perf: Add cgroup support

This kernel patch adds the ability to filter monitoring based on
container groups (cgroups). This is for use in per-cpu mode only.

The cgroup to monitor is passed as a file descriptor in the pid
argument to the syscall. The file descriptor must be opened to
the cgroup name in the cgroup filesystem. For instance, if the
cgroup name is foo and cgroupfs is mounted in /cgroup, then the
file descriptor is opened to /cgroup/foo. Cgroup mode is
activated by passing PERF_FLAG_PID_CGROUP in the flags argument
to the syscall.

For instance to measure in cgroup foo on CPU1 assuming
cgroupfs is mounted under /cgroup:

struct perf_event_attr attr;
int cgroup_fd, fd;

cgroup_fd = open("/cgroup/foo", O_RDONLY);
fd = perf_event_open(&attr, cgroup_fd, 1, -1, PERF_FLAG_PID_CGROUP);
close(cgroup_fd);

Signed-off-by: Stephane Eranian <eranian@google.com>
[ added perf_cgroup_{exit,attach} ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <4d590250.114ddf0a.689e.4482@mx.google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

include/linux/cgroup.h

@@ -627,6 +627,7 @@ bool css_is_ancestor(struct cgroup_subsys_state *cg,
 /* Get id and depth of css */
 unsigned short css_id(struct cgroup_subsys_state *css);
 unsigned short css_depth(struct cgroup_subsys_state *css);
+struct cgroup_subsys_state *cgroup_css_from_dir(struct file *f, int id);
 
 #else /* !CONFIG_CGROUPS */
 

include/linux/cgroup_subsys.h

@@ -65,4 +65,8 @@ SUBSYS(net_cls)
 SUBSYS(blkio)
 #endif
 
+#ifdef CONFIG_CGROUP_PERF
+SUBSYS(perf)
+#endif
+
 /* */

include/linux/perf_event.h

@@ -464,13 +464,15 @@ enum perf_callchain_context {
 
 #define PERF_FLAG_FD_NO_GROUP	(1U << 0)
 #define PERF_FLAG_FD_OUTPUT	(1U << 1)
+#define PERF_FLAG_PID_CGROUP	(1U << 2) /* pid=cgroup id, per-cpu mode only */
 
 #ifdef __KERNEL__
 /*
  * Kernel-internal data types and definitions:
  */
 
 #ifdef CONFIG_PERF_EVENTS
+# include <linux/cgroup.h>
 # include <asm/perf_event.h>
 # include <asm/local64.h>
 #endif
@@ -716,6 +718,22 @@ struct swevent_hlist {
 #define PERF_ATTACH_GROUP	0x02
 #define PERF_ATTACH_TASK	0x04
 
+#ifdef CONFIG_CGROUP_PERF
+/*
+ * perf_cgroup_info keeps track of time_enabled for a cgroup.
+ * This is a per-cpu dynamically allocated data structure.
+ */
+struct perf_cgroup_info {
+	u64 time;
+	u64 timestamp;
+};
+
+struct perf_cgroup {
+	struct cgroup_subsys_state css;
+	struct perf_cgroup_info *info;	/* timing info, one per cpu */
+};
+#endif
+
 /**
  * struct perf_event - performance event kernel representation:
  */
@@ -832,6 +850,11 @@ struct perf_event {
 	struct event_filter		*filter;
 #endif
 
+#ifdef CONFIG_CGROUP_PERF
+	struct perf_cgroup		*cgrp; /* cgroup event is attach to */
+	int				cgrp_defer_enabled;
+#endif
+
 #endif /* CONFIG_PERF_EVENTS */
 };
 
@@ -886,6 +909,7 @@ struct perf_event_context {
 	u64				generation;
 	int				pin_count;
 	struct rcu_head			rcu_head;
+	int				nr_cgroups; /* cgroup events present */
 };
 
 /*
@@ -905,6 +929,9 @@ struct perf_cpu_context {
 	struct list_head		rotation_list;
 	int				jiffies_interval;
 	struct pmu			*active_pmu;
+#ifdef CONFIG_CGROUP_PERF
+	struct perf_cgroup		*cgrp;
+#endif
 };
 
 struct perf_output_handle {
@@ -1040,19 +1067,19 @@ perf_sw_event(u32 event_id, u64 nr, int nmi, struct pt_regs *regs, u64 addr)
 	__perf_sw_event(event_id, nr, nmi, regs, addr);
 }
 
-extern atomic_t perf_task_events;
+extern atomic_t perf_sched_events;
 
 static inline void perf_event_task_sched_in(struct task_struct *task)
 {
-	COND_STMT(&perf_task_events, __perf_event_task_sched_in(task));
+	COND_STMT(&perf_sched_events, __perf_event_task_sched_in(task));
 }
 
 static inline
 void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next)
 {
 	perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0);
 
-	COND_STMT(&perf_task_events, __perf_event_task_sched_out(task, next));
+	COND_STMT(&perf_sched_events, __perf_event_task_sched_out(task, next));
 }
 
 extern void perf_event_mmap(struct vm_area_struct *vma);

init/Kconfig

@@ -683,6 +683,16 @@ config CGROUP_MEM_RES_CTLR_SWAP_ENABLED
 	  select this option (if, for some reason, they need to disable it
 	  then noswapaccount does the trick).
 
+config CGROUP_PERF
+	bool "Enable perf_event per-cpu per-container group (cgroup) monitoring"
+	depends on PERF_EVENTS && CGROUPS
+	help
+	  This option extends the per-cpu mode to restrict monitoring to
+	  threads which belong to the cgroup specificied and run on the
+	  designated cpu.
+
+	  Say N if unsure.
+
 menuconfig CGROUP_SCHED
 	bool "Group CPU scheduler"
 	depends on EXPERIMENTAL

kernel/cgroup.c

@@ -4818,6 +4818,29 @@ css_get_next(struct cgroup_subsys *ss, int id,
 	return ret;
 }
 
+/*
+ * get corresponding css from file open on cgroupfs directory
+ */
+struct cgroup_subsys_state *cgroup_css_from_dir(struct file *f, int id)
+{
+	struct cgroup *cgrp;
+	struct inode *inode;
+	struct cgroup_subsys_state *css;
+
+	inode = f->f_dentry->d_inode;
+	/* check in cgroup filesystem dir */
+	if (inode->i_op != &cgroup_dir_inode_operations)
+		return ERR_PTR(-EBADF);
+
+	if (id < 0 || id >= CGROUP_SUBSYS_COUNT)
+		return ERR_PTR(-EINVAL);
+
+	/* get cgroup */
+	cgrp = __d_cgrp(f->f_dentry);
+	css = cgrp->subsys[id];
+	return css ? css : ERR_PTR(-ENOENT);
+}
+
 #ifdef CONFIG_CGROUP_DEBUG
 static struct cgroup_subsys_state *debug_create(struct cgroup_subsys *ss,
 						   struct cgroup *cont)