Skip to contentSkip to conversation

PM/runtime: introduce trace points for tracing rpm_* functions

RFC PATCH In review [email protected] 3 participants 8 messages Sep 22, 2011 – Sep 27, 2011

Subsystems (5)
lore
Share & tools
lore
#0 lore raw

RFC PATCH 1/2 PM/runtime: introduce trace points for tracing rpm_* functions

From: Ming Lei <[email protected]>

This patch introduces 3 trace points to prepare for tracing rpm_idle/rpm_suspend/rpm_resume functions, so we can use these trace points to replace the current dev_dbg().

---
 include/trace/events/rpm.h |   98 ++++++++++++++++++++++++++++++++++++++++++++
 kernel/trace/Makefile      |    1 +
 kernel/trace/rpm-traces.c  |   21 +++++++++
 3 files changed, 120 insertions(+), 0 deletions(-)
 create mode 100644 include/trace/events/rpm.h
 create mode 100644 kernel/trace/rpm-traces.c
include/trace/events/rpm.h new file +98 -0
new file mode 100644
index 0000000..ecd40bb
@@ -0,0 +1,98 @@
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM rpm
+
+#if !defined(_TRACE_RUNTIME_POWER_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_RUNTIME_POWER_H
+
+#include <linux/ktime.h>
+#include <linux/tracepoint.h>
+#include <linux/device.h>
+
+/*
+ * The rpm_internal events are used for tracing some important
+ * runtime pm internal functions.
+ */
+DECLARE_EVENT_CLASS(rpm_internal,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags),
+
+	TP_STRUCT__entry(
+		__field(        const char *,   name            )
+		__field(        int,            flags           )
+		__field(        int ,   	usage_count	)
+		__field(        int ,   	disable_depth   )
+		__field(        int ,   	runtime_auto	)
+		__field(        int ,   	request_pending	)
+		__field(        int ,   	irq_safe	)
+		__field(        int ,   	child_count 	)
+	),
+
+	TP_fast_assign(
+		__entry->name = dev_name(dev);
+		__entry->flags = flags;
+		__entry->usage_count = atomic_read(
+			&dev->power.usage_count);
+		__entry->disable_depth = dev->power.disable_depth;
+		__entry->runtime_auto = dev->power.runtime_auto;
+		__entry->request_pending = dev->power.request_pending;
+		__entry->irq_safe = dev->power.irq_safe;
+		__entry->child_count = atomic_read(
+			&dev->power.child_count);
+	),
+
+	TP_printk("%s flags-%x cnt-%-2d dep-%-2d auto-%-1d p-%-1d"
+			" irq-%-1d child-%d",
+			__entry->name, __entry->flags,
+			__entry->usage_count,
+			__entry->disable_depth,
+			__entry->runtime_auto,
+			__entry->request_pending,
+			__entry->irq_safe,
+			__entry->child_count
+		 )
+);
+DEFINE_EVENT(rpm_internal, rpm_suspend,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_resume,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_idle,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags)
+);
+
+TRACE_EVENT(rpm_return_int,
+	TP_PROTO(struct device *dev, const char *func, int ret),
+	TP_ARGS(dev, func, ret),
+
+	TP_STRUCT__entry(
+		__field(	const char *,		name		)
+		__field(	const char *,		func		)
+		__field(	int,			ret		)
+	),
+
+	TP_fast_assign(
+		__entry->name = dev_name(dev);
+		__entry->ret = ret;
+		__entry->func = func;
+	),
+
+	TP_printk("%s:%s ret=%d", __entry->func, __entry->name, __entry->ret)
+);
+
+#endif /* _TRACE_RUNTIME_POWER_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
kernel/trace/Makefile +1 -0
index 761c510..56bdab5 100644
@@ -53,6 +53,7 @@ endif
 obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o
 obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o
 obj-$(CONFIG_TRACEPOINTS) += power-traces.o
+obj-$(CONFIG_TRACEPOINTS) += rpm-traces.o
 ifeq ($(CONFIG_TRACING),y)
 obj-$(CONFIG_KGDB_KDB) += trace_kdb.o
 endif
kernel/trace/rpm-traces.c new file +21 -0
new file mode 100644
index 0000000..998949e
@@ -0,0 +1,21 @@
+/*
+ * Power trace points
+ *
+ * Copyright (C) 2009 Ming Lei <[email protected]>
+ */
+
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/usb.h>
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/rpm.h>
+
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_return_int);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_idle);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_suspend);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_resume);
+
Signature · 1 line
1.7.5.4
#1 lore raw

RFC PATCH 2/2 PM/runtime: replace dev_dbg with trace_rpm_*

From: Ming Lei <[email protected]>

This patch replaces dev_dbg with trace_rpm_* inside the three important functions:

rpm_idle
rpm_suspend
rpm_resume

Trace points have the below advantages compared with dev_dbg:

- trace points include much runtime information(such as running cpu, current task, ...)

- most of linux distributions may disable "verbose debug" driver debug compile switch, so it is very difficult to report/debug runtime pm related problems from distribution users without this kind of debug information.

- for upstream kernel users, enableing the debug switch will produce many useless "rpm_resume" output, and it is very noise.

- dev_dbg inside rpm_suspend/rpm_resume may have some effects on runtime pm behaviour of console devicer

---
 drivers/base/power/runtime.c |   11 +++++++----
 1 files changed, 7 insertions(+), 4 deletions(-)
drivers/base/power/runtime.c +7 -4
index acb3f83..f41469a 100644
@@ -9,6 +9,7 @@
 
 #include <linux/sched.h>
 #include <linux/pm_runtime.h>
+#include <trace/events/rpm.h>
 #include "power.h"
 
 static int rpm_resume(struct device *dev, int rpmflags);
@@ -171,6 +172,7 @@ static int rpm_idle(struct device *dev, int rpmflags)
 	int (*callback)(struct device *);
 	int retval;
 
+	trace_rpm_idle(dev, rpmflags);
 	retval = rpm_check_suspend_allowed(dev);
 	if (retval < 0)
 		;	/* Conditions are wrong. */
@@ -243,6 +245,7 @@ static int rpm_idle(struct device *dev, int rpmflags)
 	wake_up_all(&dev->power.wait_queue);
 
  out:
+	trace_rpm_return_int(dev, __func__, retval);
 	return retval;
 }
 
@@ -295,7 +298,7 @@ static int rpm_suspend(struct device *dev, int rpmflags)
 	struct device *parent = NULL;
 	int retval;
 
-	dev_dbg(dev, "%s flags 0x%x\n", __func__, rpmflags);
+	trace_rpm_suspend(dev, rpmflags);
 
  repeat:
 	retval = rpm_check_suspend_allowed(dev);
@@ -430,7 +433,7 @@ static int rpm_suspend(struct device *dev, int rpmflags)
 	}
 
  out:
-	dev_dbg(dev, "%s returns %d\n", __func__, retval);
+	trace_rpm_return_int(dev, __func__, retval);
 
 	return retval;
 }
@@ -459,7 +462,7 @@ static int rpm_resume(struct device *dev, int rpmflags)
 	struct device *parent = NULL;
 	int retval = 0;
 
-	dev_dbg(dev, "%s flags 0x%x\n", __func__, rpmflags);
+	trace_rpm_resume(dev, rpmflags);
 
  repeat:
 	if (dev->power.runtime_error)
@@ -615,7 +618,7 @@ static int rpm_resume(struct device *dev, int rpmflags)
 		spin_lock_irq(&dev->power.lock);
 	}
 
-	dev_dbg(dev, "%s returns %d\n", __func__, retval);
+	trace_rpm_return_int(dev, __func__, retval);
 
 	return retval;
 }
Signature · 1 line
1.7.5.4
#2 lore raw
In reply to [email protected]
Quoted 7 lines · On Thursday, September 22, 2011, [email protected] wrote:

From: Ming Lei <[email protected]>

This patch introduces 3 trace points to prepare for tracing rpm_idle/rpm_suspend/rpm_resume functions, so we can use these trace points to replace the current dev_dbg().

I like the patches and would like to push them for 3.2, but I need to know if the tracing people are OK with them. Steven, Ingo?

Thanks,
Rafael

Quoted 152 lines
---
 include/trace/events/rpm.h |   98 ++++++++++++++++++++++++++++++++++++++++++++
 kernel/trace/Makefile      |    1 +
 kernel/trace/rpm-traces.c  |   21 +++++++++
 3 files changed, 120 insertions(+), 0 deletions(-)
 create mode 100644 include/trace/events/rpm.h
 create mode 100644 kernel/trace/rpm-traces.c
include/trace/events/rpm.h new file +98 -0
new file mode 100644
index 0000000..ecd40bb
@@ -0,0 +1,98 @@
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM rpm
+
+#if !defined(_TRACE_RUNTIME_POWER_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_RUNTIME_POWER_H
+
+#include <linux/ktime.h>
+#include <linux/tracepoint.h>
+#include <linux/device.h>
+
+/*
+ * The rpm_internal events are used for tracing some important
+ * runtime pm internal functions.
+ */
+DECLARE_EVENT_CLASS(rpm_internal,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags),
+
+	TP_STRUCT__entry(
+		__field(        const char *,   name            )
+		__field(        int,            flags           )
+		__field(        int ,   	usage_count	)
+		__field(        int ,   	disable_depth   )
+		__field(        int ,   	runtime_auto	)
+		__field(        int ,   	request_pending	)
+		__field(        int ,   	irq_safe	)
+		__field(        int ,   	child_count 	)
+	),
+
+	TP_fast_assign(
+		__entry->name = dev_name(dev);
+		__entry->flags = flags;
+		__entry->usage_count = atomic_read(
+			&dev->power.usage_count);
+		__entry->disable_depth = dev->power.disable_depth;
+		__entry->runtime_auto = dev->power.runtime_auto;
+		__entry->request_pending = dev->power.request_pending;
+		__entry->irq_safe = dev->power.irq_safe;
+		__entry->child_count = atomic_read(
+			&dev->power.child_count);
+	),
+
+	TP_printk("%s flags-%x cnt-%-2d dep-%-2d auto-%-1d p-%-1d"
+			" irq-%-1d child-%d",
+			__entry->name, __entry->flags,
+			__entry->usage_count,
+			__entry->disable_depth,
+			__entry->runtime_auto,
+			__entry->request_pending,
+			__entry->irq_safe,
+			__entry->child_count
+		 )
+);
+DEFINE_EVENT(rpm_internal, rpm_suspend,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_resume,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_idle,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags)
+);
+
+TRACE_EVENT(rpm_return_int,
+	TP_PROTO(struct device *dev, const char *func, int ret),
+	TP_ARGS(dev, func, ret),
+
+	TP_STRUCT__entry(
+		__field(	const char *,		name		)
+		__field(	const char *,		func		)
+		__field(	int,			ret		)
+	),
+
+	TP_fast_assign(
+		__entry->name = dev_name(dev);
+		__entry->ret = ret;
+		__entry->func = func;
+	),
+
+	TP_printk("%s:%s ret=%d", __entry->func, __entry->name, __entry->ret)
+);
+
+#endif /* _TRACE_RUNTIME_POWER_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
kernel/trace/Makefile +1 -0
index 761c510..56bdab5 100644
@@ -53,6 +53,7 @@ endif
 obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o
 obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o
 obj-$(CONFIG_TRACEPOINTS) += power-traces.o
+obj-$(CONFIG_TRACEPOINTS) += rpm-traces.o
 ifeq ($(CONFIG_TRACING),y)
 obj-$(CONFIG_KGDB_KDB) += trace_kdb.o
 endif
kernel/trace/rpm-traces.c new file +21 -0
new file mode 100644
index 0000000..998949e
@@ -0,0 +1,21 @@
+/*
+ * Power trace points
+ *
+ * Copyright (C) 2009 Ming Lei <[email protected]>
+ */
+
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/usb.h>
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/rpm.h>
+
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_return_int);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_idle);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_suspend);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_resume);
+
maintainer
#3 lore raw
In reply to [email protected]
Quoted 45 lines · On Thu, 2011-09-22 at 12:11 +0800, [email protected] wrote:

From: Ming Lei <[email protected]>

This patch introduces 3 trace points to prepare for tracing rpm_idle/rpm_suspend/rpm_resume functions, so we can use these trace points to replace the current dev_dbg().

---
 include/trace/events/rpm.h |   98 ++++++++++++++++++++++++++++++++++++++++++++
 kernel/trace/Makefile      |    1 +
 kernel/trace/rpm-traces.c  |   21 +++++++++
 3 files changed, 120 insertions(+), 0 deletions(-)
 create mode 100644 include/trace/events/rpm.h
 create mode 100644 kernel/trace/rpm-traces.c
include/trace/events/rpm.h new file +24 -0
new file mode 100644
index 0000000..ecd40bb
@@ -0,0 +1,98 @@
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM rpm
+
+#if !defined(_TRACE_RUNTIME_POWER_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_RUNTIME_POWER_H
+
+#include <linux/ktime.h>
+#include <linux/tracepoint.h>
+#include <linux/device.h>
+
+/*
+ * The rpm_internal events are used for tracing some important
+ * runtime pm internal functions.
+ */
+DECLARE_EVENT_CLASS(rpm_internal,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags),
+
+	TP_STRUCT__entry(
+		__field(        const char *,   name            )
+		__field(        int,            flags           )

include/trace/events/rpm.h line 34

Quoted 10 lines
include/trace/events/rpm.h new file +10 -0
+		__field(        int ,   	usage_count	)
+		__field(        int ,   	disable_depth   )
+		__field(        int ,   	runtime_auto	)
+		__field(        int ,   	request_pending	)
+		__field(        int ,   	irq_safe	)
+		__field(        int ,   	child_count 	)
+	),
+
+	TP_fast_assign(
+		__entry->name = dev_name(dev);

Don't use pointers, you must copy the name here. Otherwise, if the trace is read after the device has left, you will be referencing a invalid pointer and possible crash the kernel.

We have ways to express dynamic string fields. Use them.

Quoted 43 lines
include/trace/events/rpm.h new file +43 -0
+		__entry->flags = flags;
+		__entry->usage_count = atomic_read(
+			&dev->power.usage_count);
+		__entry->disable_depth = dev->power.disable_depth;
+		__entry->runtime_auto = dev->power.runtime_auto;
+		__entry->request_pending = dev->power.request_pending;
+		__entry->irq_safe = dev->power.irq_safe;
+		__entry->child_count = atomic_read(
+			&dev->power.child_count);
+	),
+
+	TP_printk("%s flags-%x cnt-%-2d dep-%-2d auto-%-1d p-%-1d"
+			" irq-%-1d child-%d",
+			__entry->name, __entry->flags,
+			__entry->usage_count,
+			__entry->disable_depth,
+			__entry->runtime_auto,
+			__entry->request_pending,
+			__entry->irq_safe,
+			__entry->child_count
+		 )
+);
+DEFINE_EVENT(rpm_internal, rpm_suspend,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_resume,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_idle,
+
+	TP_PROTO(struct device *dev, int flags),
+
+	TP_ARGS(dev, flags)
+);
+
+TRACE_EVENT(rpm_return_int,
+	TP_PROTO(struct device *dev, const char *func, int ret),

include/trace/events/rpm.h line 87

Quoted 10 lines
include/trace/events/rpm.h new file +10 -0
+	TP_ARGS(dev, func, ret),
+
+	TP_STRUCT__entry(
+		__field(	const char *,		name		)
+		__field(	const char *,		func		)
+		__field(	int,			ret		)
+	),
+
+	TP_fast_assign(
+		__entry->name = dev_name(dev);

Same here.

include/trace/events/rpm.h line 89

Quoted 2 lines
include/trace/events/rpm.h new file +2 -0
+		__entry->ret = ret;
+		__entry->func = func;

And here. But as this is just __func__, you could use %pS and save the pointer to the function. This is acceptable as kallsyms will will used to find the function, and instead pass in __THIS_IP__ to the tracepoint.

-- Steve

Quoted 48 lines
include/trace/events/rpm.h new file +9 -0
+	),
+
+	TP_printk("%s:%s ret=%d", __entry->func, __entry->name, __entry->ret)
+);
+
+#endif /* _TRACE_RUNTIME_POWER_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
kernel/trace/Makefile +1 -0
index 761c510..56bdab5 100644
@@ -53,6 +53,7 @@ endif
 obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o
 obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o
 obj-$(CONFIG_TRACEPOINTS) += power-traces.o
+obj-$(CONFIG_TRACEPOINTS) += rpm-traces.o
 ifeq ($(CONFIG_TRACING),y)
 obj-$(CONFIG_KGDB_KDB) += trace_kdb.o
 endif
kernel/trace/rpm-traces.c new file +21 -0
new file mode 100644
index 0000000..998949e
@@ -0,0 +1,21 @@
+/*
+ * Power trace points
+ *
+ * Copyright (C) 2009 Ming Lei <[email protected]>
+ */
+
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/usb.h>
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/rpm.h>
+
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_return_int);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_idle);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_suspend);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_resume);
+
author
#4 lore raw

Hi,

Quoted 52 lines · Steven Rostedt
Quoted 51 lines · On Thu, 2011-09-22 at 12:11 +0800, [email protected] wrote:

From: Ming Lei <[email protected]>

This patch introduces 3 trace points to prepare for tracing rpm_idle/rpm_suspend/rpm_resume functions, so we can use these trace points to replace the current dev_dbg().

---
 include/trace/events/rpm.h |   98 ++++++++++++++++++++++++++++++++++++++++++++ kernel/trace/Makefile      |    1 +
 kernel/trace/rpm-traces.c  |   21 +++++++++

 3 files changed, 120 insertions(+), 0 deletions(-)

 create mode 100644 include/trace/events/rpm.h
 create mode 100644 kernel/trace/rpm-traces.c

include/trace/events/rpm.h new file +30 -0
new file mode 100644
index 0000000..ecd40bb
@@ -0,0 +1,98 @@
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM rpm
+
+#if !defined(_TRACE_RUNTIME_POWER_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_RUNTIME_POWER_H
+
+#include <linux/ktime.h>
+#include <linux/tracepoint.h>
+#include <linux/device.h>
+
+/*
+ * The rpm_internal events are used for tracing some important
+ * runtime pm internal functions.
+ */
+DECLARE_EVENT_CLASS(rpm_internal,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags),
+
+     TP_STRUCT__entry(
+             __field(        const char *,   name            )
+             __field(        int,            flags           )
+             __field(        int ,           usage_count     )
+             __field(        int ,           disable_depth   )
+             __field(        int ,           runtime_auto    )
+             __field(        int ,           request_pending )
+             __field(        int ,           irq_safe        )
+             __field(        int ,           child_count     )
Quoted 8 lines

include/trace/events/rpm.h line 34

Quoted 4 lines
include/trace/events/rpm.h new file +4 -0
+     ),
+
+     TP_fast_assign(
+             __entry->name = dev_name(dev);

Don't use pointers, you must copy the name here. Otherwise, if the trace is read after the device has left, you will be referencing a invalid pointer and possible crash the kernel.

Yes, you right, but __get_str() may introduce some extra loading, could it be changed to below just like what dev_printk does?

                   __entry->name = dev ? dev_name(dev) : NULL;
Quoted 59 lines

We have ways to express dynamic string fields. Use them.

Quoted 43 lines
include/trace/events/rpm.h new file +43 -0
+             __entry->flags = flags;
+             __entry->usage_count = atomic_read(
+                     &dev->power.usage_count);
+             __entry->disable_depth = dev->power.disable_depth;
+             __entry->runtime_auto = dev->power.runtime_auto;
+             __entry->request_pending = dev->power.request_pending;
+             __entry->irq_safe = dev->power.irq_safe;
+             __entry->child_count = atomic_read(
+                     &dev->power.child_count);
+     ),
+
+     TP_printk("%s flags-%x cnt-%-2d dep-%-2d auto-%-1d p-%-1d"
+                     " irq-%-1d child-%d",
+                     __entry->name, __entry->flags,
+                     __entry->usage_count,
+                     __entry->disable_depth,
+                     __entry->runtime_auto,
+                     __entry->request_pending,
+                     __entry->irq_safe,
+                     __entry->child_count
+              )
+);
+DEFINE_EVENT(rpm_internal, rpm_suspend,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_resume,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_idle,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags)
+);
+
+TRACE_EVENT(rpm_return_int,
+     TP_PROTO(struct device *dev, const char *func, int ret),

include/trace/events/rpm.h line 87

Quoted 10 lines
include/trace/events/rpm.h new file +10 -0
+     TP_ARGS(dev, func, ret),
+
+     TP_STRUCT__entry(
+             __field(        const char *,           name            )
+             __field(        const char *,           func            )
+             __field(        int,                    ret             )
+     ),
+
+     TP_fast_assign(
+             __entry->name = dev_name(dev);

Same here.

Quoted 6 lines

include/trace/events/rpm.h line 89

Quoted 2 lines
include/trace/events/rpm.h new file +2 -0
+             __entry->ret = ret;
+             __entry->func = func;

And here. But as this is just __func__, you could use %pS and save the pointer to the function. This is acceptable as kallsyms will will used to find the function, and instead pass in __THIS_IP__ to the tracepoint.

IMO, __func__ is better than %pS here because %pS will dump some offset info (such as runtime_suspend+0x0/0x158), but which is not needed in this case.( we just need to make return value to match the previous function name, such as below)

rpm_suspend: omap_hsmmc.0 flags=9
...
rpm_return_int: rpm_suspend:omap_hsmmc.0 ret=0

Quoted 44 lines

-- Steve

Quoted 40 lines
include/trace/events/rpm.h new file +9 -0
+     ),
+
+     TP_printk("%s:%s ret=%d", __entry->func, __entry->name, __entry->ret)
+);
+
+#endif /* _TRACE_RUNTIME_POWER_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
kernel/trace/Makefile +0 -0
index 761c510..56bdab5 100644
@@ -53,6 +53,7 @@ endif

 obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o obj-$(CONFIG_TRACEPOINTS) += power-traces.o +obj-$(CONFIG_TRACEPOINTS) += rpm-traces.o ifeq ($(CONFIG_TRACING),y) obj-$(CONFIG_KGDB_KDB) += trace_kdb.o
 endif

kernel/trace/rpm-traces.c new file +13 -0
new file mode 100644
index 0000000..998949e
@@ -0,0 +1,21 @@
+/*
+ * Power trace points
+ *
+ * Copyright (C) 2009 Ming Lei <[email protected]>
+ */
+
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/usb.h>
+
Quoted 11 lines

kernel/trace/rpm-traces.c line 21

Quoted 8 lines
kernel/trace/rpm-traces.c new file +8 -0
+#define CREATE_TRACE_POINTS
+#include <trace/events/rpm.h>
+
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_return_int);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_idle);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_suspend);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_resume);
+

thanks,

Signature · 1 line
Ming Lei
maintainer
#5 lore raw
Quoted 64 lines · Ming Lei

Hi,

Quoted 52 lines · Steven Rostedt
Quoted 51 lines · On Thu, 2011-09-22 at 12:11 +0800, [email protected] wrote:

From: Ming Lei <[email protected]>

This patch introduces 3 trace points to prepare for tracing rpm_idle/rpm_suspend/rpm_resume functions, so we can use these trace points to replace the current dev_dbg().

---
 include/trace/events/rpm.h |   98 ++++++++++++++++++++++++++++++++++++++++++++
 kernel/trace/Makefile      |    1 +
 kernel/trace/rpm-traces.c  |   21 +++++++++
 3 files changed, 120 insertions(+), 0 deletions(-)
 create mode 100644 include/trace/events/rpm.h
 create mode 100644 kernel/trace/rpm-traces.c
include/trace/events/rpm.h new file +30 -0
new file mode 100644
index 0000000..ecd40bb
@@ -0,0 +1,98 @@
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM rpm
+
+#if !defined(_TRACE_RUNTIME_POWER_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_RUNTIME_POWER_H
+
+#include <linux/ktime.h>
+#include <linux/tracepoint.h>
+#include <linux/device.h>
+
+/*
+ * The rpm_internal events are used for tracing some important
+ * runtime pm internal functions.
+ */
+DECLARE_EVENT_CLASS(rpm_internal,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags),
+
+     TP_STRUCT__entry(
+             __field(        const char *,   name            )
+             __field(        int,            flags           )
+             __field(        int ,           usage_count     )
+             __field(        int ,           disable_depth   )
+             __field(        int ,           runtime_auto    )
+             __field(        int ,           request_pending )
+             __field(        int ,           irq_safe        )
+             __field(        int ,           child_count     )
Quoted 8 lines

include/trace/events/rpm.h line 34

Quoted 4 lines
include/trace/events/rpm.h new file +4 -0
+     ),
+
+     TP_fast_assign(
+             __entry->name = dev_name(dev);

Don't use pointers, you must copy the name here. Otherwise, if the trace is read after the device has left, you will be referencing a invalid pointer and possible crash the kernel.

Quoted 4 lines

Yes, you right, but __get_str() may introduce some extra loading, could it be changed to below just like what dev_printk does?

                   __entry->name = dev ? dev_name(dev) : NULL;

Doesn't help. You just stored a pointer into the buffer. Now if the device's module is removed, when you go and read the buffer you will be reading a string that no longer exists.

Not to mention, this will never work with tools such as perf or trace-cmd, that will just get a pointer address and will try to read a string that points inside the kernel.

Quoted 63 lines
Quoted 63 lines

We have ways to express dynamic string fields. Use them.

Quoted 43 lines
include/trace/events/rpm.h new file +43 -0
+             __entry->flags = flags;
+             __entry->usage_count = atomic_read(
+                     &dev->power.usage_count);
+             __entry->disable_depth = dev->power.disable_depth;
+             __entry->runtime_auto = dev->power.runtime_auto;
+             __entry->request_pending = dev->power.request_pending;
+             __entry->irq_safe = dev->power.irq_safe;
+             __entry->child_count = atomic_read(
+                     &dev->power.child_count);
+     ),
+
+     TP_printk("%s flags-%x cnt-%-2d dep-%-2d auto-%-1d p-%-1d"
+                     " irq-%-1d child-%d",
+                     __entry->name, __entry->flags,
+                     __entry->usage_count,
+                     __entry->disable_depth,
+                     __entry->runtime_auto,
+                     __entry->request_pending,
+                     __entry->irq_safe,
+                     __entry->child_count
+              )
+);
+DEFINE_EVENT(rpm_internal, rpm_suspend,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_resume,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_idle,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags)
+);
+
+TRACE_EVENT(rpm_return_int,
+     TP_PROTO(struct device *dev, const char *func, int ret),

include/trace/events/rpm.h line 87

Quoted 10 lines
include/trace/events/rpm.h new file +10 -0
+     TP_ARGS(dev, func, ret),
+
+     TP_STRUCT__entry(
+             __field(        const char *,           name            )
+             __field(        const char *,           func            )
+             __field(        int,                    ret             )
+     ),
+
+     TP_fast_assign(
+             __entry->name = dev_name(dev);

Same here.

Quoted 2 lines
include/trace/events/rpm.h new file +2 -0
+             __entry->ret = ret;
+             __entry->func = func;
Quoted 8 lines
Quoted 3 lines

And here. But as this is just __func__, you could use %pS and save the pointer to the function. This is acceptable as kallsyms will will used to find the function, and instead pass in __THIS_IP__ to the tracepoint.

IMO, __func__ is better than %pS here because %pS will dump some offset info (such as runtime_suspend+0x0/0x158), but which is not needed in this case.( we just need to make return value to match the previous function name, such as below)

This will only work with reading the trace data
from /debug/tracing/trace, but not from any tools. The pS *will* work with the tools as the tools load the kallsyms data, and is smart enough to figure this pointer out.

