Re: [PATCH 5/7] ftrace, perf: Add support to use function tracepoint in perf

[Frederic Weisbecker <fweisbec@gmail.com>]

On Wed, Feb 15, 2012 at 03:51:52PM +0100, Jiri Olsa wrote:
> Adding perf registration support for the ftrace function event,
> so it is now possible to register it via perf interface.
> 
> The perf_event struct statically contains ftrace_ops as a handle
> for function tracer. The function tracer is registered/unregistered
> in open/close actions.
> 
> To be efficient, we enable/disable ftrace_ops each time the traced
> process is scheduled in/out (via TRACE_REG_PERF_(ADD|DELL) handlers).
> This way tracing is enabled only when the process is running.
> Intentionally using this way instead of the event's hw state
> PERF_HES_STOPPED, which would not disable the ftrace_ops.
> 
> It is now possible to use function trace within perf commands
> like:
> 
>   perf record -e ftrace:function ls
>   perf stat -e ftrace:function ls
> 
> Allowed only for root.
> 
> Signed-off-by: Jiri Olsa <jolsa@redhat.com>

Acked-by: Frederic Weisbecker <fweisbec@gmail.com>

> ---
>  include/linux/perf_event.h      |    3 +
>  kernel/trace/trace.h            |   11 +++++
>  kernel/trace/trace_entries.h    |    6 ++-
>  kernel/trace/trace_event_perf.c |   86 +++++++++++++++++++++++++++++++++++++++
>  kernel/trace/trace_export.c     |    5 ++
>  5 files changed, 109 insertions(+), 2 deletions(-)
> 
> diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
> index 412b790..92a056f 100644
> --- a/include/linux/perf_event.h
> +++ b/include/linux/perf_event.h
> @@ -859,6 +859,9 @@ struct perf_event {
>  #ifdef CONFIG_EVENT_TRACING
>  	struct ftrace_event_call	*tp_event;
>  	struct event_filter		*filter;
> +#ifdef CONFIG_FUNCTION_TRACER
> +	struct ftrace_ops               ftrace_ops;
> +#endif
>  #endif
>  
>  #ifdef CONFIG_CGROUP_PERF
> diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
> index 638476a..76a1c50 100644
> --- a/kernel/trace/trace.h
> +++ b/kernel/trace/trace.h
> @@ -595,6 +595,8 @@ static inline int ftrace_trace_task(struct task_struct *task)
>  static inline int ftrace_is_dead(void) { return 0; }
>  #endif
>  
> +int ftrace_event_is_function(struct ftrace_event_call *call);
> +
>  /*
>   * struct trace_parser - servers for reading the user input separated by spaces
>   * @cont: set if the input is not complete - no final space char was found
> @@ -832,4 +834,13 @@ extern const char *__stop___trace_bprintk_fmt[];
>  	FTRACE_ENTRY(call, struct_name, id, PARAMS(tstruct), PARAMS(print))
>  #include "trace_entries.h"
>  
> +#ifdef CONFIG_PERF_EVENTS
> +#ifdef CONFIG_FUNCTION_TRACER
> +int perf_ftrace_event_register(struct ftrace_event_call *call,
> +			       enum trace_reg type, void *data);
> +#else
> +#define perf_ftrace_event_register NULL
> +#endif /* CONFIG_FUNCTION_TRACER */
> +#endif /* CONFIG_PERF_EVENTS */
> +
>  #endif /* _LINUX_KERNEL_TRACE_H */
> diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h
> index 9336590..47db7ed 100644
> --- a/kernel/trace/trace_entries.h
> +++ b/kernel/trace/trace_entries.h
> @@ -55,7 +55,7 @@
>  /*
>   * Function trace entry - function address and parent function address:
>   */
> -FTRACE_ENTRY(function, ftrace_entry,
> +FTRACE_ENTRY_REG(function, ftrace_entry,
>  
>  	TRACE_FN,
>  
> @@ -64,7 +64,9 @@ FTRACE_ENTRY(function, ftrace_entry,
>  		__field(	unsigned long,	parent_ip	)
>  	),
>  
> -	F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip)
> +	F_printk(" %lx <-- %lx", __entry->ip, __entry->parent_ip),
> +
> +	perf_ftrace_event_register
>  );
