mnlg_socket *nlg, mnl_cb_t data_cb, void *data);
+int mnlg_socket_group_add(struct mnlg_socket *nlg, const char *group_name);
+struct mnlg_socket *mnlg_socket_open(const char *family_name, uint8_t version);
+void mnlg_socket_close(struct mnlg_socket *nlg);
+
+#endif /* _MNLG_H_ */

include/linux/devlink.h

+/*
+ * include/uapi/linux/devlink.h - Network physical device Netlink interface
+ * Copyright (c) 2016 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2016 Jiri Pirko <jiri@mellanox.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef _UAPI_LINUX_DEVLINK_H_
+#define _UAPI_LINUX_DEVLINK_H_
+
+#define DEVLINK_GENL_NAME "devlink"
+#define DEVLINK_GENL_VERSION 0x1
+#define DEVLINK_GENL_MCGRP_CONFIG_NAME "config"
+
+enum devlink_command {
+	/* don't change the order or add anything between, this is ABI! */
+	DEVLINK_CMD_UNSPEC,
+
+	DEVLINK_CMD_GET,		/* can dump */
+	DEVLINK_CMD_SET,
+	DEVLINK_CMD_NEW,
+	DEVLINK_CMD_DEL,
+
+	DEVLINK_CMD_PORT_GET,		/* can dump */
+	DEVLINK_CMD_PORT_SET,
+	DEVLINK_CMD_PORT_NEW,
+	DEVLINK_CMD_PORT_DEL,
+
+	DEVLINK_CMD_PORT_SPLIT,
+	DEVLINK_CMD_PORT_UNSPLIT,
+
+	/* add new commands above here */
+
+	__DEVLINK_CMD_MAX,
+	DEVLINK_CMD_MAX = __DEVLINK_CMD_MAX - 1
+};
+
+enum devlink_port_type {
+	DEVLINK_PORT_TYPE_NOTSET,
+	DEVLINK_PORT_TYPE_AUTO,
+	DEVLINK_PORT_TYPE_ETH,
+	DEVLINK_PORT_TYPE_IB,
+};
+
+enum devlink_attr {
+	/* don't change the order or add anything between, this is ABI! */
+	DEVLINK_ATTR_UNSPEC,
+
+	/* bus name + dev name together are a handle for devlink entity */
+	DEVLINK_ATTR_BUS_NAME,			/* string */
+	DEVLINK_ATTR_DEV_NAME,			/* string */
+
+	DEVLINK_ATTR_PORT_INDEX,		/* u32 */
+	DEVLINK_ATTR_PORT_TYPE,			/* u16 */
+	DEVLINK_ATTR_PORT_DESIRED_TYPE,		/* u16 */
+	DEVLINK_ATTR_PORT_NETDEV_IFINDEX,	/* u32 */
+	DEVLINK_ATTR_PORT_NETDEV_NAME,		/* string */
+	DEVLINK_ATTR_PORT_IBDEV_NAME,		/* string */
+	DEVLINK_ATTR_PORT_SPLIT_COUNT,		/* u32 */
+	DEVLINK_ATTR_PORT_SPLIT_GROUP,		/* u32 */
+
+	/* add new attributes above here, update the policy in devlink.c */
+
+	__DEVLINK_ATTR_MAX,
+	DEVLINK_ATTR_MAX = __DEVLINK_ATTR_MAX - 1
+};
+
+#endif /* _UAPI_LINUX_DEVLINK_H_ */

man/man8/devlink-dev.8

+.TH DEVLINK\-DEV 8 "14 Mar 2016" "iproute2" "Linux"
+.SH NAME
+devlink-dev \- devlink device configuration
+.SH SYNOPSIS
+.sp
+.ad l
+.in +8
+.ti -8
+.B devlink
+.RI "[ " OPTIONS " ]"
+.B dev
+.RI  " { " COMMAND " | "
+.BR help " }"
+.sp
+
+.ti -8
+.IR OPTIONS " := { "
+\fB\-V\fR[\fIersion\fR] |
+
+.ti -8
+.B devlink dev show
+.RI "[ " DEV " ]"
+
+.ti -8
+.B devlink dev help
+
+.SH "DESCRIPTION"
+.SS devlink dev show - display devlink device attributes
+
+.PP
+.I "DEV"
+- specifies the devlink device to show.
+If this argument is omitted all devices are listed.
+
+.in +4
+Format is:
+.in +2
+BUS_NAME/BUS_ADDRESS
+
+.SH "EXAMPLES"
+.PP
+devlink dev show
+.RS 4
+Shows the state of all devlink devices on the system.
+.RE
+.PP
+devlink dev show pci/0000:01:00.0
+.RS 4
+Shows the state of specified devlink device.
+
+.SH SEE ALSO
+.BR devlink (8),
+.BR devlink-port (8),
+.BR devlink-monitor (8),
+.br
+
+.SH AUTHOR
+Jiri Pirko <jiri@mellanox.com>

man/man8/devlink-monitor.8

+.TH DEVLINK\-MONITOR 8 "14 Mar 2016" "iproute2" "Linux"
+.SH "NAME"
+devlink-monitor \- state monitoring
+.SH SYNOPSIS
+.sp
+.ad l
+.in +8
+.ti -8
+.BR "devlink monitor" " [ " all " |"
+.IR OBJECT-LIST " ]"
+.sp
+
+.SH DESCRIPTION
+The
+.B devlink
+utility can monitor the state of devlink devices and ports
+continuously. This option has a slightly different format. Namely, the
+.B monitor
+command is the first in the command line and then the object list.
+
+.I OBJECT-LIST
+is the list of object types that we want to monitor.
+It may contain
+.BR dev ", " port ".
+
+.B devlink
+opens Devlink Netlink socket, listens on it and dumps state changes.
+
+.SH SEE ALSO
+.BR devlink (8),
+.BR devlink-dev (8),
+.BR devlink-port (8),
+.br
+
+.SH AUTHOR
+Jiri Pirko <jiri@mellanox.com>

man/man8/devlink-port.8

+.TH DEVLINK\-PORT 8 "14 Mar 2016" "iproute2" "Linux"
+.SH NAME
+devlink-port \- devlink port configuration
+.SH SYNOPSIS
+.sp
+.ad l
+.in +8
+.ti -8
+.B devlink
+.RI "[ " OPTIONS " ]"
+.B port
+.RI  " { " COMMAND " | "
+.BR help " }"
+.sp
+
+.ti -8
+.IR OPTIONS " := { "
+\fB\-V\fR[\fIersion\fR] |
+
+.ti -8
+.BR "devlink port set "
+.IR DEV/PORT_INDEX
+.RI "[ "
+.BR type " { " eth " | " ib " | " auto " }"
+.RI "]"
+
+.ti -8
+.BR "devlink port split "
+.IR DEV/PORT_INDEX
+.BR count
+.IR COUNT
+
+.ti -8
+.BR "devlink port unsplit "
+.IR DEV/PORT_INDEX
+
+.ti -8
+.B devlink port show
+.RI "[ " DEV/PORT_INDEX " ]"
+
+.ti -8
+.B devlink port help
+
+.SH "DESCRIPTION"
+.SS devlink port set - change devlink port attributes
+
+.PP
+.B "DEV/PORT_INDEX"
+- specifies the devlink port to operate on.
+
+.in +4
+Format is:
+.in +2
+BUS_NAME/BUS_ADDRESS/PORT_INDEX
+
+.TP
+.BR type " { " eth " | " ib " | " auto " } "
+set port type
+
+.I eth
+- Ethernet
+
+.I ib
+- Infiniband
+
+.I auto
+- autoselect
+
+.SS devlink port split - split devlink port into more
+
+.PP
+.B "DEV/PORT_INDEX"
+- specifies the devlink port to operate on.
+
+.TP
+.BI count " COUNT"
+number of ports to split to.
+
+.SS devlink port unsplit - unsplit previously split devlink port
+Could be performed on any split port of the same split group.
+
+.PP
+.B "DEV/PORT_INDEX"
+- specifies the devlink port to operate on.
+
+.SS devlink port show - display devlink port attributes
+
+.PP
+.I "DEV/PORT_INDEX"
+- specifies the devlink port to show.
+If this argument is omitted all ports are listed.
+
+.SH "EXAMPLES"
+.PP
+devlink port show
+.RS 4
+Shows the state of all devlink ports on the system.
+.RE
+.PP
+devlink port show pci/0000:01:00.0/1
+.RS 4
+Shows the state of specified devlink port.
+.RE
+.PP
+devlink port set pci/0000:01:00.0/1 type eth
+.RS 4
+Set type of specified devlink port to Ethernet.
+.RE
+.PP
+devlink port split pci/0000:01:00.0/1 count 4
+.RS 4
+Split the specified devlink port into four ports.
+.RE
+.PP
+devlink port unsplit pci/0000:01:00.0/1
+.RS 4
+Unplit the specified previously split devlink port.
+
+.SH SEE ALSO
+.BR devlink (8),
+.BR devlink-dev (8),
+.BR devlink-monitor (8),
+.br
+
+.SH AUTHOR
+Jiri Pirko <jiri@mellanox.com>

man/man8/devlink.8

+.TH DEVLINK 8 "14 Mar 2016" "iproute2" "Linux"
+.SH NAME
+devlink \- Devlink tool
+.SH SYNOPSIS
+.sp
+.ad l
+.in +8
+.ti -8
+.B devlink
+.RI "[ " OPTIONS " ] " OBJECT " { " COMMAND " | "
+.BR help " }"
+.sp
+
+.ti -8
+.IR OBJECT " := { "
+.BR dev " | " port " | " monitor " }"
+.sp
+
+.ti -8
+.IR OPTIONS " := { "
+\fB\-V\fR[\fIersion\fR] |
+
+.SH OPTIONS
+
+.TP
+.BR "\-V" , " -Version"
+Print the version of the
+.B devlink
+utility and exit.
+
+.SS
+.I OBJECT
+
+.TP
+.B dev
+- devlink device.
+
+.TP
+.B port
+- devlink port.
+
+.TP
+.B monitor
+- watch for netlink messages.
+
+.SS
+.I COMMAND
+
+Specifies the action to perform on the object.
+The set of possible actions depends on the object type.
+As a rule, it is possible to
+.B show
+(or
+.B list
+) objects, but some objects do not allow all of these operations
+or have some additional commands. The
+.B help
+command is available for all objects. It prints
+out a list of available commands and argument syntax conventions.
+.sp
+If no command is given, some default command is assumed.
+Usually it is
+.B list
+or, if the objects of this class cannot be listed,
+.BR "help" .
+
+.SH EXIT STATUS
+Exit status is 0 if command was successful or a positive integer upon failure.
+
+.SH SEE ALSO
+.BR devlink-dev (8),
+.BR devlink-port (8),
+.BR devlink-monitor (8),
+.br
+
+.SH REPORTING BUGS
+Report any bugs to the Network Developers mailing list
+.B <netdev@vger.kernel.org>
+where the development and maintenance is primarily done.
+You do not have to be subscribed to the list to send a message there.
+
+.SH AUTHOR
+Jiri Pirko <jiri@mellanox.com>