[2/4] cpu: expose pm_qos_resume_latency for each cpu

Message ID 57C7A597.4050001@linaro.org
State New
Headers show

Commit Message

Alex Shi Sept. 1, 2016, 3:50 a.m.
Few commits and patch changed according to Greg's comments.

Regards
Alex

-- 
2.8.1.101.g72d917a


 
--
To unsubscribe from this list: send the line "unsubscribe linux-pm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Ulf Hansson Sept. 1, 2016, 9:26 a.m. | #1
On 1 September 2016 at 05:50, Alex Shi <alex.shi@linaro.org> wrote:
>

> Few commits and patch changed according to Greg's comments.

>

> Regards

> Alex

>

> ====

>

> From 186c534b0b8b9649fbfce05b0b4f90f764c571a4 Mon Sep 17 00:00:00 2001

> From: Alex Shi <alex.shi@linaro.org>

> Date: Tue, 16 Aug 2016 15:29:01 +0800

> Subject: [PATCH 2/4] cpu: expose pm_qos_resume_latency for each cpu

>

> Adding /sys/devices/system/cpu/cpux/power/pm_qos_resume_latency_us for

> each of cpus. The pm_qos_resume_latency usage defined in

> Documentation/ABI/testing/sysfs-devices-power

>

> The cpu-dma PM QoS constraint impacts all the cpus in the system. There

> is no way to let the user to choose a PM QoS constraint per cpu.

>

> The following patch exposes to the userspace a per cpu based sysfs file

> in order to let the userspace to change the value of the PM QoS latency

> constraint.

>

> This change is inoperative in its form and the cpuidle governors have to

> take into account the per cpu latency constraint in addition to the

> global cpu-dma latency constraint in order to operate properly.


In general I think the change makes sense, although it's this last
piece here that I wonder about.

Is it okay that we expose sysfs attributes to userspace that don't
have any effect if they change the values? Perhaps it should be the
responsibility of the menu governor somehow to expose the sysfs nodes
instead? Unless there are some difficulties that prevents us from that
of course.


Kind regards
Uffe

>

> Signed-off-by: Alex Shi <alex.shi@linaro.org>

> To: linux-kernel@vger.kernel.org

> To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

> Cc: linux-pm@vger.kernel.org

> Cc: Ulf Hansson <ulf.hansson@linaro.org>

> Cc: Daniel Lezcano <daniel.lezcano@linaro.org>

> ---

>  drivers/base/cpu.c | 2 ++

>  1 file changed, 2 insertions(+)

>

> diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c

> index 4c28e1a..2c3b359 100644

> --- a/drivers/base/cpu.c

> +++ b/drivers/base/cpu.c

> @@ -17,6 +17,7 @@

>  #include <linux/of.h>

>  #include <linux/cpufeature.h>

>  #include <linux/tick.h>

> +#include <linux/pm_qos.h>

>

>  #include "base.h"

>

> @@ -376,6 +377,7 @@ int register_cpu(struct cpu *cpu, int num)

>

>         per_cpu(cpu_sys_devices, num) = &cpu->dev;

>         register_cpu_under_node(num, cpu_to_node(num));

> +       dev_pm_qos_expose_latency_limit(&cpu->dev, 0);

>

>         return 0;

>  }

> --

> 2.8.1.101.g72d917a

>

>

>

--
To unsubscribe from this list: send the line "unsubscribe linux-pm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Alex Shi Sept. 13, 2016, 1:04 a.m. | #2
Cc Rafael.

On 09/01/2016 05:26 PM, Ulf Hansson wrote:
> In general I think the change makes sense, although it's this last

> piece here that I wonder about.

> 

> Is it okay that we expose sysfs attributes to userspace that don't

> have any effect if they change the values? Perhaps it should be the

> responsibility of the menu governor somehow to expose the sysfs nodes

> instead? Unless there are some difficulties that prevents us from that

> of course.

> 


Hi Ulf,

Sorry for response so late. The pm QoS designed to expose this interface
in userspace. Root user can change this value and made effect on device
sleeping status. That's required.

Since this is per device interface, set it on menu governor isn't so good.

Regards
Alex
--
To unsubscribe from this list: send the line "unsubscribe linux-pm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Ulf Hansson Sept. 13, 2016, 7:17 a.m. | #3
On 13 September 2016 at 03:04, Alex Shi <alex.shi@linaro.org> wrote:
> Cc Rafael.

>

> On 09/01/2016 05:26 PM, Ulf Hansson wrote:

>> In general I think the change makes sense, although it's this last

>> piece here that I wonder about.

>>

>> Is it okay that we expose sysfs attributes to userspace that don't

>> have any effect if they change the values? Perhaps it should be the

>> responsibility of the menu governor somehow to expose the sysfs nodes

>> instead? Unless there are some difficulties that prevents us from that

>> of course.

>>

>

> Hi Ulf,

>

> Sorry for response so late. The pm QoS designed to expose this interface

> in userspace. Root user can change this value and made effect on device

> sleeping status. That's required.


Sure, I understand that. Although, my point is that it's only when the
menu governor is enabled, that it would make sense for userspace to
modify the PM QoS values for the cpu device.

>

> Since this is per device interface, set it on menu governor isn't so good.


You may be right, and it's not a big deal for me! It was just an idea.

Kind regards
Uffe
--
To unsubscribe from this list: send the line "unsubscribe linux-pm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

====

From 186c534b0b8b9649fbfce05b0b4f90f764c571a4 Mon Sep 17 00:00:00 2001
From: Alex Shi <alex.shi@linaro.org>
Date: Tue, 16 Aug 2016 15:29:01 +0800
Subject: [PATCH 2/4] cpu: expose pm_qos_resume_latency for each cpu

Adding /sys/devices/system/cpu/cpux/power/pm_qos_resume_latency_us for
each of cpus. The pm_qos_resume_latency usage defined in
Documentation/ABI/testing/sysfs-devices-power

The cpu-dma PM QoS constraint impacts all the cpus in the system. There
is no way to let the user to choose a PM QoS constraint per cpu.

The following patch exposes to the userspace a per cpu based sysfs file
in order to let the userspace to change the value of the PM QoS latency
constraint.

This change is inoperative in its form and the cpuidle governors have to
take into account the per cpu latency constraint in addition to the
global cpu-dma latency constraint in order to operate properly.

Signed-off-by: Alex Shi <alex.shi@linaro.org>
To: linux-kernel@vger.kernel.org
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: linux-pm@vger.kernel.org
Cc: Ulf Hansson <ulf.hansson@linaro.org>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
---
 drivers/base/cpu.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c
index 4c28e1a..2c3b359 100644
--- a/drivers/base/cpu.c
+++ b/drivers/base/cpu.c
@@ -17,6 +17,7 @@ 
 #include <linux/of.h>
 #include <linux/cpufeature.h>
 #include <linux/tick.h>
+#include <linux/pm_qos.h>
 
 #include "base.h"
 
@@ -376,6 +377,7 @@  int register_cpu(struct cpu *cpu, int num)
 
 	per_cpu(cpu_sys_devices, num) = &cpu->dev;
 	register_cpu_under_node(num, cpu_to_node(num));
+	dev_pm_qos_expose_latency_limit(&cpu->dev, 0);
 
 	return 0;
 }