>  
>  /* Function call entry */
> diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
> index d72af0b..fdeeb5c 100644
> --- a/kernel/trace/trace_event_perf.c
> +++ b/kernel/trace/trace_event_perf.c
> @@ -24,6 +24,11 @@ static int	total_ref_count;
>  static int perf_trace_event_perm(struct ftrace_event_call *tp_event,
>  				 struct perf_event *p_event)
>  {
> +	/* The ftrace function trace is allowed only for root. */
> +	if (ftrace_event_is_function(tp_event) &&
> +	    perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN))
> +		return -EPERM;
> +
>  	/* No tracing, just counting, so no obvious leak */
>  	if (!(p_event->attr.sample_type & PERF_SAMPLE_RAW))
>  		return 0;
> @@ -250,3 +255,84 @@ __kprobes void *perf_trace_buf_prepare(int size, unsigned short type,
>  	return raw_data;
>  }
>  EXPORT_SYMBOL_GPL(perf_trace_buf_prepare);
> +
> +#ifdef CONFIG_FUNCTION_TRACER
> +static void
> +perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip)
> +{
> +	struct ftrace_entry *entry;
> +	struct hlist_head *head;
> +	struct pt_regs regs;
> +	int rctx;
> +
> +#define ENTRY_SIZE (ALIGN(sizeof(struct ftrace_entry) + sizeof(u32), \
> +		    sizeof(u64)) - sizeof(u32))
> +
> +	BUILD_BUG_ON(ENTRY_SIZE > PERF_MAX_TRACE_SIZE);
> +
> +	perf_fetch_caller_regs(&regs);
> +
> +	entry = perf_trace_buf_prepare(ENTRY_SIZE, TRACE_FN, NULL, &rctx);
> +	if (!entry)
> +		return;
> +
> +	entry->ip = ip;
> +	entry->parent_ip = parent_ip;
> +
> +	head = this_cpu_ptr(event_function.perf_events);
> +	perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, 0,
> +			      1, &regs, head);
> +
> +#undef ENTRY_SIZE
> +}
> +
> +static int perf_ftrace_function_register(struct perf_event *event)
> +{
> +	struct ftrace_ops *ops = &event->ftrace_ops;
> +
> +	ops->flags |= FTRACE_OPS_FL_CONTROL;
> +	ops->func = perf_ftrace_function_call;
> +	return register_ftrace_function(ops);
> +}
> +
> +static int perf_ftrace_function_unregister(struct perf_event *event)
> +{
> +	struct ftrace_ops *ops = &event->ftrace_ops;
> +	return unregister_ftrace_function(ops);
> +}
> +
> +static void perf_ftrace_function_enable(struct perf_event *event)
> +{
> +	ftrace_function_local_enable(&event->ftrace_ops);
> +}
> +
> +static void perf_ftrace_function_disable(struct perf_event *event)
> +{
> +	ftrace_function_local_disable(&event->ftrace_ops);
> +}
> +
> +int perf_ftrace_event_register(struct ftrace_event_call *call,
> +			       enum trace_reg type, void *data)
> +{
> +	switch (type) {
> +	case TRACE_REG_REGISTER:
> +	case TRACE_REG_UNREGISTER:
> +		break;
> +	case TRACE_REG_PERF_REGISTER:
> +	case TRACE_REG_PERF_UNREGISTER:
> +		return 0;
> +	case TRACE_REG_PERF_OPEN:
> +		return perf_ftrace_function_register(data);
> +	case TRACE_REG_PERF_CLOSE:
> +		return perf_ftrace_function_unregister(data);
> +	case TRACE_REG_PERF_ADD:
> +		perf_ftrace_function_enable(data);
> +		return 0;
> +	case TRACE_REG_PERF_DEL:
> +		perf_ftrace_function_disable(data);
> +		return 0;
> +	}
> +
> +	return -EINVAL;
> +}
> +#endif /* CONFIG_FUNCTION_TRACER */
> diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c
> index f74de86..a3dbee6 100644
> --- a/kernel/trace/trace_export.c
> +++ b/kernel/trace/trace_export.c
> @@ -184,4 +184,9 @@ __attribute__((section("_ftrace_events"))) *__event_##call = &event_##call;
>  	FTRACE_ENTRY_REG(call, struct_name, etype,			\
>  			 PARAMS(tstruct), PARAMS(print), NULL)
>  
> +int ftrace_event_is_function(struct ftrace_event_call *call)
> +{
> +	return call == &event_function;
> +}
> +
>  #include "trace_entries.h"
> -- 
> 1.7.1
>