-- Steve

Quoted 63 lines

rpm_suspend: omap_hsmmc.0 flags=9
...
rpm_return_int: rpm_suspend:omap_hsmmc.0 ret=0

Quoted 44 lines

-- Steve

Quoted 40 lines
include/trace/events/rpm.h new file +9 -0
+     ),
+
+     TP_printk("%s:%s ret=%d", __entry->func, __entry->name, __entry->ret)
+);
+
+#endif /* _TRACE_RUNTIME_POWER_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
kernel/trace/Makefile +1 -0
index 761c510..56bdab5 100644
@@ -53,6 +53,7 @@ endif
 obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o
 obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o
 obj-$(CONFIG_TRACEPOINTS) += power-traces.o
+obj-$(CONFIG_TRACEPOINTS) += rpm-traces.o
 ifeq ($(CONFIG_TRACING),y)
 obj-$(CONFIG_KGDB_KDB) += trace_kdb.o
 endif
kernel/trace/rpm-traces.c new file +13 -0
new file mode 100644
index 0000000..998949e
@@ -0,0 +1,21 @@
+/*
+ * Power trace points
+ *
+ * Copyright (C) 2009 Ming Lei <[email protected]>
+ */
+
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/usb.h>
+
Quoted 11 lines

kernel/trace/rpm-traces.c line 21

Quoted 8 lines
kernel/trace/rpm-traces.c new file +8 -0
+#define CREATE_TRACE_POINTS
+#include <trace/events/rpm.h>
+
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_return_int);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_idle);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_suspend);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_resume);
+

thanks,

author
#6 lore raw
Quoted 70 lines · Steven Rostedt
Quoted 64 lines · Ming Lei

Hi,

Quoted 52 lines · Steven Rostedt
Quoted 51 lines · On Thu, 2011-09-22 at 12:11 +0800, [email protected] wrote:

From: Ming Lei <[email protected]>

This patch introduces 3 trace points to prepare for tracing rpm_idle/rpm_suspend/rpm_resume functions, so we can use these trace points to replace the current dev_dbg().

---
 include/trace/events/rpm.h |   98 ++++++++++++++++++++++++++++++++++++++++++++ kernel/trace/Makefile      |    1 +
 kernel/trace/rpm-traces.c  |   21 +++++++++

 3 files changed, 120 insertions(+), 0 deletions(-)

 create mode 100644 include/trace/events/rpm.h
 create mode 100644 kernel/trace/rpm-traces.c

include/trace/events/rpm.h new file +30 -0
new file mode 100644
index 0000000..ecd40bb
@@ -0,0 +1,98 @@
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM rpm
+
+#if !defined(_TRACE_RUNTIME_POWER_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_RUNTIME_POWER_H
+
+#include <linux/ktime.h>
+#include <linux/tracepoint.h>
+#include <linux/device.h>
+
+/*
+ * The rpm_internal events are used for tracing some important
+ * runtime pm internal functions.
+ */
+DECLARE_EVENT_CLASS(rpm_internal,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags),
+
+     TP_STRUCT__entry(
+             __field(        const char *,   name            )
+             __field(        int,            flags           )
+             __field(        int ,           usage_count     )
+             __field(        int ,           disable_depth   )
+             __field(        int ,           runtime_auto    )
+             __field(        int ,           request_pending )
+             __field(        int ,           irq_safe        )
+             __field(        int ,           child_count     )
Quoted 8 lines

include/trace/events/rpm.h line 34

Quoted 4 lines
include/trace/events/rpm.h new file +4 -0
+     ),
+
+     TP_fast_assign(
+             __entry->name = dev_name(dev);

Don't use pointers, you must copy the name here. Otherwise, if the trace is read after the device has left, you will be referencing a invalid pointer and possible crash the kernel.

Quoted 4 lines

Yes, you right, but __get_str() may introduce some extra loading, could it be changed to below just like what dev_printk does?

                   __entry->name = dev ? dev_name(dev) : NULL;

Quoted 7 lines

Doesn't help. You just stored a pointer into the buffer. Now if the device's module is removed, when you go and read the buffer you will be reading a string that no longer exists.

Not to mention, this will never work with tools such as perf or trace-cmd, that will just get a pointer address and will try to read a string that points inside the kernel.

Ok, sounds like dev_name does not work for the case, I will update the patch, but I am wondering why no such problem for dev_dbg? printk still may buffer these for sometime in some cases.

Quoted 73 lines
Quoted 64 lines
Quoted 64 lines

We have ways to express dynamic string fields. Use them.

