[PATCH 1/3] perf: add PERF_RECORD_NAMESPACES to include namespaces related info

[Hari Bathini <hbathini@linux.vnet.ibm.com>]

With the advert of container technologies like docker, that depend
on namespaces for isolation, there is a need for tracing support for
namespaces. This patch introduces new PERF_RECORD_NAMESPACES event
for tracing based on namespaces related info.

Signed-off-by: Hari Bathini <hbathini@linux.vnet.ibm.com>
---
 fs/mount.h                      |   17 -----
 include/linux/mnt_namespace.h   |   18 +++++
 include/linux/perf_event.h      |    1 
 include/uapi/linux/perf_event.h |   21 ++++++
 kernel/events/core.c            |  136 +++++++++++++++++++++++++++++++++++++++
 kernel/fork.c                   |    4 +
 kernel/nsproxy.c                |    5 +
 7 files changed, 184 insertions(+), 18 deletions(-)

diff --git a/fs/mount.h b/fs/mount.h
index d2e25d7..5ec592b 100644
--- a/fs/mount.h
+++ b/fs/mount.h
@@ -1,22 +1,7 @@
 #include <linux/mount.h>
 #include <linux/seq_file.h>
-#include <linux/poll.h>
-#include <linux/ns_common.h>
 #include <linux/fs_pin.h>
-
-struct mnt_namespace {
-	atomic_t		count;
-	struct ns_common	ns;
-	struct mount *	root;
-	struct list_head	list;
-	struct user_namespace	*user_ns;
-	struct ucounts		*ucounts;
-	u64			seq;	/* Sequence number to prevent loops */
-	wait_queue_head_t poll;
-	u64 event;
-	unsigned int		mounts; /* # of mounts in the namespace */
-	unsigned int		pending_mounts;
-};
+#include <linux/mnt_namespace.h>
 
 struct mnt_pcp {
 	int mnt_count;
diff --git a/include/linux/mnt_namespace.h b/include/linux/mnt_namespace.h
index 12b2ab5..b911ca6 100644
--- a/include/linux/mnt_namespace.h
+++ b/include/linux/mnt_namespace.h
@@ -2,10 +2,26 @@
 #define _NAMESPACE_H_
 #ifdef __KERNEL__
 
-struct mnt_namespace;
+#include <linux/poll.h>
+#include <linux/ns_common.h>
+
 struct fs_struct;
 struct user_namespace;
 
+struct mnt_namespace {
+	atomic_t		count;
+	struct ns_common	ns;
+	struct mount 		*root;
+	struct list_head	list;
+	struct user_namespace	*user_ns;
+	struct ucounts		*ucounts;
+	u64			seq;	/* Sequence number to prevent loops */
+	wait_queue_head_t	poll;
+	u64			event;
+	unsigned int		mounts; /* # of mounts in the namespace */
+	unsigned int		pending_mounts;
+};
+
 extern struct mnt_namespace *copy_mnt_ns(unsigned long, struct mnt_namespace *,
 		struct user_namespace *, struct fs_struct *);
 extern void put_mnt_ns(struct mnt_namespace *ns);
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 4741ecd..243b988 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -1110,6 +1110,7 @@ extern int perf_unregister_guest_info_callbacks(struct perf_guest_info_callbacks
 
 extern void perf_event_exec(void);
 extern void perf_event_comm(struct task_struct *tsk, bool exec);
+extern void perf_event_namespaces(struct task_struct *tsk);
 extern void perf_event_fork(struct task_struct *tsk);
 
 /* Callchains */
diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h
index c66a485..575aed6 100644
--- a/include/uapi/linux/perf_event.h
+++ b/include/uapi/linux/perf_event.h
@@ -344,7 +344,8 @@ struct perf_event_attr {
 				use_clockid    :  1, /* use @clockid for time fields */
 				context_switch :  1, /* context switch data */
 				write_backward :  1, /* Write ring buffer from end to beginning */
-				__reserved_1   : 36;
+				namespaces     :  1, /* include namespaces data */
+				__reserved_1   : 35;
 
 	union {
 		__u32		wakeup_events;	  /* wakeup every n events */
@@ -862,6 +863,24 @@ enum perf_event_type {
 	 */
 	PERF_RECORD_SWITCH_CPU_WIDE		= 15,
 
+	/*
+	 * struct {
+	 *	struct perf_event_header	header;
+	 *
+	 *	u32				pid, tid;
+	 *	u64				time;
+	 *	u32				uts_ns_inum;
+	 *	u32				ipc_ns_inum;
+	 *	u32				mnt_ns_inum;
+	 *	u32				pid_ns_inum;
+	 *	u32				net_ns_inum;
+	 *	u32				cgroup_ns_inum;
+	 *	u32				user_ns_inum;
+	 * 	struct sample_id		sample_id;
+	 * };
+	 */
+	PERF_RECORD_NAMESPACES			= 16,
+
 	PERF_RECORD_MAX,			/* non-ABI */
 };
 
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 0e29213..3bdc2e7 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -46,6 +46,9 @@
 #include <linux/filter.h>
 #include <linux/namei.h>
 #include <linux/parser.h>
+#include <linux/ipc_namespace.h>
+#include <linux/utsname.h>
+#include <linux/mnt_namespace.h>
 
 #include "internal.h"
 
@@ -375,6 +378,7 @@ static DEFINE_PER_CPU(struct pmu_event_list, pmu_sb_events);
 
 static atomic_t nr_mmap_events __read_mostly;
 static atomic_t nr_comm_events __read_mostly;
+static atomic_t nr_namespaces_events __read_mostly;
 static atomic_t nr_task_events __read_mostly;
 static atomic_t nr_freq_events __read_mostly;
 static atomic_t nr_switch_events __read_mostly;
@@ -3874,6 +3878,8 @@ static void unaccount_event(struct perf_event *event)
 		atomic_dec(&nr_mmap_events);
 	if (event->attr.comm)
 		atomic_dec(&nr_comm_events);
+	if (event->attr.namespaces)
+		atomic_dec(&nr_namespaces_events);
 	if (event->attr.task)
 		atomic_dec(&nr_task_events);
 	if (event->attr.freq)
@@ -6476,6 +6482,134 @@ void perf_event_comm(struct task_struct *task, bool exec)
 }
 
 /*
+ * namespaces tracking
+ */
+
+struct perf_namespaces_event {
+	struct task_struct	*task;
+
+	struct {
+		struct perf_event_header	header;
+		u32				pid;
+		u32				tid;
+		u64				time;
+		u32				uts_ns_inum;
+		u32				ipc_ns_inum;
+		u32				mnt_ns_inum;
+		u32				pid_ns_inum;
+		u32				net_ns_inum;
+		u32				cgroup_ns_inum;
+		u32				user_ns_inum;
+	} event_id;
+};
+
+static int perf_event_namespaces_match(struct perf_event *event)
+{
+	return event->attr.namespaces;
+}
+
+static void perf_event_namespaces_output(struct perf_event *event,
+					 void *data)
+{
+	struct perf_namespaces_event *namespaces_event = data;
+	struct perf_output_handle handle;
+	struct perf_sample_data sample;
+	int size = namespaces_event->event_id.header.size;
+	struct nsproxy *nsproxy;
+	int ret;
+
+	if (!perf_event_namespaces_match(event))
+		return;
+
+	perf_event_header__init_id(&namespaces_event->event_id.header,
+				   &sample, event);
+	ret = perf_output_begin(&handle, event,
+				namespaces_event->event_id.header.size);
+
+	if (ret)
+		goto out;
+
+	namespaces_event->event_id.pid = perf_event_pid(event,
+							namespaces_event->task);
+	namespaces_event->event_id.tid = perf_event_tid(event,
+							namespaces_event->task);
+
+	if (namespaces_event->task != current)
+		task_lock(namespaces_event->task);
+
+	nsproxy = namespaces_event->task->nsproxy;
+	if (nsproxy != NULL) {
+		namespaces_event->event_id.uts_ns_inum =
+					nsproxy->uts_ns->ns.inum;
+#ifdef CONFIG_IPC_NS
+		namespaces_event->event_id.ipc_ns_inum =
+					nsproxy->ipc_ns->ns.inum;
+#endif
+		namespaces_event->event_id.mnt_ns_inum =
+					nsproxy->mnt_ns->ns.inum;
+		namespaces_event->event_id.pid_ns_inum =
+					nsproxy->pid_ns_for_children->ns.inum;
+#ifdef CONFIG_NET
+		namespaces_event->event_id.net_ns_inum =
+					nsproxy->net_ns->ns.inum;
+#endif
+#ifdef CONFIG_CGROUPS
+		namespaces_event->event_id.cgroup_ns_inum =
+					nsproxy->cgroup_ns->ns.inum;
+#endif
+	}
+
+	namespaces_event->event_id.user_ns_inum =
+			__task_cred(namespaces_event->task)->user_ns->ns.inum;
+
+	if (namespaces_event->task != current)
+		task_unlock(namespaces_event->task);
+
+	namespaces_event->event_id.time = perf_event_clock(event);
+
+	perf_output_put(&handle, namespaces_event->event_id);
+
+	perf_event__output_id_sample(event, &handle, &sample);
+
+	perf_output_end(&handle);
+out:
+	namespaces_event->event_id.header.size = size;
+}
+
+void perf_event_namespaces(struct task_struct *task)
+{
+	struct perf_namespaces_event namespaces_event;
+
+	if (!atomic_read(&nr_namespaces_events))
+		return;
+
+	namespaces_event = (struct perf_namespaces_event){
+		.task	= task,
+		.event_id  = {
+			.header = {
+				.type = PERF_RECORD_NAMESPACES,
+				.misc = 0,
+				.size = sizeof(namespaces_event.event_id),
+			},
+			/* .pid */
+			/* .tid */
+			/* .time */
+			/* .uts_ns_inum */
+			/* .ipc_ns_inum */
+			/* .mnt_ns_inum */
+			/* .pid_ns_inum */
+			/* .net_ns_inum */
+			/* .cgroup_ns_inum */
+			/* .user_ns_inum */
+		},
+	};
+
+	perf_iterate_sb(perf_event_namespaces_output,
+			&namespaces_event,
+			NULL);
+}
+
+/*
  * mmap tracking
  */
 
@@ -9018,6 +9152,8 @@ static void account_event(struct perf_event *event)
 		atomic_inc(&nr_mmap_events);
 	if (event->attr.comm)
 		atomic_inc(&nr_comm_events);
+	if (event->attr.namespaces)
+		atomic_inc(&nr_namespaces_events);
 	if (event->attr.task)
 		atomic_inc(&nr_task_events);
 	if (event->attr.freq)
diff --git a/kernel/fork.c b/kernel/fork.c
index 997ac1d..3faca3d 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1818,6 +1818,7 @@ static __latent_entropy struct task_struct *copy_process(
 	cgroup_post_fork(p);
 	threadgroup_change_end(current);
 	perf_event_fork(p);
+	perf_event_namespaces(p);
 
 	trace_task_newtask(p, clone_flags);
 	uprobe_copy_process(p, clone_flags);
@@ -2280,6 +2281,9 @@ SYSCALL_DEFINE1(unshare, unsigned long, unshare_flags)
 		free_fs_struct(new_fs);
 
 bad_unshare_out:
+	if (!err)
+		perf_event_namespaces(current);
+
 	return err;
 }
 
diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c
index 782102e..4c25e6e 100644
--- a/kernel/nsproxy.c
+++ b/kernel/nsproxy.c
@@ -26,6 +26,7 @@
 #include <linux/file.h>
 #include <linux/syscalls.h>
 #include <linux/cgroup.h>
+#include <linux/perf_event.h>
 
 static struct kmem_cache *nsproxy_cachep;
 
@@ -264,6 +265,10 @@ SYSCALL_DEFINE2(setns, int, fd, int, nstype)
 	switch_task_namespaces(tsk, new_nsproxy);
 out:
 	fput(file);
+
+	if (!err)
+		perf_event_namespaces(tsk);
+
 	return err;
 }