[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <201206162236.21611.rjw@sisk.pl>
Date: Sat, 16 Jun 2012 22:36:21 +0200
From: "Rafael J. Wysocki" <rjw@...k.pl>
To: Greg KH <gregkh@...uxfoundation.org>,
Sameer Nanda <snanda@...omium.org>
Cc: rob@...dley.net, len.brown@...el.com, pavel@....cz,
linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-doc@...r.kernel.org, Steven Rostedt <rostedt@...dmis.org>
Subject: [PATCH] (was: Re: [PATCH v2] power: add knob for printing device resume times)
On Saturday, June 16, 2012, Rafael J. Wysocki wrote:
> On Friday, June 15, 2012, Greg KH wrote:
> > On Fri, Jun 15, 2012 at 12:00:20PM +0200, Rafael J. Wysocki wrote:
> > > On Friday, June 15, 2012, Greg KH wrote:
> > > > On Wed, May 23, 2012 at 09:45:32AM -0700, Sameer Nanda wrote:
> > > > > Added a new knob called /sys/power/pm_print_times. Setting it to 1
> > > > > enables printing of time taken by devices to suspend and resume.
> > > > > Setting it to 0 disables this printing (unless overridden by
> > > > > initcall_debug kernel command line option).
> > > > >
> > > > > Signed-off-by: Sameer Nanda <snanda@...omium.org>
> > > > > cc: Greg KH <gregkh@...uxfoundation.org>
> > > > > cc: Rafael J. Wysocki <rjw@...k.pl>
> > > > > cc: Steven Rostedt <rostedt@...dmis.org>
> > > > > ---
> > > > > Documentation/ABI/testing/sysfs-power | 13 ++++++++++++
> > > > > drivers/base/power/main.c | 4 +-
> > > > > drivers/base/power/power.h | 11 ++++++++++
> > > > > kernel/power/main.c | 34 +++++++++++++++++++++++++++++++++
> > > > > 4 files changed, 60 insertions(+), 2 deletions(-)
> > > >
> > > > This patch fails against the linux-next tree, care to fix it up so that
> > > > I can apply it?
> > >
> > > I'd prefer it to go through my tree if you don't mind.
> >
> > Not at all, feel free to put:
> >
> > Acked-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
> >
> > When it shows up again.
>
> I've just used the original one, it only needed a documentation merge conflict
> fixed.
>
> Queued up for 3.6.
I propose the following patch on top of this one. The details are in the
changelog.
If anyone has any objections, please let me know.
Thanks,
Rafael
---
From: Rafael J. Wysocki <rjw@...k.pl>
Subject: PM / Sleep: Separate printing suspend times from initcall_debug
Change the behavior of the newly introduced
/sys/power/pm_print_times attribute so that its initial value
depends on initcall_debug, but setting it to 0 will cause device
suspend/resume times not to be printed, even if initcall_debug has
been set. This way, the people who use initcall_debug for reasons
other than PM debugging will be able to switch the suspend/resume
times printing off, if need be.
Signed-off-by: Rafael J. Wysocki <rjw@...k.pl>
---
drivers/base/power/main.c | 4 ++--
drivers/base/power/power.h | 11 -----------
include/linux/suspend.h | 4 ++++
kernel/power/main.c | 14 +++++++++++---
4 files changed, 17 insertions(+), 16 deletions(-)
Index: linux/kernel/power/main.c
===================================================================
--- linux.orig/kernel/power/main.c
+++ linux/kernel/power/main.c
@@ -139,7 +139,7 @@ power_attr(pm_test);
* show() returns whether printing of suspend and resume times is enabled.
* store() accepts 0 or 1. 0 disables printing and 1 enables it.
*/
-int pm_print_times_enabled;
+bool pm_print_times_enabled;
static ssize_t pm_print_times_show(struct kobject *kobj,
struct kobj_attribute *attr, char *buf)
@@ -159,12 +159,19 @@ static ssize_t pm_print_times_store(stru
if (val > 1)
return -EINVAL;
- pm_print_times_enabled = val;
+ pm_print_times_enabled = !!val;
return n;
}
+static inline void pm_print_times_init(void)
+{
+ pm_print_times_enabled = !!initcall_debug;
+}
+
power_attr(pm_print_times);
-#endif /* CONFIG_PM_DEBUG */
+#else /* !CONFIG_PM_DEBUG */
+static inline void pm_print_times_init(void) {}
+#endif /* !CONFIG_PM_DEBUG */
#ifdef CONFIG_DEBUG_FS
static char *suspend_step_name(enum suspend_stat_step step)
@@ -599,6 +606,7 @@ static int __init pm_init(void)
error = sysfs_create_group(power_kobj, &attr_group);
if (error)
return error;
+ pm_print_times_init();
return pm_autosleep_init();
}
Index: linux/include/linux/suspend.h
===================================================================
--- linux.orig/include/linux/suspend.h
+++ linux/include/linux/suspend.h
@@ -387,6 +387,8 @@ static inline void unlock_system_sleep(v
mutex_unlock(&pm_mutex);
}
+extern bool pm_print_times_enabled;
+
#else /* !CONFIG_PM_SLEEP */
static inline int register_pm_notifier(struct notifier_block *nb)
@@ -406,6 +408,8 @@ static inline bool pm_wakeup_pending(voi
static inline void lock_system_sleep(void) {}
static inline void unlock_system_sleep(void) {}
+#define pm_print_times_enabled (false)
+
#endif /* !CONFIG_PM_SLEEP */
#ifdef CONFIG_PM_AUTOSLEEP
Index: linux/drivers/base/power/main.c
===================================================================
--- linux.orig/drivers/base/power/main.c
+++ linux/drivers/base/power/main.c
@@ -166,7 +166,7 @@ static ktime_t initcall_debug_start(stru
{
ktime_t calltime = ktime_set(0, 0);
- if (pm_print_times) {
+ if (pm_print_times_enabled) {
pr_info("calling %s+ @ %i, parent: %s\n",
dev_name(dev), task_pid_nr(current),
dev->parent ? dev_name(dev->parent) : "none");
@@ -181,7 +181,7 @@ static void initcall_debug_report(struct
{
ktime_t delta, rettime;
- if (pm_print_times) {
+ if (pm_print_times_enabled) {
rettime = ktime_get();
delta = ktime_sub(rettime, calltime);
pr_info("call %s+ returned %d after %Ld usecs\n", dev_name(dev),
Index: linux/drivers/base/power/power.h
===================================================================
--- linux.orig/drivers/base/power/power.h
+++ linux/drivers/base/power/power.h
@@ -85,14 +85,3 @@ static inline int pm_qos_sysfs_add(struc
static inline void pm_qos_sysfs_remove(struct device *dev) {}
#endif
-
-#ifdef CONFIG_PM_DEBUG
-
-extern int pm_print_times_enabled;
-#define pm_print_times (initcall_debug || pm_print_times_enabled)
-
-#else /* CONFIG_PM_DEBUG */
-
-#define pm_print_times initcall_debug
-
-#endif /* CONFIG_PM_DEBUG */
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists