public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Steven Rostedt <rostedt@goodmis.org>
To: linux-kernel@vger.kernel.org
Cc: Ingo Molnar <mingo@kernel.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Fengguang Wu <fengguang.wu@intel.com>
Subject: [for-next][PATCH 5/6] tracing: Fix trace events build without modules
Date: Mon, 04 Mar 2013 12:20:10 -0500	[thread overview]
Message-ID: <20130304172616.025880747@goodmis.org> (raw)
In-Reply-To: 20130304172005.362537932@goodmis.org

[-- Attachment #1: 0005-tracing-Fix-trace-events-build-without-modules.patch --]
[-- Type: text/plain, Size: 4295 bytes --]

From: "Steven Rostedt (Red Hat)" <srostedt@redhat.com>

The new multi-buffers added a descriptor that kept track of module
events, and the directories they use, with struct ftace_module_file_ops.
This is used to add a ref count to keep modules from unloading while
their files are being accessed.

As the descriptor is only needed when CONFIG_MODULES is enabled, it
is only declared when the config is enabled. But that struct is
dereferenced in a few areas outside the #ifdef CONFIG_MODULES.

By adding some helper routines and moving code around a little,
events can be compiled again without modules.

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 kernel/trace/trace_events.c |   55 ++++++++++++++++++++++++++++---------------
 1 file changed, 36 insertions(+), 19 deletions(-)

diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index 63b4bdf..0f1307a 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -1546,9 +1546,18 @@ struct ftrace_module_file_ops {
 	struct file_operations		filter;
 };
 
-static struct ftrace_module_file_ops *find_ftrace_file_ops(struct module *mod)
+static struct ftrace_module_file_ops *
+find_ftrace_file_ops(struct ftrace_module_file_ops *file_ops, struct module *mod)
 {
-	struct ftrace_module_file_ops *file_ops;
+	/*
+	 * As event_calls are added in groups by module,
+	 * when we find one file_ops, we don't need to search for
+	 * each call in that module, as the rest should be the
+	 * same. Only search for a new one if the last one did
+	 * not match.
+	 */
+	if (file_ops && mod == file_ops->mod)
+		return file_ops;
 
 	list_for_each_entry(file_ops, &ftrace_module_file_list, list) {
 		if (file_ops->mod == mod)
@@ -1664,16 +1673,35 @@ static int trace_module_notify(struct notifier_block *self,
 
 	return 0;
 }
+
+static int
+__trace_add_new_mod_event(struct ftrace_event_call *call,
+			  struct trace_array *tr,
+			  struct ftrace_module_file_ops *file_ops)
+{
+	return __trace_add_new_event(call, tr,
+				     &file_ops->id, &file_ops->enable,
+				     &file_ops->filter, &file_ops->format);
+}
+
 #else
-static struct ftrace_module_file_ops *find_ftrace_file_ops(struct module *mod)
+static inline struct ftrace_module_file_ops *
+find_ftrace_file_ops(struct ftrace_module_file_ops *file_ops, struct module *mod)
 {
 	return NULL;
 }
-static int trace_module_notify(struct notifier_block *self,
-			       unsigned long val, void *data)
+static inline int trace_module_notify(struct notifier_block *self,
+				      unsigned long val, void *data)
 {
 	return 0;
 }
+static inline int
+__trace_add_new_mod_event(struct ftrace_event_call *call,
+			  struct trace_array *tr,
+			  struct ftrace_module_file_ops *file_ops)
+{
+	return -ENODEV;
+}
 #endif /* CONFIG_MODULES */
 
 /* Create a new event directory structure for a trace directory. */
@@ -1692,20 +1720,11 @@ __trace_add_event_dirs(struct trace_array *tr)
 			 * want the module to disappear when reading one
 			 * of these files). The file_ops keep account of
 			 * the module ref count.
-			 *
-			 * As event_calls are added in groups by module,
-			 * when we find one file_ops, we don't need to search for
-			 * each call in that module, as the rest should be the
-			 * same. Only search for a new one if the last one did
-			 * not match.
 			 */
-			if (!file_ops || call->mod != file_ops->mod)
-				file_ops = find_ftrace_file_ops(call->mod);
+			file_ops = find_ftrace_file_ops(file_ops, call->mod);
 			if (!file_ops)
 				continue; /* Warn? */
-			ret = __trace_add_new_event(call, tr,
-					&file_ops->id, &file_ops->enable,
-					&file_ops->filter, &file_ops->format);
+			ret = __trace_add_new_mod_event(call, tr, file_ops);
 			if (ret < 0)
 				pr_warning("Could not create directory for event %s\n",
 					   call->name);
@@ -1794,9 +1813,7 @@ __add_event_to_tracers(struct ftrace_event_call *call,
 
 	list_for_each_entry(tr, &ftrace_trace_arrays, list) {
 		if (file_ops)
-			__trace_add_new_event(call, tr,
-					      &file_ops->id, &file_ops->enable,
-					      &file_ops->filter, &file_ops->format);
+			__trace_add_new_mod_event(call, tr, file_ops);
 		else
 			__trace_add_new_event(call, tr,
 					      &ftrace_event_id_fops,
-- 
1.7.10.4



  parent reply	other threads:[~2013-03-04 17:27 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-04 17:20 [for-next][PATCH 0/6] tracing: memory savings and minor fixes Steven Rostedt
2013-03-04 17:20 ` [for-next][PATCH 1/6] tracing: Add a helper function for event print functions Steven Rostedt
2013-03-04 17:20 ` [for-next][PATCH 2/6] tracing: Annotate event field-defining functions with __init Steven Rostedt
2013-03-04 17:20 ` [for-next][PATCH 3/6] tracing/syscalls: Annotate " Steven Rostedt
2013-03-04 17:20 ` [for-next][PATCH 4/6] tracing: Add __per_cpu annotation to trace array percpu data pointer Steven Rostedt
2013-03-04 17:20 ` Steven Rostedt [this message]
2013-03-04 17:20 ` [for-next][PATCH 6/6] tracing: Fix some section mismatch warnings Steven Rostedt

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20130304172616.025880747@goodmis.org \
    --to=rostedt@goodmis.org \
    --cc=akpm@linux-foundation.org \
    --cc=fengguang.wu@intel.com \
    --cc=fweisbec@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox