Skip to contentSkip to conversation

mfd: intel-lpss: Avoid resuming runtime-suspended lpss unnecessarily

RFC PATCH Superseded Chen Yu 3 participants 7 messages Sep 4, 2016 – Sep 27, 2016

Subsystems (2)

v1 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Chen Yu

author
#0 lore raw

RFC PATCH 1/1 mfd: intel-lpss: Avoid resuming runtime-suspended lpss unnecessarily

We have report that the intel_lpss_prepare() takes too much time during suspend, and this is because we first resume the devices from runtime suspend by resume_lpss_device(), to make sure they are in proper state before system suspend, which takes 100ms for each LPSS devices(PCI power state from D3_cold to D0). And since resume_lpss_device() resumes the devices synchronously, we might get huge latency if we have many LPSS devices.

So first try is to use pm_request_resume() instead, to make the runtime resume process asynchronously. Unfortunately the asynchronous runtime resume relies on pm_wq, which is freezed at early stage. So we choose another method, that is to avoid resuming runtime-suspended devices, if they are already runtime suspended. This is safe because for LPSS driver, the runtime suspend and system suspend are of the same hook - i.e., intel_lpss_suspend(). And moreover, this device is neither runtime wakeup source nor system wakeup source.

Cc Andy Shevchenko <[email protected]>
Cc Mika Westerberg <[email protected]>
Cc Rafael J. Wysocki <[email protected]>
---
 drivers/mfd/intel-lpss.c | 9 +++++++++
 1 file changed, 9 insertions(+)
drivers/mfd/intel-lpss.c +9 -0
index 41b1138..6dcc9a0 100644
@@ -485,6 +485,15 @@ static int resume_lpss_device(struct device *dev, void *data)
 int intel_lpss_prepare(struct device *dev)
 {
 	/*
+	 * This is safe because:
+	 * 1. The runtime suspend and system suspend
+	 * are of the same hook.
+	 * 2. This device is neither runtime wakeup source
+	 * nor system wakeup source.
+	 */
+	if (pm_runtime_status_suspended(dev))
+		return 1;
+	/*
 	 * Resume both child devices before entering system sleep. This
 	 * ensures that they are in proper state before they get suspended.
 	 */
Signature · 1 line
2.7.4
#1 lore raw
Quoted 31 lines · Chen Yu

We have report that the intel_lpss_prepare() takes too much time during suspend, and this is because we first resume the devices from runtime suspend by resume_lpss_device(), to make sure they are in proper state before system suspend, which takes 100ms for each LPSS devices(PCI power state from D3_cold to D0). And since resume_lpss_device() resumes the devices synchronously, we might get huge latency if we have many LPSS devices.

So first try is to use pm_request_resume() instead, to make the runtime resume process asynchronously. Unfortunately the asynchronous runtime resume relies on pm_wq, which is freezed at early stage. So we choose another method, that is to avoid resuming runtime-suspended devices, if they are already runtime suspended. This is safe because for LPSS driver, the runtime suspend and system suspend are of the same hook - i.e., intel_lpss_suspend(). And moreover, this device is neither runtime wakeup source nor system wakeup source.

Cc Andy Shevchenko <[email protected]>
Cc Mika Westerberg <[email protected]>
Cc Rafael J. Wysocki <[email protected]>
---
 drivers/mfd/intel-lpss.c | 9 +++++++++
 1 file changed, 9 insertions(+)
drivers/mfd/intel-lpss.c +0 -0
index 41b1138..6dcc9a0 100644
@@ -485,6 +485,15 @@ static int resume_lpss_device(struct device *dev, void *data)
 int intel_lpss_prepare(struct device *dev)

drivers/mfd/intel-lpss.c static int resume_lpss_device(struct device *dev, void *data) · line 495

Quoted 10 lines
drivers/mfd/intel-lpss.c +8 -0
 {
 	/*
+	 * This is safe because:
+	 * 1. The runtime suspend and system suspend
+	 * are of the same hook.
+	 * 2. This device is neither runtime wakeup source
+	 * nor system wakeup source.
+	 */
+	if (pm_runtime_status_suspended(dev))
+		return 1;

What's '1'?

Quoted 4 lines
drivers/mfd/intel-lpss.c +1 -0
+	/*
 	 * Resume both child devices before entering system sleep. This
 	 * ensures that they are in proper state before they get suspended.
 	 */
Signature · 4 lines
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog
author
#2 lore raw

Hi,

Quoted 37 lines · Lee Jones
Quoted 35 lines · Chen Yu

We have report that the intel_lpss_prepare() takes too much time during suspend, and this is because we first resume the devices from runtime suspend by resume_lpss_device(), to make sure they are in proper state before system suspend, which takes 100ms for each LPSS devices(PCI power state from D3_cold to D0). And since resume_lpss_device() resumes the devices synchronously, we might get huge latency if we have many LPSS devices.

So first try is to use pm_request_resume() instead, to make the runtime resume process asynchronously. Unfortunately the asynchronous runtime resume relies on pm_wq, which is freezed at early stage. So we choose another method, that is to avoid resuming runtime-suspended devices, if they are already runtime suspended. This is safe because for LPSS driver, the runtime suspend and system suspend are of the same hook - i.e., intel_lpss_suspend(). And moreover, this device is neither runtime wakeup source nor system wakeup source.

Cc Andy Shevchenko <[email protected]>
Cc Mika Westerberg <[email protected]>
Cc Rafael J. Wysocki <[email protected]>
---
 drivers/mfd/intel-lpss.c | 9 +++++++++
 1 file changed, 9 insertions(+)
drivers/mfd/intel-lpss.c +2 -0
index 41b1138..6dcc9a0 100644
@@ -485,6 +485,15 @@ static int resume_lpss_device(struct device *dev, void *data)
 int intel_lpss_prepare(struct device *dev)
 {
 	/*
+	 * This is safe because:
+	 * 1. The runtime suspend and system suspend
Quoted 9 lines

drivers/mfd/intel-lpss.c static int resume_lpss_device(struct device *dev, void *data) · line 495

Quoted 6 lines
drivers/mfd/intel-lpss.c +6 -0
+	 * are of the same hook.
+	 * 2. This device is neither runtime wakeup source
+	 * nor system wakeup source.
+	 */
+	if (pm_runtime_status_suspended(dev))
+		return 1;

What's '1'?

According to the comment in device_prepare():

A positive return value from ->prepare() means "this device appears to be runtime-suspended and its state is fine, so if it really is runtime-suspended, you can leave it in that state provided that you will do the same thing with all of its descendants".

Thanks,
Yu

#3 lore raw
Quoted 47 lines · Chen Yu
Quoted 37 lines · Lee Jones
Quoted 35 lines · Chen Yu

We have report that the intel_lpss_prepare() takes too much time during suspend, and this is because we first resume the devices from runtime suspend by resume_lpss_device(), to make sure they are in proper state before system suspend, which takes 100ms for each LPSS devices(PCI power state from D3_cold to D0). And since resume_lpss_device() resumes the devices synchronously, we might get huge latency if we have many LPSS devices.

So first try is to use pm_request_resume() instead, to make the runtime resume process asynchronously. Unfortunately the asynchronous runtime resume relies on pm_wq, which is freezed at early stage. So we choose another method, that is to avoid resuming runtime-suspended devices, if they are already runtime suspended. This is safe because for LPSS driver, the runtime suspend and system suspend are of the same hook - i.e., intel_lpss_suspend(). And moreover, this device is neither runtime wakeup source nor system wakeup source.

Cc Andy Shevchenko <[email protected]>
Cc Mika Westerberg <[email protected]>
Cc Rafael J. Wysocki <[email protected]>
---
 drivers/mfd/intel-lpss.c | 9 +++++++++
 1 file changed, 9 insertions(+)
drivers/mfd/intel-lpss.c +2 -0
index 41b1138..6dcc9a0 100644
@@ -485,6 +485,15 @@ static int resume_lpss_device(struct device *dev, void *data)
 int intel_lpss_prepare(struct device *dev)
 {
 	/*
+	 * This is safe because:
+	 * 1. The runtime suspend and system suspend
Quoted 9 lines

drivers/mfd/intel-lpss.c static int resume_lpss_device(struct device *dev, void *data) · line 495

Quoted 6 lines
drivers/mfd/intel-lpss.c +6 -0
+	 * are of the same hook.
+	 * 2. This device is neither runtime wakeup source
+	 * nor system wakeup source.
+	 */
+	if (pm_runtime_status_suspended(dev))
+		return 1;

What's '1'?

Quoted 6 lines

According to the comment in device_prepare():

A positive return value from ->prepare() means "this device appears to be runtime-suspended and its state is fine, so if it really is runtime-suspended, you can leave it in that state provided that you will do the same thing with all of its descendants".

Are there no defines for this?

Signature · 4 lines
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog
#4 lore raw
Quoted 47 lines · Lee Jones
Quoted 46 lines · Chen Yu
Quoted 45 lines · Lee Jones
Quoted 31 lines · Chen Yu

We have report that the intel_lpss_prepare() takes too much time during suspend, and this is because we first resume the devices from runtime suspend by resume_lpss_device(), to make sure they are in proper state before system suspend, which takes 100ms for each LPSS devices(PCI power state from D3_cold to D0). And since resume_lpss_device() resumes the devices synchronously, we might get huge latency if we have many LPSS devices.

So first try is to use pm_request_resume() instead, to make the runtime resume process asynchronously. Unfortunately the asynchronous runtime resume relies on pm_wq, which is freezed at early stage. So we choose another method, that is to avoid resuming runtime-suspended devices, if they are already runtime suspended. This is safe because for LPSS driver, the runtime suspend and system suspend are of the same hook - i.e., intel_lpss_suspend(). And moreover, this device is neither runtime wakeup source nor system wakeup source.

Cc Andy Shevchenko <[email protected]>
Cc Mika Westerberg <[email protected]>
Cc Rafael J. Wysocki <[email protected]>
---
  drivers/mfd/intel-lpss.c | 9 +++++++++
  1 file changed, 9 insertions(+)
drivers/mfd/intel-lpss.c +0 -0
index 41b1138..6dcc9a0 100644
@@ -485,6 +485,15 @@ static int resume_lpss_device(struct device *dev, void *data)
  int intel_lpss_prepare(struct device *dev)

drivers/mfd/intel-lpss.c static int resume_lpss_device(struct device *dev, void *data) · line 495

Quoted 10 lines
drivers/mfd/intel-lpss.c +8 -0
  {
  	/*
+	 * This is safe because:
+	 * 1. The runtime suspend and system suspend
+	 * are of the same hook.
+	 * 2. This device is neither runtime wakeup source
+	 * nor system wakeup source.
+	 */
+	if (pm_runtime_status_suspended(dev))
+		return 1;

What's '1'?

Quoted 8 lines
Quoted 6 lines

According to the comment in device_prepare():

A positive return value from ->prepare() means "this device appears to be runtime-suspended and its state is fine, so if it really is runtime-suspended, you can leave it in that state provided that you will do the same thing with all of its descendants".

Are there no defines for this?

Not at the moment, but I guess they can be added if really necessary. :-)

But that said it doesn't have to be 1 or any specific value. Any positive number will have the same effect.

Thanks,

Rafael

author
#5 lore raw

Hi,

Quoted 64 lines

-----Original Message-----
From: Wysocki, Rafael J
Sent: Tuesday, September 13, 2016 8:16 PM
To: Lee Jones; Chen, Yu C
Cc: [email protected]; Andy Shevchenko; Mika Westerberg
Subject: Re: [PATCH][RFC] mfd: intel-lpss: Avoid resuming runtime-suspended lpss unnecessarily

Quoted 12 lines · Lee Jones
Quoted 11 lines · Chen Yu
Quoted 10 lines · Lee Jones
Quoted 8 lines · Chen Yu

We have report that the intel_lpss_prepare() takes too much time during suspend, and this is because we first resume the devices from runtime suspend by resume_lpss_device(), to make sure they are in proper state before system suspend, which takes 100ms for each LPSS devices(PCI power state from D3_cold to D0). And since resume_lpss_device() resumes the devices synchronously, we might get huge latency if we have many LPSS devices.

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

So first try is to use pm_request_resume() instead, to make the runtime resume process asynchronously. Unfortunately the asynchronous runtime resume relies on pm_wq, which is freezed at early stage. So we choose another method, that is to avoid resuming runtime-suspended devices, if they are already runtime suspended. This is safe because for LPSS driver, the runtime suspend and system suspend are of the same hook - i.e., intel_lpss_suspend(). And moreover, this device is neither runtime wakeup source nor system

wakeup source.

Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Cc Andy Shevchenko <[email protected]>
Cc Mika Westerberg <[email protected]>
Cc Rafael J. Wysocki <[email protected]>
---
  drivers/mfd/intel-lpss.c | 9 +++++++++
  1 file changed, 9 insertions(+)
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

drivers/mfd/intel-lpss.c static int resume_lpss_device(struct device *dev,

Quoted 5 lines
drivers/mfd/intel-lpss.c +0 -0
index 41b1138..6dcc9a0 100644
@@ -485,6 +485,15 @@ static int resume_lpss_device(struct device *dev,

void *data)

Quoted 19 lines
Quoted 19 lines
Quoted 4 lines
Quoted 4 lines
drivers/mfd/intel-lpss.c +1 -0
  int intel_lpss_prepare(struct device *dev)
  {
  	/*
+	 * This is safe because:
Quoted 9 lines

drivers/mfd/intel-lpss.c static int resume_lpss_device(struct device *dev, void *data) · line 495

Quoted 7 lines
drivers/mfd/intel-lpss.c +7 -0
+	 * 1. The runtime suspend and system suspend
+	 * are of the same hook.
+	 * 2. This device is neither runtime wakeup source
+	 * nor system wakeup source.
+	 */
+	if (pm_runtime_status_suspended(dev))
+		return 1;

What's '1'?

