RDMA/nldev: provide detailed CM_ID information

Implement RDMA nldev netlink interface to get detailed CM_ID information.

Because cm_id's are attached to rdma devices in various work queue
contexts, the pid and task information at restrak_add() time is sometimes
not useful.  For example, an nvme/f host connection cm_id ends up being
bound to a device in a work queue context and the resulting pid at attach
time no longer exists after connection setup.  So instead we mark all
cm_id's created via the rdma_ucm as "user", and all others as "kernel".
This required tweaking the restrack code a little.  It also required
wrapping some rdma_cm functions to allow passing the module name string.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

1 files changed, 98 insertions, 9 deletions

diff --git a/drivers/infiniband/core/nldev.c b/drivers/infiniband/core/nldev.c
index f38c6838bb31..3fd3f9e99e11 100644
--- a/drivers/infiniband/core/nldev.c
+++ b/drivers/infiniband/core/nldev.c
@@ -34,9 +34,11 @@
 #include <linux/pid.h>
 #include <linux/pid_namespace.h>
 #include <net/netlink.h>
+#include <rdma/rdma_cm.h>
 #include <rdma/rdma_netlink.h>
 
 #include "core_priv.h"
+#include "cma_priv.h"
 
 static const struct nla_policy nldev_policy[RDMA_NLDEV_ATTR_MAX] = {
 	[RDMA_NLDEV_ATTR_DEV_INDEX]     = { .type = NLA_U32 },
@@ -71,6 +73,13 @@ static const struct nla_policy nldev_policy[RDMA_NLDEV_ATTR_MAX] = {
 	[RDMA_NLDEV_ATTR_RES_PID]		= { .type = NLA_U32 },
 	[RDMA_NLDEV_ATTR_RES_KERN_NAME]		= { .type = NLA_NUL_STRING,
 						    .len = TASK_COMM_LEN },
+	[RDMA_NLDEV_ATTR_RES_CM_ID]		= { .type = NLA_NESTED },
+	[RDMA_NLDEV_ATTR_RES_CM_ID_ENTRY]	= { .type = NLA_NESTED },
+	[RDMA_NLDEV_ATTR_RES_PS]		= { .type = NLA_U32 },
+	[RDMA_NLDEV_ATTR_RES_SRC_ADDR]	= {
+			.len = sizeof(struct __kernel_sockaddr_storage) },
+	[RDMA_NLDEV_ATTR_RES_DST_ADDR]	= {
+			.len = sizeof(struct __kernel_sockaddr_storage) },
 };
 
 static int fill_nldev_handle(struct sk_buff *msg, struct ib_device *device)
@@ -182,6 +191,7 @@ static int fill_res_info(struct sk_buff *msg, struct ib_device *device)
 		[RDMA_RESTRACK_PD] = "pd",
 		[RDMA_RESTRACK_CQ] = "cq",
 		[RDMA_RESTRACK_QP] = "qp",
+		[RDMA_RESTRACK_CM_ID] = "cm_id",
 	};
 
 	struct rdma_restrack_root *res = &device->res;
@@ -212,6 +222,25 @@ err:
 	return ret;
 }
 
+static int fill_res_name_pid(struct sk_buff *msg,
+			     struct rdma_restrack_entry *res)
+{
+	/*
+	 * For user resources, user is should read /proc/PID/comm to get the
+	 * name of the task file.
+	 */
+	if (rdma_is_kernel_res(res)) {
+		if (nla_put_string(msg, RDMA_NLDEV_ATTR_RES_KERN_NAME,
+		    res->kern_name))
+			return -EMSGSIZE;
+	} else {
+		if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_PID,
+		    task_pid_vnr(res->task)))
+			return -EMSGSIZE;
+	}
+	return 0;
+}
+
 static int fill_res_qp_entry(struct sk_buff *msg, struct netlink_callback *cb,
 			     struct rdma_restrack_entry *res, uint32_t port)
 {
@@ -262,19 +291,65 @@ static int fill_res_qp_entry(struct sk_buff *msg, struct netlink_callback *cb,
 	if (nla_put_u8(msg, RDMA_NLDEV_ATTR_RES_STATE, qp_attr.qp_state))
 		goto err;
 
-	/*
-	 * Existence of task means that it is user QP and netlink
-	 * user is invited to go and read /proc/PID/comm to get name
-	 * of the task file and res->task_com should be NULL.
-	 */
-	if (rdma_is_kernel_res(res)) {
-		if (nla_put_string(msg, RDMA_NLDEV_ATTR_RES_KERN_NAME, res->kern_name))
+	if (fill_res_name_pid(msg, res))
+		goto err;
+
+	nla_nest_end(msg, entry_attr);
+	return 0;
+
+err:
+	nla_nest_cancel(msg, entry_attr);
+out:
+	return -EMSGSIZE;
+}
+
+static int fill_res_cm_id_entry(struct sk_buff *msg,
+				struct netlink_callback *cb,
+				struct rdma_restrack_entry *res, uint32_t port)
+{
+	struct rdma_id_private *id_priv =
+				container_of(res, struct rdma_id_private, res);
+	struct rdma_cm_id *cm_id = &id_priv->id;
+	struct nlattr *entry_attr;
+
+	if (port && port != cm_id->port_num)
+		return 0;
+
+	entry_attr = nla_nest_start(msg, RDMA_NLDEV_ATTR_RES_CM_ID_ENTRY);
+	if (!entry_attr)
+		goto out;
+
+	if (cm_id->port_num &&
+	    nla_put_u32(msg, RDMA_NLDEV_ATTR_PORT_INDEX, cm_id->port_num))
+		goto err;
+
+	if (id_priv->qp_num) {
+		if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_LQPN, id_priv->qp_num))
 			goto err;
-	} else {
-		if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_PID, task_pid_vnr(res->task)))
+		if (nla_put_u8(msg, RDMA_NLDEV_ATTR_RES_TYPE, cm_id->qp_type))
 			goto err;
 	}
 
+	if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_PS, cm_id->ps))
+		goto err;
+
+	if (nla_put_u8(msg, RDMA_NLDEV_ATTR_RES_STATE, id_priv->state))
+		goto err;
+
+	if (cm_id->route.addr.src_addr.ss_family &&
+	    nla_put(msg, RDMA_NLDEV_ATTR_RES_SRC_ADDR,
+		    sizeof(cm_id->route.addr.src_addr),
+		    &cm_id->route.addr.src_addr))
+		goto err;
+	if (cm_id->route.addr.dst_addr.ss_family &&
+	    nla_put(msg, RDMA_NLDEV_ATTR_RES_DST_ADDR,
+		    sizeof(cm_id->route.addr.dst_addr),
+		    &cm_id->route.addr.dst_addr))
+		goto err;
+
+	if (fill_res_name_pid(msg, res))
+		goto err;
+
 	nla_nest_end(msg, entry_attr);
 	return 0;
 
@@ -571,6 +646,11 @@ static const struct nldev_fill_res_entry fill_entries[RDMA_RESTRACK_MAX] = {
 		.nldev_cmd = RDMA_NLDEV_CMD_RES_QP_GET,
 		.nldev_attr = RDMA_NLDEV_ATTR_RES_QP,
 	},
+	[RDMA_RESTRACK_CM_ID] = {
+		.fill_res_func = fill_res_cm_id_entry,
+		.nldev_cmd = RDMA_NLDEV_CMD_RES_CM_ID_GET,
+		.nldev_attr = RDMA_NLDEV_ATTR_RES_CM_ID,
+	},
 };
 
 static int res_get_common_dumpit(struct sk_buff *skb,
@@ -713,6 +793,12 @@ static int nldev_res_get_qp_dumpit(struct sk_buff *skb,
 	return res_get_common_dumpit(skb, cb, RDMA_RESTRACK_QP);
 }
 
+static int nldev_res_get_cm_id_dumpit(struct sk_buff *skb,
+				      struct netlink_callback *cb)
+{
+	return res_get_common_dumpit(skb, cb, RDMA_RESTRACK_CM_ID);
+}
+
 static const struct rdma_nl_cbs nldev_cb_table[RDMA_NLDEV_NUM_OPS] = {
 	[RDMA_NLDEV_CMD_GET] = {
 		.doit = nldev_get_doit,
@@ -739,6 +825,9 @@ static const struct rdma_nl_cbs nldev_cb_table[RDMA_NLDEV_NUM_OPS] = {
 		 * too.
 		 */
 	},
+	[RDMA_NLDEV_CMD_RES_CM_ID_GET] = {
+		.dump = nldev_res_get_cm_id_dumpit,
+	},
 };
 
 void __init nldev_init(void)