Quoted 43 lines
include/trace/events/rpm.h new file +43 -0
+             __entry->flags = flags;
+             __entry->usage_count = atomic_read(
+                     &dev->power.usage_count);
+             __entry->disable_depth = dev->power.disable_depth;
+             __entry->runtime_auto = dev->power.runtime_auto;
+             __entry->request_pending = dev->power.request_pending;
+             __entry->irq_safe = dev->power.irq_safe;
+             __entry->child_count = atomic_read(
+                     &dev->power.child_count);
+     ),
+
+     TP_printk("%s flags-%x cnt-%-2d dep-%-2d auto-%-1d p-%-1d"
+                     " irq-%-1d child-%d",
+                     __entry->name, __entry->flags,
+                     __entry->usage_count,
+                     __entry->disable_depth,
+                     __entry->runtime_auto,
+                     __entry->request_pending,
+                     __entry->irq_safe,
+                     __entry->child_count
+              )
+);
+DEFINE_EVENT(rpm_internal, rpm_suspend,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_resume,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags)
+);
+DEFINE_EVENT(rpm_internal, rpm_idle,
+
+     TP_PROTO(struct device *dev, int flags),
+
+     TP_ARGS(dev, flags)
+);
+
+TRACE_EVENT(rpm_return_int,
+     TP_PROTO(struct device *dev, const char *func, int ret),

include/trace/events/rpm.h line 87

Quoted 10 lines
include/trace/events/rpm.h new file +10 -0
+     TP_ARGS(dev, func, ret),
+
+     TP_STRUCT__entry(
+             __field(        const char *,           name            )
+             __field(        const char *,           func            )
+             __field(        int,                    ret             )
+     ),
+
+     TP_fast_assign(
+             __entry->name = dev_name(dev);

Same here.

Quoted 2 lines
include/trace/events/rpm.h new file +2 -0
+             __entry->ret = ret;
+             __entry->func = func;
Quoted 8 lines
Quoted 3 lines

And here. But as this is just __func__, you could use %pS and save the pointer to the function. This is acceptable as kallsyms will will used to find the function, and instead pass in __THIS_IP__ to the tracepoint.

IMO, __func__ is better than %pS here because %pS will dump some offset info (such as runtime_suspend+0x0/0x158), but which is not needed in this case.( we just need to make return value to match the previous function name, such as below)

Quoted 4 lines

This will only work with reading the trace data
from /debug/tracing/trace, but not from any tools. The pS *will* work with the tools as the tools load the kallsyms data, and is smart enough to figure this pointer out.

__func__ points to a const string, which will be stored in kernel global read-only section, I think perf can read correct string from the pointer anytime, don't I?

thanks,
--
Ming Lei

Quoted 69 lines

-- Steve

Quoted 63 lines

          rpm_suspend: omap_hsmmc.0 flags=9
          ...
          rpm_return_int: rpm_suspend:omap_hsmmc.0 ret=0

Quoted 44 lines

-- Steve

Quoted 40 lines
include/trace/events/rpm.h new file +9 -0
+     ),
+
+     TP_printk("%s:%s ret=%d", __entry->func, __entry->name, __entry->ret)
+);
+
+#endif /* _TRACE_RUNTIME_POWER_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
kernel/trace/Makefile +0 -0
index 761c510..56bdab5 100644
@@ -53,6 +53,7 @@ endif

 obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o obj-$(CONFIG_TRACEPOINTS) += power-traces.o +obj-$(CONFIG_TRACEPOINTS) += rpm-traces.o ifeq ($(CONFIG_TRACING),y) obj-$(CONFIG_KGDB_KDB) += trace_kdb.o
 endif

kernel/trace/rpm-traces.c new file +13 -0
new file mode 100644
index 0000000..998949e
@@ -0,0 +1,21 @@
+/*
+ * Power trace points
+ *
+ * Copyright (C) 2009 Ming Lei <[email protected]>
+ */
+
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/workqueue.h>
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/usb.h>
+
Quoted 11 lines

kernel/trace/rpm-traces.c line 21

Quoted 8 lines
kernel/trace/rpm-traces.c new file +8 -0
+#define CREATE_TRACE_POINTS
+#include <trace/events/rpm.h>
+
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_return_int);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_idle);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_suspend);
+EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_resume);
+

thanks,

Signature · 1 line
Ming Lei
maintainer
#7 lore raw
Quoted 8 lines · Ming Lei
Quoted 3 lines

Not to mention, this will never work with tools such as perf or trace-cmd, that will just get a pointer address and will try to read a string that points inside the kernel.

Ok, sounds like dev_name does not work for the case, I will update the patch, but I am wondering why no such problem for dev_dbg? printk still may buffer these for sometime in some cases.

printk will push out to the console immediately. Whatever is in the buffer has already been processed into a string. With tracing, we save the data as defined by TP_fast_assign() and is printed later when the user asks in TP_printk().

Quoted 9 lines
Quoted 5 lines

This will only work with reading the trace data
from /debug/tracing/trace, but not from any tools. The pS *will* work with the tools as the tools load the kallsyms data, and is smart enough to figure this pointer out.

__func__ points to a const string, which will be stored in kernel global read-only section, I think perf can read correct string from the pointer anytime, don't I?

That would be very dangerous to let a userspace app have any access to reading the kernel memory. Not to mention, this data may be sent out over the network to another machine, which must be able to read this data too.

-- Steve