linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Chase Douglas <chase.douglas@canonical.com>
To: linux-kernel@vger.kernel.org
Cc: Steven Rostedt <rostedt@goodmis.org>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>,
	Randy Dunlap <randy.dunlap@oracle.com>
Subject: [PATCH 1/3 v4] Add tracing_off_event() to stop tracing when a bug or warning occur
Date: Wed, 14 Apr 2010 12:20:14 -0400	[thread overview]
Message-ID: <1271262016-18650-1-git-send-email-chase.douglas@canonical.com> (raw)

The tracing_off_event() function calls tracing_off() to stop tracing
when an event occurs. By default, only BUG-type events stop tracing,
while WARNING type events do not. This is controlled through the
tracing_off={none,warn,bug} commandline parameter.

Call this function from bug and warning event handlers to enable a user
to debug their kernel by starting a trace, hitting an event, and then
retrieving trace info knowing that the trace was stopped right after the
event was hit.

Signed-off-by: Chase Douglas <chase.douglas@canonical.com>
---
 include/linux/kernel.h     |    7 +++++++
 kernel/trace/ring_buffer.c |   34 ++++++++++++++++++++++++++++++++++
 2 files changed, 41 insertions(+), 0 deletions(-)

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 9365227..80c67ad 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -478,16 +478,23 @@ static inline char *pack_hex_byte(char *buf, u8 byte)
  *
  * Most likely, you want to use tracing_on/tracing_off.
  */
+enum trace_off_event {
+	TRACE_EVENT_BUG = 0,
+	TRACE_EVENT_WARN,
+};
+
 #ifdef CONFIG_RING_BUFFER
 void tracing_on(void);
 void tracing_off(void);
 /* trace_off_permanent stops recording with no way to bring it back */
 void tracing_off_permanent(void);
+void tracing_off_event(enum trace_off_event event);
 int tracing_is_on(void);
 #else
 static inline void tracing_on(void) { }
 static inline void tracing_off(void) { }
 static inline void tracing_off_permanent(void) { }
+static inline void tracing_off_event(enum trace_off_event event) { }
 static inline int tracing_is_on(void) { return 0; }
 #endif
 #ifdef CONFIG_TRACING
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 41ca394..f6be195 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -194,6 +194,40 @@ void tracing_off_permanent(void)
 	set_bit(RB_BUFFERS_DISABLED_BIT, &ring_buffer_flags);
 }
 
+static enum trace_off_event tracing_off_event_ctrl = TRACE_EVENT_BUG;
+
+/**
+ * tracing_off_event - turn off tracing depending on event type
+ * @event: type of event that occurred
+ *
+ * This function checks the event type to determine whether tracing should be
+ * disabled. Useful for disabling tracing on bugs or warnings.
+ */
+void tracing_off_event(enum trace_off_event event)
+{
+	if (event <= tracing_off_event_ctrl)
+		tracing_off();
+}
+EXPORT_SYMBOL_GPL(tracing_off_event);
+
+static int __init tracing_off_event_setup(char *str)
+{
+	if (!strcmp("none", str))
+		tracing_off_event_ctrl = -1;
+	else if (!strcmp("bug", str))
+		tracing_off_event_ctrl = TRACE_EVENT_BUG;
+	else if (!strcmp("warn", str))
+		tracing_off_event_ctrl = TRACE_EVENT_WARN;
+	else {
+		printk(KERN_NOTICE "Invalid value passed for tracing_off parameter\n");
+		return 1;
+	}
+
+	return 0;
+}
+
+__setup("tracing_off=", tracing_off_event_setup);
+
 /**
  * tracing_is_on - show state of ring buffers enabled
  */
-- 
1.7.0


             reply	other threads:[~2010-04-14 16:20 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-04-14 16:20 Chase Douglas [this message]
2010-04-14 16:20 ` [PATCH 2/3] Add tracing_off_event() calls to BUG() and WARN() paths Chase Douglas
2010-04-15 20:57   ` Frederic Weisbecker
2010-04-15 21:49     ` Chase Douglas
2010-04-15 23:15       ` Frederic Weisbecker
2010-04-16  4:13         ` Chase Douglas
2010-04-14 16:20 ` [PATCH 3/3] Stop tracing on a schedule bug Chase Douglas
2010-04-15 21:03   ` Frederic Weisbecker
2010-04-15 21:45     ` Chase Douglas
2010-04-15 23:01       ` Thomas Gleixner
2010-04-15 23:10         ` Frederic Weisbecker
2010-04-15 23:27         ` Chase Douglas
2010-04-15 23:50           ` Thomas Gleixner
2010-04-16  0:21             ` Thomas Gleixner
2010-04-16  1:49               ` Frederic Weisbecker
2010-04-16 16:41                 ` Steven Rostedt
2010-04-16  4:01               ` Chase Douglas
2010-04-16 16:46                 ` Steven Rostedt
2010-04-16 17:14                   ` Chase Douglas
2010-04-16 18:14                   ` Frederic Weisbecker
2010-04-16 19:58                   ` Thomas Gleixner
2010-04-19 22:30               ` Chase Douglas
2010-04-16  3:52             ` Chase Douglas
2010-04-15 20:13 ` [PATCH 1/3 v4] Add tracing_off_event() to stop tracing when a bug or warning occur Frederic Weisbecker
  -- strict thread matches above, loose matches on Subject: below --
2010-03-22 20:12 Chase Douglas

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=1271262016-18650-1-git-send-email-chase.douglas@canonical.com \
    --to=chase.douglas@canonical.com \
    --cc=fweisbec@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=randy.dunlap@oracle.com \
    --cc=rostedt@goodmis.org \
    --cc=tglx@linutronix.de \
    /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;
as well as URLs for NNTP newsgroup(s).