From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932108Ab2LMRIJ (ORCPT ); Thu, 13 Dec 2012 12:08:09 -0500 Received: from hrndva-omtalb.mail.rr.com ([71.74.56.122]:28767 "EHLO hrndva-omtalb.mail.rr.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752872Ab2LMRIH (ORCPT ); Thu, 13 Dec 2012 12:08:07 -0500 X-Authority-Analysis: v=2.0 cv=QPnqt33L c=1 sm=0 a=rXTBtCOcEpjy1lPqhTCpEQ==:17 a=mNMOxpOpBa8A:10 a=_aGYK8jRc_AA:10 a=5SG0PmZfjMsA:10 a=Q9fys5e9bTEA:10 a=meVymXHHAAAA:8 a=SI-C17Dh1R0A:10 a=EimCnZnUSnrWQxtqR14A:9 a=PUjeQqilurYA:10 a=rXTBtCOcEpjy1lPqhTCpEQ==:117 X-Cloudmark-Score: 0 X-Authenticated-User: X-Originating-IP: 74.67.115.198 Message-ID: <1355418485.17101.377.camel@gandalf.local.home> Subject: Re: [BUG] ftrace: available_filter_functions empty after unloading module From: Steven Rostedt To: Piotr Haber Cc: Frederic Weisbecker , Ingo Molnar , Linux Kernel Mailing List Date: Thu, 13 Dec 2012 12:08:05 -0500 In-Reply-To: <1355405698.17101.373.camel@gandalf.local.home> References: <50C0B236.4090200@broadcom.com> <1354817428.17101.10.camel@gandalf.local.home> <50C1A3A3.8030107@broadcom.com> <1354888679.17101.36.camel@gandalf.local.home> <50C5CB74.10008@broadcom.com> <1355405698.17101.373.camel@gandalf.local.home> Content-Type: text/plain; charset="ISO-8859-15" X-Mailer: Evolution 3.4.4-1 Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 2012-12-13 at 08:34 -0500, Steven Rostedt wrote: > On Mon, 2012-12-10 at 12:45 +0100, Piotr Haber wrote: > > > > So you unloaded a module that you were tracing, correct? Hmm, the > > > warning that you hit was an accounting error. The accounting of what > > > functions are traced and what isn't. For some reason it tried to disable > > > a function that wasn't being traced. > > Yes, it happens when i unload a module that i traced before > > > > > Can you easily reproduce this? If so, can you give my your .config and > > > the steps you used to trigger it. > > Not 100% but it happens often enough to be annoying > > Attached is the config > > As for reproduction, i load the brcmsmac module, add all it symbols to > > set_ftrace_filter, associate with AP and unload the module > > nothing fancy, i use 'function' tracer > > Thanks, I'll see if I can reproduce it. > I'm having trouble reproducing. Can you apply the below patch and try to reproduce it. And then show me the contents of /debug/tracing/enabled_functions. Thanks, -- Steve diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 4451aa3..e069227 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -2390,9 +2390,6 @@ t_next(struct seq_file *m, void *v, loff_t *pos) struct ftrace_ops *ops = iter->ops; struct dyn_ftrace *rec = NULL; - if (unlikely(ftrace_disabled)) - return NULL; - if (iter->flags & FTRACE_ITER_HASH) return t_hash_next(m, pos); @@ -2449,9 +2446,6 @@ static void *t_start(struct seq_file *m, loff_t *pos) mutex_lock(&ftrace_lock); - if (unlikely(ftrace_disabled)) - return NULL; - /* * If an lseek was done, then reset and start from beginning. */ @@ -2557,9 +2551,6 @@ ftrace_enabled_open(struct inode *inode, struct file *file) { struct ftrace_iterator *iter; - if (unlikely(ftrace_disabled)) - return -ENODEV; - iter = __seq_open_private(file, &show_ftrace_seq_ops, sizeof(*iter)); if (iter) { iter->pg = ftrace_pages_start;