public inbox for linux-omap@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH 1/3] OMAP: omap_device: fix nsec/usec conversion in latency calculations
@ 2009-11-23 17:33 Kevin Hilman
  2009-11-23 17:33 ` [PATCH 2/3] OMAP: omap_device: track latency in nanoseconds Kevin Hilman
  2009-12-01 12:55 ` [PATCH 1/3] OMAP: omap_device: fix nsec/usec conversion in latency calculations Paul Walmsley
  0 siblings, 2 replies; 6+ messages in thread
From: Kevin Hilman @ 2009-11-23 17:33 UTC (permalink / raw)
  To: linux-omap

Use

   usecs = nsecs / NSEC_PER_USEC;

instead of

   usecs = nsecs * NSEC_PER_USEC;

Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
---
 arch/arm/plat-omap/omap_device.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm/plat-omap/omap_device.c b/arch/arm/plat-omap/omap_device.c
index 288fd12..722f2c8 100644
--- a/arch/arm/plat-omap/omap_device.c
+++ b/arch/arm/plat-omap/omap_device.c
@@ -142,7 +142,7 @@ static int _omap_device_activate(struct omap_device *od, u8 ignore_lat)
 		read_persistent_clock(&b);
 
 		c = timespec_sub(b, a);
-		act_lat = timespec_to_ns(&c) * NSEC_PER_USEC;
+		act_lat = timespec_to_ns(&c) / NSEC_PER_USEC;
 
 		pr_debug("omap_device: %s: pm_lat %d: activate: elapsed time "
 			 "%llu usec\n", od->pdev.name, od->pm_lat_level,
@@ -198,7 +198,7 @@ static int _omap_device_deactivate(struct omap_device *od, u8 ignore_lat)
 		read_persistent_clock(&b);
 
 		c = timespec_sub(b, a);
-		deact_lat = timespec_to_ns(&c) * NSEC_PER_USEC;
+		deact_lat = timespec_to_ns(&c) / NSEC_PER_USEC;
 
 		pr_debug("omap_device: %s: pm_lat %d: deactivate: elapsed time "
 			 "%llu usec\n", od->pdev.name, od->pm_lat_level,
-- 
1.6.5.1


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* [PATCH 2/3] OMAP: omap_device: track latency in nanoseconds
  2009-11-23 17:33 [PATCH 1/3] OMAP: omap_device: fix nsec/usec conversion in latency calculations Kevin Hilman
@ 2009-11-23 17:33 ` Kevin Hilman
  2009-11-23 17:33   ` [PATCH 3/3] OMAP: omap_device: optionally auto-adjust device activate/deactivate latencies Kevin Hilman
  2009-12-01 12:55   ` [PATCH 2/3] OMAP: omap_device: track latency in nanoseconds Paul Walmsley
  2009-12-01 12:55 ` [PATCH 1/3] OMAP: omap_device: fix nsec/usec conversion in latency calculations Paul Walmsley
  1 sibling, 2 replies; 6+ messages in thread
From: Kevin Hilman @ 2009-11-23 17:33 UTC (permalink / raw)
  To: linux-omap

Rather than having to do a usecs = nsecs / NSECS_PER_USEC to
track latency in usecs, just track it in nanoseconds.

Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
---
 arch/arm/plat-omap/include/plat/omap_device.h |    4 ++--
 arch/arm/plat-omap/omap_device.c              |    8 ++++----
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/arch/arm/plat-omap/include/plat/omap_device.h b/arch/arm/plat-omap/include/plat/omap_device.h
index c504780..3b2f3a6 100644
--- a/arch/arm/plat-omap/include/plat/omap_device.h
+++ b/arch/arm/plat-omap/include/plat/omap_device.h
@@ -50,8 +50,8 @@
  * @pm_lats: ptr to an omap_device_pm_latency table
  * @pm_lats_cnt: ARRAY_SIZE() of what is passed to @pm_lats
  * @pm_lat_level: array index of the last odpl entry executed - -1 if never
- * @dev_wakeup_lat: dev wakeup latency in microseconds
- * @_dev_wakeup_lat_limit: dev wakeup latency limit in usec - set by OMAP PM
+ * @dev_wakeup_lat: dev wakeup latency in nanoseconds
+ * @_dev_wakeup_lat_limit: dev wakeup latency limit in nsec - set by OMAP PM
  * @_state: one of OMAP_DEVICE_STATE_* (see above)
  * @flags: device flags
  *
diff --git a/arch/arm/plat-omap/omap_device.c b/arch/arm/plat-omap/omap_device.c
index 722f2c8..7bf665c 100644
--- a/arch/arm/plat-omap/omap_device.c
+++ b/arch/arm/plat-omap/omap_device.c
@@ -142,10 +142,10 @@ static int _omap_device_activate(struct omap_device *od, u8 ignore_lat)
 		read_persistent_clock(&b);
 
 		c = timespec_sub(b, a);
-		act_lat = timespec_to_ns(&c) / NSEC_PER_USEC;
+		act_lat = timespec_to_ns(&c);
 
 		pr_debug("omap_device: %s: pm_lat %d: activate: elapsed time "
-			 "%llu usec\n", od->pdev.name, od->pm_lat_level,
+			 "%llu nsec\n", od->pdev.name, od->pm_lat_level,
 			 act_lat);
 
 		WARN(act_lat > odpl->activate_lat, "omap_device: %s.%d: "
@@ -198,10 +198,10 @@ static int _omap_device_deactivate(struct omap_device *od, u8 ignore_lat)
 		read_persistent_clock(&b);
 
 		c = timespec_sub(b, a);
-		deact_lat = timespec_to_ns(&c) / NSEC_PER_USEC;
+		deact_lat = timespec_to_ns(&c);
 
 		pr_debug("omap_device: %s: pm_lat %d: deactivate: elapsed time "
-			 "%llu usec\n", od->pdev.name, od->pm_lat_level,
+			 "%llu nsec\n", od->pdev.name, od->pm_lat_level,
 			 deact_lat);
 
 		WARN(deact_lat > odpl->deactivate_lat, "omap_device: %s.%d: "
-- 
1.6.5.1


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* [PATCH 3/3] OMAP: omap_device: optionally auto-adjust device activate/deactivate latencies
  2009-11-23 17:33 ` [PATCH 2/3] OMAP: omap_device: track latency in nanoseconds Kevin Hilman
@ 2009-11-23 17:33   ` Kevin Hilman
  2009-11-23 18:35     ` Kevin Hilman
  2009-12-01 12:55   ` [PATCH 2/3] OMAP: omap_device: track latency in nanoseconds Paul Walmsley
  1 sibling, 1 reply; 6+ messages in thread
From: Kevin Hilman @ 2009-11-23 17:33 UTC (permalink / raw)
  To: linux-omap

First, this patch adds new worst-case latency values to the
omap_device_pm_latency struct.  Here the worst-case measured latencies
for the activate and deactivate hooks are stored.

In addition, add an option to auto-adjust the latency values used for
device activate/deactivate.

By setting a new 'OMAP_DEVICE_LATENCY_AUTO_ADJUST' flag in the
omap_device_pm_latency struct, the omap_device layer automatically
adjusts the activate/deactivate latencies to the worst-case measured
values.

Anytime a new worst-case value is found, it is printed to the console.
Here is an example log during boot using UART2 s an example.  After
boot, the OPP is manually changed to the 125MHz OPP:

[...]
Freeing init memory: 128K
omap_device: serial8250.2: new worst case deactivate latency 0: 30517
omap_device: serial8250.2: new worst case activate latency 0: 30517
omap_device: serial8250.2: new worst case activate latency 0: 218139648
omap_device: serial8250.2: new worst case deactivate latency 0: 61035
omap_device: serial8250.2: new worst case activate latency 0: 278076171
omap_device: serial8250.2: new worst case activate latency 0: 298614501
omap_device: serial8250.2: new worst case activate latency 0: 327331542

/ # echo 125000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed

omap_device: serial8250.2: new worst case deactivate latency 0: 91552

Motivation: this can be used as a technique to automatically determine
the worst case latency values.  The current method of printing a
warning on every violation is too noisy to actually interact the
console in order to set low OPP to discover latencies.

Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
---
 arch/arm/plat-omap/include/plat/omap_device.h |    4 ++
 arch/arm/plat-omap/omap_device.c              |   41 ++++++++++++++++++++-----
 2 files changed, 37 insertions(+), 8 deletions(-)

diff --git a/arch/arm/plat-omap/include/plat/omap_device.h b/arch/arm/plat-omap/include/plat/omap_device.h
index 3b2f3a6..e0634fe 100644
--- a/arch/arm/plat-omap/include/plat/omap_device.h
+++ b/arch/arm/plat-omap/include/plat/omap_device.h
@@ -132,11 +132,15 @@ int omap_device_enable_clocks(struct omap_device *od);
  */
 struct omap_device_pm_latency {
 	u32 deactivate_lat;
+	u32 deactivate_lat_worst;
 	int (*deactivate_func)(struct omap_device *od);
 	u32 activate_lat;
+	u32 activate_lat_worst;
 	int (*activate_func)(struct omap_device *od);
+	u32 flags;
 };
 
+#define OMAP_DEVICE_LATENCY_AUTO_ADJUST BIT(1)
 
 /* Get omap_device pointer from platform_device pointer */
 #define to_omap_device(x) container_of((x), struct omap_device, pdev)
diff --git a/arch/arm/plat-omap/omap_device.c b/arch/arm/plat-omap/omap_device.c
index 7bf665c..8ba12a6 100644
--- a/arch/arm/plat-omap/omap_device.c
+++ b/arch/arm/plat-omap/omap_device.c
@@ -148,10 +148,22 @@ static int _omap_device_activate(struct omap_device *od, u8 ignore_lat)
 			 "%llu nsec\n", od->pdev.name, od->pm_lat_level,
 			 act_lat);
 
-		WARN(act_lat > odpl->activate_lat, "omap_device: %s.%d: "
-		     "activate step %d took longer than expected (%llu > %d)\n",
-		     od->pdev.name, od->pdev.id, od->pm_lat_level,
-		     act_lat, odpl->activate_lat);
+		if (act_lat > odpl->activate_lat) {
+			odpl->activate_lat_worst = act_lat;
+			if (odpl->flags & OMAP_DEVICE_LATENCY_AUTO_ADJUST) {
+				odpl->activate_lat = act_lat;
+				pr_warning("omap_device: %s.%d: new worst case "
+					   "activate latency %d: %llu\n",
+					   od->pdev.name, od->pdev.id, 
+					   od->pm_lat_level, act_lat);
+			} else
+				pr_warning("omap_device: %s.%d: activate "
+					   "latency %d higher than exptected. "
+					   "(%llu > %d)\n",
+					   od->pdev.name, od->pdev.id, 
+					   od->pm_lat_level, act_lat,
+					   odpl->activate_lat);
+		}
 
 		od->dev_wakeup_lat -= odpl->activate_lat;
 	}
@@ -204,10 +216,23 @@ static int _omap_device_deactivate(struct omap_device *od, u8 ignore_lat)
 			 "%llu nsec\n", od->pdev.name, od->pm_lat_level,
 			 deact_lat);
 
-		WARN(deact_lat > odpl->deactivate_lat, "omap_device: %s.%d: "
-		     "deactivate step %d took longer than expected "
-		     "(%llu > %d)\n", od->pdev.name, od->pdev.id,
-		     od->pm_lat_level, deact_lat, odpl->deactivate_lat);
+		if (deact_lat > odpl->deactivate_lat) {
+			odpl->deactivate_lat_worst = deact_lat;
+			if (odpl->flags & OMAP_DEVICE_LATENCY_AUTO_ADJUST) {
+				odpl->deactivate_lat = deact_lat;
+				pr_warning("omap_device: %s.%d: new worst case "
+					   "deactivate latency %d: %llu\n",
+					   od->pdev.name, od->pdev.id,
+					   od->pm_lat_level, deact_lat);
+			} else
+				pr_warning("omap_device: %s.%d: deactivate "
+					   "latency %d higher than exptected. "
+					   "(%llu > %d)\n",
+					   od->pdev.name, od->pdev.id, 
+					   od->pm_lat_level, deact_lat,
+					   odpl->deactivate_lat);
+		}
+ 
 
 		od->dev_wakeup_lat += odpl->activate_lat;
 
-- 
1.6.5.1


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH 3/3] OMAP: omap_device: optionally auto-adjust device activate/deactivate latencies
  2009-11-23 17:33   ` [PATCH 3/3] OMAP: omap_device: optionally auto-adjust device activate/deactivate latencies Kevin Hilman
@ 2009-11-23 18:35     ` Kevin Hilman
  0 siblings, 0 replies; 6+ messages in thread
From: Kevin Hilman @ 2009-11-23 18:35 UTC (permalink / raw)
  To: linux-omap

Kevin Hilman wrote:
> First, this patch adds new worst-case latency values to the
> omap_device_pm_latency struct.  Here the worst-case measured latencies
> for the activate and deactivate hooks are stored.
> 
> In addition, add an option to auto-adjust the latency values used for
> device activate/deactivate.
> 
> By setting a new 'OMAP_DEVICE_LATENCY_AUTO_ADJUST' flag in the
> omap_device_pm_latency struct, the omap_device layer automatically
> adjusts the activate/deactivate latencies to the worst-case measured
> values.
> 
> Anytime a new worst-case value is found, it is printed to the console.
> Here is an example log during boot using UART2 s an example.  After
> boot, the OPP is manually changed to the 125MHz OPP:
> 
> [...]
> Freeing init memory: 128K
> omap_device: serial8250.2: new worst case deactivate latency 0: 30517
> omap_device: serial8250.2: new worst case activate latency 0: 30517
> omap_device: serial8250.2: new worst case activate latency 0: 218139648
> omap_device: serial8250.2: new worst case deactivate latency 0: 61035
> omap_device: serial8250.2: new worst case activate latency 0: 278076171
> omap_device: serial8250.2: new worst case activate latency 0: 298614501
> omap_device: serial8250.2: new worst case activate latency 0: 327331542
> 
> / # echo 125000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed
> 
> omap_device: serial8250.2: new worst case deactivate latency 0: 91552
> 
> Motivation: this can be used as a technique to automatically determine
> the worst case latency values.  The current method of printing a
> warning on every violation is too noisy to actually interact the
> console in order to set low OPP to discover latencies.

Another motivation for this patch is that the activate/deactivate 
latenices can vary depending on the idlemode of the device.  While 
working on the UART I noticed that when using no-idle, the activate 
latencies were as high as several hundred msecs as shown above.  When 
the UARTs are in smart-idle, the max latency is well under 100 usecs.

Kevin

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH 1/3] OMAP: omap_device: fix nsec/usec conversion in latency calculations
  2009-11-23 17:33 [PATCH 1/3] OMAP: omap_device: fix nsec/usec conversion in latency calculations Kevin Hilman
  2009-11-23 17:33 ` [PATCH 2/3] OMAP: omap_device: track latency in nanoseconds Kevin Hilman
@ 2009-12-01 12:55 ` Paul Walmsley
  1 sibling, 0 replies; 6+ messages in thread
From: Paul Walmsley @ 2009-12-01 12:55 UTC (permalink / raw)
  To: Kevin Hilman; +Cc: linux-omap

On Mon, 23 Nov 2009, Kevin Hilman wrote:

> Use
> 
>    usecs = nsecs / NSEC_PER_USEC;
> 
> instead of
> 
>    usecs = nsecs * NSEC_PER_USEC;
> 
> Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>

Thanks, looks good.  Will queue this up with the rest of the hwmod 
patches for .33.

- Paul

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH 2/3] OMAP: omap_device: track latency in nanoseconds
  2009-11-23 17:33 ` [PATCH 2/3] OMAP: omap_device: track latency in nanoseconds Kevin Hilman
  2009-11-23 17:33   ` [PATCH 3/3] OMAP: omap_device: optionally auto-adjust device activate/deactivate latencies Kevin Hilman
@ 2009-12-01 12:55   ` Paul Walmsley
  1 sibling, 0 replies; 6+ messages in thread
From: Paul Walmsley @ 2009-12-01 12:55 UTC (permalink / raw)
  To: Kevin Hilman; +Cc: linux-omap

On Mon, 23 Nov 2009, Kevin Hilman wrote:

> Rather than having to do a usecs = nsecs / NSECS_PER_USEC to
> track latency in usecs, just track it in nanoseconds.
> 
> Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>

Thanks, looks good.  Will queue this up with the rest of the hwmod
patches for .33.

- Paul

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2009-12-01 12:55 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-11-23 17:33 [PATCH 1/3] OMAP: omap_device: fix nsec/usec conversion in latency calculations Kevin Hilman
2009-11-23 17:33 ` [PATCH 2/3] OMAP: omap_device: track latency in nanoseconds Kevin Hilman
2009-11-23 17:33   ` [PATCH 3/3] OMAP: omap_device: optionally auto-adjust device activate/deactivate latencies Kevin Hilman
2009-11-23 18:35     ` Kevin Hilman
2009-12-01 12:55   ` [PATCH 2/3] OMAP: omap_device: track latency in nanoseconds Paul Walmsley
2009-12-01 12:55 ` [PATCH 1/3] OMAP: omap_device: fix nsec/usec conversion in latency calculations Paul Walmsley

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox