From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752765Ab0FXF6x (ORCPT ); Thu, 24 Jun 2010 01:58:53 -0400 Received: from casper.infradead.org ([85.118.1.10]:36897 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751266Ab0FXF6w convert rfc822-to-8bit (ORCPT ); Thu, 24 Jun 2010 01:58:52 -0400 Subject: Re: perf regression: failed to stat syscalls events From: Peter Zijlstra To: Wu Fengguang Cc: LKML , Frederic Weisbecker , Steven Rostedt In-Reply-To: <20100624014422.GA17476@localhost> References: <20100624013915.GA16965@localhost> <20100624014422.GA17476@localhost> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8BIT Date: Thu, 24 Jun 2010 07:58:45 +0200 Message-ID: <1277359125.1875.828.camel@laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.28.3 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 2010-06-24 at 09:44 +0800, Wu Fengguang wrote: > On Thu, Jun 24, 2010 at 09:39:15AM +0800, Wu Fengguang wrote: > > Greetings, > > > > I noticed that > > > > # perf stat -e syscalls:sys_enter_mmap true > > > > refuse to work on 2.6.35-rc1. The last working kernel is 2.6.34. > > Other events like lock:* or block:* or whatever continue to work. > So this is a problem specific to syscalls:*. Does your tree contain: --- commit a8fb2608053547bc3152ea61a5ec7cdfce5d942c Author: Steven Rostedt Date: Thu Jun 10 14:53:16 2010 -0400 perf/tracing: Fix regression of perf losing kprobe events With the addition of the code to shrink the kernel tracepoint infrastructure, we lost kprobes being traced by perf. The reason is that I tested if the "tp_event->class->perf_probe" existed before enabling it. This prevents "ftrace only" events (like the function trace events) from being enabled by perf. Unfortunately, kprobe events do not use perf_probe. This causes kprobes to be missed by perf. To fix this, we add the test to see if "tp_event->class->reg" exists as well as perf_probe. Normal trace events have only "perf_probe" but no "reg" function, and kprobes and syscalls have the "reg" but no "perf_probe". The ftrace unique events do not have either, so this is a valid test. If a kprobe or syscall is not to be probed by perf, the "reg" function is called anyway, and will return a failure and prevent perf from probing it. Reported-by: Srikar Dronamraju Tested-by: Srikar Dronamraju Acked-by: Peter Zijlstra Signed-off-by: Steven Rostedt diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index e6f6588..8a2b73f 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -96,7 +96,9 @@ int perf_trace_init(struct perf_event *p_event) mutex_lock(&event_mutex); list_for_each_entry(tp_event, &ftrace_events, list) { if (tp_event->event.type == event_id && - tp_event->class && tp_event->class->perf_probe && + tp_event->class && + (tp_event->class->perf_probe || + tp_event->class->reg) && try_module_get(tp_event->mod)) { ret = perf_trace_event_init(tp_event, p_event); break;