According to the comment in device_prepare():

A positive return value from ->prepare() means "this device appears to be runtime-suspended and its state is fine, so if it really is runtime-suspended, you can leave it in that state provided that you will do the same thing with all of its descendants".

Quoted 7 lines
Quoted 2 lines

Are there no defines for this?

Not at the moment, but I guess they can be added if really necessary. :-)

But that said it doesn't have to be 1 or any specific value. Any positive number will have the same effect.

Thanks for point it out, Hi Lee, should I repost a patch set to define the return value and make this one based on that?

Thanks,
Yu

#6 lore raw
Quoted 67 lines · Yu C
Quoted 67 lines

-----Original Message-----
From: Wysocki, Rafael J
Sent: Tuesday, September 13, 2016 8:16 PM
To: Lee Jones; Chen, Yu C
Cc: [email protected]; Andy Shevchenko; Mika Westerberg
Subject: Re: [PATCH][RFC] mfd: intel-lpss: Avoid resuming runtime-suspended lpss unnecessarily

Quoted 12 lines · Lee Jones
Quoted 11 lines · Chen Yu
Quoted 10 lines · Lee Jones
Quoted 8 lines · Chen Yu

We have report that the intel_lpss_prepare() takes too much time during suspend, and this is because we first resume the devices from runtime suspend by resume_lpss_device(), to make sure they are in proper state before system suspend, which takes 100ms for each LPSS devices(PCI power state from D3_cold to D0). And since resume_lpss_device() resumes the devices synchronously, we might get huge latency if we have many LPSS devices.

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

So first try is to use pm_request_resume() instead, to make the runtime resume process asynchronously. Unfortunately the asynchronous runtime resume relies on pm_wq, which is freezed at early stage. So we choose another method, that is to avoid resuming runtime-suspended devices, if they are already runtime suspended. This is safe because for LPSS driver, the runtime suspend and system suspend are of the same hook - i.e., intel_lpss_suspend(). And moreover, this device is neither runtime wakeup source nor system

wakeup source.

Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Quoted 9 lines
Cc Andy Shevchenko <[email protected]>
Cc Mika Westerberg <[email protected]>
Cc Rafael J. Wysocki <[email protected]>
---
  drivers/mfd/intel-lpss.c | 9 +++++++++
  1 file changed, 9 insertions(+)
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

drivers/mfd/intel-lpss.c static int resume_lpss_device(struct device *dev,

Quoted 5 lines
drivers/mfd/intel-lpss.c +0 -0
index 41b1138..6dcc9a0 100644
@@ -485,6 +485,15 @@ static int resume_lpss_device(struct device *dev,

void *data)

Quoted 21 lines
Quoted 13 lines
Quoted 4 lines
Quoted 4 lines
drivers/mfd/intel-lpss.c +1 -0
  int intel_lpss_prepare(struct device *dev)
  {
  	/*
+	 * This is safe because:
Quoted 9 lines

drivers/mfd/intel-lpss.c static int resume_lpss_device(struct device *dev, void *data) · line 495

Quoted 7 lines
drivers/mfd/intel-lpss.c +7 -0
+	 * 1. The runtime suspend and system suspend
+	 * are of the same hook.
+	 * 2. This device is neither runtime wakeup source
+	 * nor system wakeup source.
+	 */
+	if (pm_runtime_status_suspended(dev))
+		return 1;

What's '1'?

Quoted 6 lines

According to the comment in device_prepare():

A positive return value from ->prepare() means "this device appears to be runtime-suspended and its state is fine, so if it really is runtime-suspended, you can leave it in that state provided that you will do the same thing with all of its descendants".

Are there no defines for this?

Quoted 6 lines
Quoted 4 lines

Not at the moment, but I guess they can be added if really necessary. :-)

But that said it doesn't have to be 1 or any specific value. Any positive number will have the same effect.

Thanks for point it out, Hi Lee, should I repost a patch set to define the return value and make this one based on that?

I think that would be a great way to move forward.

Signature · 4 lines
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog