PM / wakeup : remove misuse of IRQF_NO_SUSPEND flag

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags and uses newly introduce PM wakeup APIs dev_pm_{set,clear}_wake_irq.

Cc Linus Walleij <[email protected]>

Cc Alessandro Zummo <[email protected]>

Cc Alexandre Belloni <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

Quoted 9 lines · Sudeep Holla

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags and uses newly introduce PM wakeup APIs dev_pm_{set,clear}_wake_irq.

Quoted 5 lines

Cc Linus Walleij <[email protected]>

Cc Alessandro Zummo <[email protected]>

Cc Alexandre Belloni <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

Looks correct to me.

Acked-by Linus Walleij <[email protected]>

Quoted 17 lines · Linus Walleij

Quoted 14 lines · Sudeep Holla

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags and uses newly introduce PM wakeup APIs dev_pm_{set,clear}_wake_irq.

Cc Linus Walleij <[email protected]>

Cc Alessandro Zummo <[email protected]>

Cc Alexandre Belloni <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

Quoted 3 lines

Looks correct to me.

Acked-by Linus Walleij <[email protected]>

Thanks, while you are at it, do you know what interrupt stv2_pl031 shares so that I can fix the TODO in the comment or the code(whichever applicable).

Quoted 2 lines

Looks correct to me.

Acked-by Linus Walleij <[email protected]>

Thanks, while you are at it, do you know what interrupt stv2_pl031 shares so that I can fix the TODO in the comment or the code(whichever applicable).

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags and uses newly introduce PM wakeup APIs dev_pm_{set,clear}_wake_irq.

Cc Linus Walleij <[email protected]>

Cc Alessandro Zummo <[email protected]>

Cc Alexandre Belloni <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
 drivers/rtc/rtc-pl031.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags and uses newly introduce PM wakeup APIs dev_pm_{set,clear}_wake_irq.

Cc Linus Walleij <[email protected]>

Cc Alessandro Zummo <[email protected]>

Cc Alexandre Belloni <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags and uses newly introduce PM wakeup APIs dev_pm_{set,clear}_wake_irq.

Cc Linus Walleij <[email protected]>

Cc Alessandro Zummo <[email protected]>

Cc Alexandre Belloni <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
 drivers/rtc/rtc-ab8500.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

The driver handles wakeup irq correctly using device_init_wakeup and enable_irq_wake. There's no need to use IRQF_NO_SUSPEND while registering the interrupt.

This patch removes the use of IRQF_NO_SUSPEND flag.

Cc Laxman Dewangan <[email protected]>

Cc Stephen Warren <[email protected]>

Cc Thierry Reding <[email protected]>

Cc [email protected]

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
 drivers/input/keyboard/tegra-kbc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/input/keyboard/tegra-kbc.c static int tegra_kbc_probe(struct platform_device · line 709

Quoted 13 lines · Sudeep Holla

drivers/input/keyboard/tegra-kbc.c +1 -1

index f97c73bd14f8..5f5dbbd51991 100644
@@ -705,7 +705,7 @@ static int tegra_kbc_probe(struct platform_device
 *pdev)
      input_set_drvdata(kbc->idev, kbc);
      err = devm_request_irq(&pdev->dev, kbc->irq, tegra_kbc_isr,
-              IRQF_NO_SUSPEND | IRQF_TRIGGER_HIGH, pdev->name, kbc);
+                   IRQF_TRIGGER_HIGH, pdev->name, kbc);
 

Acked-by Laxman Dewangan <[email protected]>

Hi Dmitry,

Quoted 18 lines · Laxman Dewangan

drivers/input/keyboard/tegra-kbc.c static int tegra_kbc_probe(struct platform_device · line 709

Quoted 13 lines · Sudeep Holla

drivers/input/keyboard/tegra-kbc.c +1 -1

index f97c73bd14f8..5f5dbbd51991 100644
@@ -705,7 +705,7 @@ static int tegra_kbc_probe(struct platform_device
 *pdev)
     input_set_drvdata(kbc->idev, kbc);
     err = devm_request_irq(&pdev->dev, kbc->irq, tegra_kbc_isr,
-              IRQF_NO_SUSPEND | IRQF_TRIGGER_HIGH, pdev->name, kbc);
+                   IRQF_TRIGGER_HIGH, pdev->name, kbc);
 

Acked-by Laxman Dewangan <[email protected]>

Can you pick this up with all the Acks ?

mpic_irq_set_wake return -ENXIO for non FSL MPIC and sets IRQF_NO_SUSPEND flag for FSL ones. enable_irq_wake already returns -ENXIO if irq_set_wak is not implemented. Also there's no need to set the IRQF_NO_SUSPEND flag as it doesn't guarantee wakeup for that interrupt.

This patch removes the redundant mpic_irq_set_wake and sets the IRQCHIP_SKIP_SET_WAKE for only FSL MPIC.

Cc Benjamin Herrenschmidt <[email protected]>

Cc Paul Mackerras <[email protected]>

Cc Michael Ellerman <[email protected]>

Cc Scott Wood <[email protected]>

Cc Hongtao Jia <[email protected]>

Cc Marc Zyngier <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
 arch/powerpc/sysdev/mpic.c | 23 ++++-------------------
 1 file changed, 4 insertions(+), 19 deletions(-)

arch/powerpc/sysdev/mpic.c +0 -16

index 537e5db85a06..123e43612f0a 100644
@@ -924,22 +924,6 @@ int mpic_set_irq_type(struct irq_data *d, unsigned int 
 flow_type)
      return IRQ_SET_MASK_OK_NOCOPY;
 }
 
-static int mpic_irq_set_wake(struct irq_data *d, unsigned int on)
-{
-     struct irq_desc *desc = container_of(d, struct irq_desc, irq_data);
-     struct mpic *mpic = mpic_from_irq_data(d);
-
-     if (!(mpic->flags & MPIC_FSL))
-             return -ENXIO;
-
-     if (on)
-             desc->action->flags |= IRQF_NO_SUSPEND;
-     else
-             desc->action->flags &= ~IRQF_NO_SUSPEND;
-
-     return 0;
-}
-
 void mpic_set_vector(unsigned int virq, unsigned int vector)
 {

      struct mpic *mpic = mpic_from_irq(virq);

arch/powerpc/sysdev/mpic.c +0 -2

@@ -977,7 +961,6 @@ static struct irq_chip mpic_irq_chip = {
      .irq_unmask     = mpic_unmask_irq,
      .irq_eoi        = mpic_end_irq,
      .irq_set_type   = mpic_set_irq_type,
-     .irq_set_wake   = mpic_irq_set_wake,
 };
 
 #ifdef CONFIG_SMP

@@ -992,7 +975,6 @@ static struct irq_chip mpic_tm_chip = {
      .irq_mask       = mpic_mask_tm,
      .irq_unmask     = mpic_unmask_tm,
      .irq_eoi        = mpic_end_irq,
-     .irq_set_wake   = mpic_irq_set_wake,
 };
 
 #ifdef CONFIG_MPIC_U3_HT_IRQS

-----Original Message-----
From: Wood Scott-B07421
Sent: Wednesday, September 23, 2015 7:50 AM
To: Sudeep Holla
Cc: [email protected]; [email protected]; Thomas Gleixner; Rafael J. Wysocki; Benjamin Herrenschmidt; Paul Mackerras; Michael Ellerman; Jia Hongtao-B38951; Marc Zyngier; [email protected]; Wang Dongsheng- B40534
Subject: Re: [PATCH 04/17] powerpc: mpic: use IRQCHIP_SKIP_SET_WAKE instead of redundant mpic_irq_set_wake

Quoted 5 lines · Sudeep Holla

mpic_irq_set_wake return -ENXIO for non FSL MPIC and sets IRQF_NO_SUSPEND flag for FSL ones. enable_irq_wake already returns -ENXIO if irq_set_wak is not implemented. Also there's no need to set the IRQF_NO_SUSPEND flag as it doesn't guarantee wakeup for that interrupt.

Quoted 6 lines · Sudeep Holla

Quoted 5 lines

mpic_irq_set_wake return -ENXIO for non FSL MPIC and sets IRQF_NO_SUSPEND flag for FSL ones. enable_irq_wake already returns -ENXIO if irq_set_wak is not implemented. Also there's no need to set the IRQF_NO_SUSPEND flag as it doesn't guarantee wakeup for that interrupt.

Non-freescale return -ENXIO, is there any issue? If non-freescale platform does not support it, but IPs still use enable/disable_irq_wake, we should return a error number.

@Scott:
If set this flag we cannot keep a irq as a wakeup source when system going to SUSPEND or MEM.

irq_set_wake() means we can set this irq as a wake source. IRQCHIP_SKIP_SET_WAKE is ignore irq_set_wake() feature.

-----Original Message-----
From: [email protected] [mailto:linux-kernel- [email protected]] On Behalf Of Thomas Gleixner
Sent: Wednesday, September 23, 2015 11:49 AM
To: Wang Dongsheng-B40534
Cc: Wood Scott-B07421; Sudeep Holla; [email protected]; linux- [email protected]; Rafael J. Wysocki; Benjamin Herrenschmidt; Paul Mackerras; Michael Ellerman; Jia Hongtao-B38951; Marc Zyngier; linuxppc- [email protected]
Subject: RE: [PATCH 04/17] powerpc: mpic: use IRQCHIP_SKIP_SET_WAKE instead of redundant mpic_irq_set_wake

Quoted 10 lines · Wang Dongsheng

Quoted 6 lines · Sudeep Holla

Quoted 5 lines

mpic_irq_set_wake return -ENXIO for non FSL MPIC and sets IRQF_NO_SUSPEND flag for FSL ones. enable_irq_wake already returns -ENXIO if irq_set_wak is not implemented. Also there's no need to set the IRQF_NO_SUSPEND flag as it doesn't guarantee wakeup for that interrupt.

Non-freescale return -ENXIO, is there any issue? If non-freescale platform does not support it, but IPs still use enable/disable_irq_wake, we should return a error number.

You can just set IRQCHIP_SKIP_SET_WAKE for FSL chips and not for the others.

Quoted 4 lines

@Scott:
If set this flag we cannot keep a irq as a wakeup source when system going to SUSPEND or MEM.

Quoted 2 lines

irq_set_wake() means we can set this irq as a wake source. IRQCHIP_SKIP_SET_WAKE is ignore irq_set_wake() feature.

Nonsense. IRQCHIP_SKIP_SET_WAKE merily tells the core not to bail on !chip->irq_set_wake(), but its still marking the interrupt as wakeup source and therefor not masking it on suspend.

Quoted 11 lines

-----Original Message-----
From: [email protected] [mailto:linux-kernel- [email protected]] On Behalf Of Thomas Gleixner
Sent: Wednesday, September 23, 2015 11:49 AM
To: Wang Dongsheng-B40534
Cc: Wood Scott-B07421; Sudeep Holla; [email protected]; linux- [email protected]; Rafael J. Wysocki; Benjamin Herrenschmidt; Paul Mackerras; Michael Ellerman; Jia Hongtao-B38951; Marc Zyngier; linuxppc- [email protected]
Subject: RE: [PATCH 04/17] powerpc: mpic: use IRQCHIP_SKIP_SET_WAKE instead of redundant mpic_irq_set_wake

Quoted 19 lines · Wang Dongsheng

Quoted 10 lines

Quoted 6 lines · Sudeep Holla

Quoted 5 lines

mpic_irq_set_wake return -ENXIO for non FSL MPIC and sets IRQF_NO_SUSPEND flag for FSL ones. enable_irq_wake already returns -ENXIO if irq_set_wak is not implemented. Also there's no need to set the IRQF_NO_SUSPEND flag as it doesn't guarantee wakeup for that interrupt.

Non-freescale return -ENXIO, is there any issue? If non-freescale platform does not support it, but IPs still use enable/disable_irq_wake, we should return a error number.

You can just set IRQCHIP_SKIP_SET_WAKE for FSL chips and not for the others.

Quoted 4 lines

@Scott:
If set this flag we cannot keep a irq as a wakeup source when system going to SUSPEND or MEM.

Quoted 7 lines

Quoted 2 lines

irq_set_wake() means we can set this irq as a wake source. IRQCHIP_SKIP_SET_WAKE is ignore irq_set_wake() feature.

Nonsense. IRQCHIP_SKIP_SET_WAKE merily tells the core not to bail on !chip->irq_set_wake(), but its still marking the interrupt as wakeup source and therefor not masking it on suspend.

Sorry, I just check irq_set_irq_wake() code, right, IRQCHIP_SKIP_SET_WAKE also can going to irqd_set to mask IRQD_WAKEUP_STATE.

Yes, this flag just skip the irq_set_wake() not this feature.

mpic_irq_set_wake return -ENXIO for non FSL MPIC and sets IRQF_NO_SUSPEND flag for FSL ones. enable_irq_wake already returns -ENXIO if irq_set_wak is not implemented. Also there's no need to set the IRQF_NO_SUSPEND flag as it doesn't guarantee wakeup for that interrupt.

This patch removes the redundant mpic_irq_set_wake and sets the IRQCHIP_SKIP_SET_WAKE for only FSL MPIC.

Cc Benjamin Herrenschmidt <[email protected]>

Cc Paul Mackerras <[email protected]>

Cc Michael Ellerman <[email protected]>

Cc Scott Wood <[email protected]>

Cc Hongtao Jia <[email protected]>

Cc Marc Zyngier <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
 arch/powerpc/sysdev/mpic.c | 23 ++++-------------------
 1 file changed, 4 insertions(+), 19 deletions(-)

Quoted 16 lines · Sudeep Holla

mpic_irq_set_wake return -ENXIO for non FSL MPIC and sets IRQF_NO_SUSPEND flag for FSL ones. enable_irq_wake already returns -ENXIO if irq_set_wak is not implemented. Also there's no need to set the IRQF_NO_SUSPEND flag as it doesn't guarantee wakeup for that interrupt.

This patch removes the redundant mpic_irq_set_wake and sets the IRQCHIP_SKIP_SET_WAKE for only FSL MPIC.

Cc Benjamin Herrenschmidt <[email protected]>

Cc Paul Mackerras <[email protected]>

Cc Michael Ellerman <[email protected]>

Cc Scott Wood <[email protected]>

Cc Hongtao Jia <[email protected]>

Cc Marc Zyngier <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

Quoted 3 lines

---
  arch/powerpc/sysdev/mpic.c | 23 ++++-------------------
  1 file changed, 4 insertions(+), 19 deletions(-)

Acked-by Scott Wood <[email protected]>

Hi Ben,

Quoted 22 lines · Scott Wood

Quoted 20 lines · Sudeep Holla

mpic_irq_set_wake return -ENXIO for non FSL MPIC and sets IRQF_NO_SUSPEND
flag for FSL ones. enable_irq_wake already returns -ENXIO if irq_set_wak is not implemented. Also there's no need to set the IRQF_NO_SUSPEND flag as it doesn't guarantee wakeup for that interrupt.

This patch removes the redundant mpic_irq_set_wake and sets the IRQCHIP_SKIP_SET_WAKE for only FSL MPIC.

Cc Benjamin Herrenschmidt <[email protected]>

Cc Paul Mackerras <[email protected]>

Cc Michael Ellerman <[email protected]>

Cc Scott Wood <[email protected]>

Cc Hongtao Jia <[email protected]>

Cc Marc Zyngier <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
  arch/powerpc/sysdev/mpic.c | 23 ++++-------------------
  1 file changed, 4 insertions(+), 19 deletions(-)

Quoted 2 lines

Acked-by Scott Wood <[email protected]>

Can you pick this up via your tree ?

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags and uses newly introduce PM wakeup APIs dev_pm_{set,clear}_wake_irq.

Cc Darren Hart <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags and uses newly introduce PM wakeup APIs dev_pm_{set,clear}_wake_irq.

Cc Alessandro Zummo <[email protected]>

Cc Alexandre Belloni <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
 drivers/rtc/rtc-ds1343.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

The driver handles wakeup irq correctly using irq_set_irq_wake. There's no need to use IRQF_NO_SUSPEND while registering the interrupt.

This patch removes the use of IRQF_NO_SUSPEND flag.

Cc Samuel Ortiz <[email protected]>

Cc Lee Jones <[email protected]>

Signed-off-by Sudeep Holla <[email protected]>

---
 drivers/mfd/qcom_rpm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/mfd/qcom_rpm.c +1 -1

index 6afc9fabd94c..207a3bd68559 100644
@@ -550,7 +550,7 @@ static int qcom_rpm_probe(struct platform_device *pdev)
 	ret = devm_request_irq(&pdev->dev,
 			       irq_ack,
 			       qcom_rpm_ack_interrupt,
-			       IRQF_TRIGGER_RISING | IRQF_NO_SUSPEND,
+			       IRQF_TRIGGER_RISING,
 			       "qcom_rpm_ack",
 			       rpm);
 	if (ret) {

The $SUBJECT is not correct.

Quoted 11 lines

The driver handles wakeup irq correctly using irq_set_irq_wake. There's no need to use IRQF_NO_SUSPEND while registering the interrupt.

This patch removes the use of IRQF_NO_SUSPEND flag.

Cc Samuel Ortiz <[email protected]>

Cc Lee Jones <[email protected]>

Signed-off-by Sudeep Holla <[email protected]>

---
 drivers/mfd/qcom_rpm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

... code is fine though:

Acked-by Lee Jones <[email protected]>

Please re-submit with the subject line fixed.

Quoted 14 lines · Lee Jones

The $SUBJECT is not correct.

Quoted 11 lines

The driver handles wakeup irq correctly using irq_set_irq_wake. There's no need to use IRQF_NO_SUSPEND while registering the interrupt.

This patch removes the use of IRQF_NO_SUSPEND flag.

Cc Samuel Ortiz <[email protected]>

Cc Lee Jones <[email protected]>

Signed-off-by Sudeep Holla <[email protected]>

---
 drivers/mfd/qcom_rpm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Quoted 5 lines

... code is fine though:

Acked-by Lee Jones <[email protected]>

Please re-submit with the subject line fixed.

What's wrong with the subject?

Quoted 17 lines · Dmitry Torokhov

Quoted 16 lines · Lee Jones

The $SUBJECT is not correct.

Quoted 11 lines

The driver handles wakeup irq correctly using irq_set_irq_wake. There's no need to use IRQF_NO_SUSPEND while registering the interrupt.

This patch removes the use of IRQF_NO_SUSPEND flag.

Cc Samuel Ortiz <[email protected]>

Cc Lee Jones <[email protected]>

Signed-off-by Sudeep Holla <[email protected]>

---
  drivers/mfd/qcom_rpm.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

... code is fine though:

Quoted 5 lines

Quoted 3 lines

Acked-by Lee Jones <[email protected]>

Please re-submit with the subject line fixed.

What's wrong with the subject?

This patch has nothing to do with Input and the first word of the explanatory sentence should start with an uppercase character.

The $SUBJECT is not correct.

The driver handles wakeup irq correctly using irq_set_irq_wake. There's no need to use IRQF_NO_SUSPEND while registering the interrupt.

This patch removes the use of IRQF_NO_SUSPEND flag.

Cc Samuel Ortiz <[email protected]>

Acked-by Lee Jones <[email protected]>

Signed-off-by Sudeep Holla <[email protected]>

---
 drivers/mfd/qcom_rpm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Updated version with the $subject fixed and Lee's ACK.

drivers/mfd/qcom_rpm.c +1 -1

index 6afc9fabd94c..207a3bd68559 100644
@@ -550,7 +550,7 @@ static int qcom_rpm_probe(struct platform_device *pdev)
 	ret = devm_request_irq(&pdev->dev,
 			       irq_ack,
 			       qcom_rpm_ack_interrupt,
-			       IRQF_TRIGGER_RISING | IRQF_NO_SUSPEND,
+			       IRQF_TRIGGER_RISING,
 			       "qcom_rpm_ack",
 			       rpm);
 	if (ret) {

The description in the driver states: "ABX500 does not provide auto ADC, so to monitor the required temperatures, a periodic work is used. It is more important to not wake up the CPU... If the chip gets too hot during a sleep state it's most likely due to external factors, such as the surrounding temperature and nothing can be done in S/W."

So it makes no sense to keep IRQs enabled as it need not be wakeup source. This patch removes the use of IRQF_NO_SUSPEND flag

Cc Jean Delvare <[email protected]>

Cc Guenter Roeck <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
  drivers/hwmon/abx500.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

drivers/hwmon/abx500.c +1 -1

index 6cb89c0ebab6..dcab258f60c1 100644
@@ -377,7 +377,7 @@ static int setup_irqs(struct platform_device *pdev)
  	}
 
  	ret = devm_request_threaded_irq(&pdev->dev, irq, NULL,
-		abx500_temp_irq_handler, IRQF_NO_SUSPEND, "abx500-temp", pdev);
+		abx500_temp_irq_handler, 0, "abx500-temp", pdev);
  	if (ret < 0)
  		dev_err(&pdev->dev, "Request threaded irq failed (%d)\n", ret);
 

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Cc Steven Miao <[email protected]>

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Quoted 8 lines · Sudeep Holla

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Applying into omap-for-v4.4/cleanup thanks.

Quoted 7 lines · Tony Lindgren

Quoted 6 lines · Sudeep Holla

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

Quoted 4 lines

Quoted 2 lines

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Applying into omap-for-v4.4/cleanup thanks.

Actually I don't think this does the right thing. The interrupts in the $subject patch are in the always on powerdomain, and we really

want them to be excluded from the suspend.

So not applying without further explanations.

Quoted 8 lines · Tony Lindgren

Quoted 7 lines · Tony Lindgren

Quoted 6 lines · Sudeep Holla

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

Quoted 7 lines

Quoted 4 lines

Quoted 2 lines

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Applying into omap-for-v4.4/cleanup thanks.

Actually I don't think this does the right thing. The interrupts in the $subject patch are in the always on powerdomain, and we really

Agreed

Quoted 2 lines

want them to be excluded from the suspend.

OK but what's wrong with this patch. At-least the name suggest it's a wakeup interrupt. And using IRQF_NO_SUSPEND for the wakeup interrupt is simply wrong.

Quoted 2 lines

So not applying without further explanations.

But I don't understand the real need for IRQF_NO_SUSPEND over wakeup APIs ?

Quoted 19 lines · Sudeep Holla

Quoted 8 lines · Tony Lindgren

Quoted 7 lines · Tony Lindgren

Quoted 6 lines · Sudeep Holla

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

Quoted 7 lines

Quoted 4 lines

Quoted 2 lines

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Applying into omap-for-v4.4/cleanup thanks.

Actually I don't think this does the right thing. The interrupts in the $subject patch are in the always on powerdomain, and we really

Quoted 8 lines

Agreed

Quoted 2 lines

want them to be excluded from the suspend.

OK but what's wrong with this patch. At-least the name suggest it's a wakeup interrupt. And using IRQF_NO_SUSPEND for the wakeup interrupt is simply wrong.

Hmm so if we have a separate always on irq controller for the wake-up events and we want to keep it always on and exclude it from any suspend related things.. Why would we not use IRQF_NO_SUSPEND on it?

Above you say "The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle..." and that's exactly what we want to do here :)

For the dedicated wake-up interrupts, we have separate registers to enable and disable them. The $subject irq is the shared interrupt that allows making use of the pin specific wake-up interrupts, and for those yes we are using enable_irq_wake().

Quoted 4 lines

Quoted 2 lines

So not applying without further explanations.

But I don't understand the real need for IRQF_NO_SUSPEND over wakeup APIs ?

Because in the $subject case we just want to always keep it on and never suspend it. It's unrelated to the wakeup APIs at least for the omap related SoCs.

-----Original Message-----
From: Sudeep Holla [mailto:[email protected]]
Sent: Monday, September 21, 2015 6:47 PM
To: [email protected]; [email protected]
Cc: Sudeep Holla <[email protected]>; Thomas Gleixner <[email protected]>; Rafael J. Wysocki <[email protected]>; David S. Miller <[email protected]>; Manoil Claudiu-B08782 <[email protected]>; Kevin Hao <[email protected]>; [email protected]
Subject: [PATCH 13/17] net: gianfar: remove misuse of IRQF_NO_SUSPEND flag

The device is set as wakeup capable using proper wakeup API but the driver misuses IRQF_NO_SUSPEND to set the interrupt as wakeup source which is incorrect.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Quoted 16 lines

-----Original Message-----
From: Sudeep Holla [mailto:[email protected]]
Sent: Monday, September 21, 2015 6:47 PM
To: [email protected]; [email protected]
Cc: Sudeep Holla <[email protected]>; Thomas Gleixner <[email protected]>; Rafael J. Wysocki <[email protected]>; David S. Miller <[email protected]>; Manoil Claudiu-B08782 <[email protected]>; Kevin Hao <[email protected]>; [email protected]
Subject: [PATCH 13/17] net: gianfar: remove misuse of IRQF_NO_SUSPEND flag

The device is set as wakeup capable using proper wakeup API but the driver misuses IRQF_NO_SUSPEND to set the interrupt as wakeup source which is incorrect.

Quoted 3 lines

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

What would be the purpose of IRQF_NO_SUSPEND flag then?

The flag is a friendlier API compared to calling enable_irq_wake().

For older kernels, on PPC architectures, the flag did the job.

When did this change? Since when using IRQF_NO_SUSPEND is a "misuse"?

Quoted 7 lines

The device is set as wakeup capable using proper wakeup API but the driver misuses IRQF_NO_SUSPEND to set the interrupt as wakeup source which is incorrect.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

What would be the purpose of IRQF_NO_SUSPEND flag then? The flag is a friendlier API compared to calling enable_irq_wake(). For older kernels,

on PPC architectures, the flag did the job. When did this change? Since when using IRQF_NO_SUSPEND is a "misuse"?

-----Original Message-----
From: Thomas Gleixner [mailto:[email protected]]
Sent: Monday, September 21, 2015 7:52 PM
To: Manoil Claudiu-B08782 <[email protected]>
Cc: Sudeep Holla <[email protected]>; [email protected]; linux- [email protected]; Rafael J. Wysocki <[email protected]>; David S. Miller <[email protected]>; Kevin Hao <[email protected]>; [email protected]
Subject: RE: [PATCH 13/17] net: gianfar: remove misuse of IRQF_NO_SUSPEND flag

Quoted 10 lines · Manoil Claudiu

Quoted 7 lines

The device is set as wakeup capable using proper wakeup API but the driver misuses IRQF_NO_SUSPEND to set the interrupt as wakeup source which is incorrect.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

What would be the purpose of IRQF_NO_SUSPEND flag then? The flag is a friendlier API compared to calling enable_irq_wake(). For older kernels,

It's not an API, it's just a bandaid for lazy programmers.

Quoted 2 lines

on PPC architectures, the flag did the job. When did this change? Since when using IRQF_NO_SUSPEND is a "misuse"?

It always was. Simply because IRQF_NO_SUSPEND has absolutely nothing to do with wakeup interrupt sources. It's a flag which excludes the interrupt from the suspend mechanism, but it does not flag it a wakeup source.

Quoted 2 lines

-----Original Message-----
From: Thomas Gleixner [mailto:[email protected]]

Quoted 8 lines

Quoted 2 lines

on PPC architectures, the flag did the job. When did this change? Since when using IRQF_NO_SUSPEND is a "misuse"?

It always was. Simply because IRQF_NO_SUSPEND has absolutely nothing to do with wakeup interrupt sources. It's a flag which excludes the interrupt from the suspend mechanism, but it does not flag it a wakeup source.

I'm seeing also a "powerpc: mpic" patch in the series, unfortunately I can't

afford to test it right now. However I ran a quick test with this gianfar patch in isolation on a powerpc system, and seen some difference in the behavior (with and w/o the patch). In both cases the system wakes up from standby by magic packet. However, without the IRQF_NO_SUSPEND flag 2 wake-up interrupts are reported in /proc/interrupts for one magic packet; with the

flag on there's just 1 interrupt. Maybe this is not relevant, maybe the "powerpc: mpic" patch from this series changes this behavior.

But if this is the API, what can I say? We'll see in time. Btw, enable_irq_wake() returns an error code, normally it should be handled by printing a warning message at least, right? But since most drivers don't handle that, I'm assuming it should be left unhandled to avoid overcomplicating things.

FWIW

Acked-by Claudiu Manoil <[email protected]>

Quoted 10 lines · Manoil Claudiu

Quoted 7 lines

The device is set as wakeup capable using proper wakeup API but the driver misuses IRQF_NO_SUSPEND to set the interrupt as wakeup source which is incorrect.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

What would be the purpose of IRQF_NO_SUSPEND flag then? The flag is a friendlier API compared to calling enable_irq_wake(). For older kernels,

It's not an API, it's just a bandaid for lazy programmers.

Quoted 2 lines

on PPC architectures, the flag did the job. When did this change? Since when using IRQF_NO_SUSPEND is a "misuse"?

It always was. Simply because IRQF_NO_SUSPEND has absolutely nothing to do with wakeup interrupt sources. It's a flag which excludes the interrupt from the suspend mechanism, but it does not flag it a wakeup source.

Quoted 14 lines · Thomas Gleixner

Quoted 10 lines · Manoil Claudiu

Quoted 7 lines

The device is set as wakeup capable using proper wakeup API but the driver misuses IRQF_NO_SUSPEND to set the interrupt as wakeup source which is incorrect.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

What would be the purpose of IRQF_NO_SUSPEND flag then? The flag is a friendlier API compared to calling enable_irq_wake(). For older kernels,

It's not an API, it's just a bandaid for lazy programmers.

Quoted 7 lines

Quoted 2 lines

on PPC architectures, the flag did the job. When did this change? Since when using IRQF_NO_SUSPEND is a "misuse"?

It always was. Simply because IRQF_NO_SUSPEND has absolutely nothing to do with wakeup interrupt sources. It's a flag which excludes the interrupt from the suspend mechanism, but it does not flag it a wakeup source.

It was not very clear on the intended use of IRQF_NO_SUSPEND at the beginning as it was not documented anywhere. It's good that we have the Documentation/power/suspend-and-interrupts.txt to clarify that now.

Quoted 23 lines · Li Yang

Quoted 14 lines · Thomas Gleixner

Quoted 10 lines · Manoil Claudiu

Quoted 7 lines

The device is set as wakeup capable using proper wakeup API but the driver misuses IRQF_NO_SUSPEND to set the interrupt as wakeup source which is incorrect.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

What would be the purpose of IRQF_NO_SUSPEND flag then? The flag is a friendlier API compared to calling enable_irq_wake(). For older kernels,

It's not an API, it's just a bandaid for lazy programmers.

Quoted 7 lines

Quoted 2 lines

on PPC architectures, the flag did the job. When did this change? Since when using IRQF_NO_SUSPEND is a "misuse"?

It always was. Simply because IRQF_NO_SUSPEND has absolutely nothing to do with wakeup interrupt sources. It's a flag which excludes the interrupt from the suspend mechanism, but it does not flag it a wakeup source.

Quoted 4 lines

It was not very clear on the intended use of IRQF_NO_SUSPEND at the beginning as it was not documented anywhere. It's good that we have the Documentation/power/suspend-and-interrupts.txt to clarify that now.

This is complete and utter bullshit to put it blunt.

The documentation of IRQF_NO_SUSPEND as non suitable for marking a particular device interrupt as wakeup source was added on Feb 20 2015 with commit 737eb0301f2. Further clarification was added with commit 7438b633a6b on Mar 4 2015.

The IRQF_NO_SUSPEND crap was introduced to gianfar with commit 614b42426cc3 on Jul 31 2015.

The device is set as wakeup capable using proper wakeup API but the driver misuses IRQF_NO_SUSPEND to set the interrupt as wakeup source which is incorrect.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Cc "David S. Miller" <[email protected]>

Cc Claudiu Manoil <[email protected]>

Cc Kevin Hao <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

The device is set as wakeup capable using proper wakeup API but the driver misuses IRQF_NO_SUSPEND to set the interrupt as wakeup source which is incorrect.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Cc Srinivas Kandagatla <[email protected]>

Cc Maxime Coquelin <[email protected]>

Cc Patrice Chotard <[email protected]>

Cc Mauro Carvalho Chehab <[email protected]>

This driver doesn't claim the IR transmitter to be wakeup source. It even disables the clock and the IR during suspend-resume cycle.

This patch removes yet another misuse of IRQF_NO_SUSPEND.

Cc Mauro Carvalho Chehab <[email protected]>

Cc Zhangfei Gao <[email protected]>

Cc Patrice Chotard <[email protected]>

Cc Fabio Estevam <[email protected]>

Cc Guoxiong Yan <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
  drivers/media/rc/ir-hix5hd2.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

drivers/media/rc/ir-hix5hd2.c +0 -0

index 1c087cb76815..d0549fba711c 100644

Quoted 23 lines · Sudeep Holla

This driver doesn't claim the IR transmitter to be wakeup source. It even disables the clock and the IR during suspend-resume cycle.

This patch removes yet another misuse of IRQF_NO_SUSPEND.

Cc Mauro Carvalho Chehab <[email protected]>

Cc Zhangfei Gao <[email protected]>

Cc Patrice Chotard <[email protected]>

Cc Fabio Estevam <[email protected]>

Cc Guoxiong Yan <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
   drivers/media/rc/ir-hix5hd2.c | 2 +-
   1 file changed, 1 insertion(+), 1 deletion(-)

drivers/media/rc/ir-hix5hd2.c +0 -0

index 1c087cb76815..d0549fba711c 100644
@@ -257,7 +257,7 @@ static int hix5hd2_ir_probe(struct platform_device *pdev)
   		goto clkerr;
 

drivers/media/rc/ir-hix5hd2.c static int hix5hd2_ir_probe(struct platform_device *pdev) · line 263

Quoted 7 lines

drivers/media/rc/ir-hix5hd2.c +1 -1

   	if (devm_request_irq(dev, priv->irq, hix5hd2_ir_rx_interrupt,
-			     IRQF_NO_SUSPEND, pdev->name, priv) < 0) {
+			     0, pdev->name, priv) < 0) {
   		dev_err(dev, "IRQ %d register failed\n", priv->irq);
   		ret = -EINVAL;
   		goto regerr;

ir is wakeup source for hix5hd2, so we use IRQF_NO_SUSPEND.

However, it is true the wakeup mechanism is not realized on hix5hd2 yet.

I am fine with either using IRQF_NO_SUSPEND or not.

Quoted 8 lines · zhangfei

Quoted 5 lines · Sudeep Holla

This driver doesn't claim the IR transmitter to be wakeup source. It even disables the clock and the IR during suspend-resume cycle.

This patch removes yet another misuse of IRQF_NO_SUSPEND.

Quoted 7 lines

Quoted 7 lines

Cc Mauro Carvalho Chehab <[email protected]>

Cc Zhangfei Gao <[email protected]>

Cc Patrice Chotard <[email protected]>

Cc Fabio Estevam <[email protected]>

Cc Guoxiong Yan <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

Quoted 13 lines

Quoted 13 lines

---
   drivers/media/rc/ir-hix5hd2.c | 2 +-
   1 file changed, 1 insertion(+), 1 deletion(-)

drivers/media/rc/ir-hix5hd2.c +0 -0

index 1c087cb76815..d0549fba711c 100644
@@ -257,7 +257,7 @@ static int hix5hd2_ir_probe(struct
 platform_device *pdev)
           goto clkerr;
 

Quoted 9 lines

Quoted 7 lines

drivers/media/rc/ir-hix5hd2.c +1 -1

       if (devm_request_irq(dev, priv->irq, hix5hd2_ir_rx_interrupt,
-                 IRQF_NO_SUSPEND, pdev->name, priv) < 0) {
+                 0, pdev->name, priv) < 0) {
           dev_err(dev, "IRQ %d register failed\n", priv->irq);
           ret = -EINVAL;

           goto regerr;

ir is wakeup source for hix5hd2, so we use IRQF_NO_SUSPEND.

OK, but from the existing implementation of suspend/resume callbacks, I read that the clocks as well as the IP block is powered off. Is part of the logic always-on ?

Quoted 1 line

However, it is true the wakeup mechanism is not realized on hix5hd2 yet.

OK, then I assume you can add the right APIs(enable_irq_wake and friends) when you add that feature.

Quoted 2 lines

I am fine with either using IRQF_NO_SUSPEND or not.

No using IRQF_NO_SUSPEND for wakeup is simply wrong and hence this patch series removes all those misuse. If you need it as wakeup, then you need to use right APIs for that. Since I don't see any support for wakeup in this driver I decided to just remove the flag. Please feel free to add the support making use of right APIs.

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with irq_set_irq_wake instead.

Cc Linus Walleij <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
 drivers/pinctrl/pinctrl-single.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

drivers/pinctrl/pinctrl-single.c +0 -0

index ef04b962c3d5..a5d0f0e52de1 100644
@@ -1615,12 +1615,14 @@ static void pcs_irq_unmask(struct irq_data *d)
  */
 static int pcs_irq_set_wake(struct irq_data *d, unsigned int state)

Quoted 31 lines · Sudeep Holla

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with irq_set_irq_wake instead.

Cc Linus Walleij <[email protected]>

Cc [email protected]

Signed-off-by Sudeep Holla <[email protected]>

---
  drivers/pinctrl/pinctrl-single.c | 7 ++++---
  1 file changed, 4 insertions(+), 3 deletions(-)

drivers/pinctrl/pinctrl-single.c +2 -0

index ef04b962c3d5..a5d0f0e52de1 100644
@@ -1615,12 +1615,14 @@ static void pcs_irq_unmask(struct irq_data *d)
   */
  static int pcs_irq_set_wake(struct irq_data *d, unsigned int state)
  {
+	struct pcs_soc_data *pcs_soc = irq_data_get_irq_chip_data(d);
+
  	if (state)
  		pcs_irq_unmask(d);
  	else
  		pcs_irq_mask(d);
 

drivers/pinctrl/pinctrl-single.c static void pcs_irq_unmask(struct irq_data *d) · line 1625

Quoted 2 lines

drivers/pinctrl/pinctrl-single.c +1 -1

-	return 0;
+	return irq_set_irq_wake(pcs_soc->irq, state);

That's going to create at least a lockdep splat due to nesting irq_desc->lock.

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Cc Linus Walleij <[email protected]>

Cc linux-arm-kernel at lists.infradead.org

Signed-off-by Sudeep Holla <[email protected]>

---
 drivers/mfd/ab8500-core.c        | 11 +++++++++--
 drivers/mfd/ab8500-debugfs.c     |  2 +-
 drivers/mfd/ab8500-gpadc.c       | 15 +++++++++++----
 drivers/mfd/db8500-prcmu.c       | 24 +++++++++++++++++-------
 drivers/power/ab8500_btemp.c     |  6 ++++--
 drivers/power/ab8500_charger.c   |  6 ++++--
 drivers/power/ab8500_fg.c        |  9 ++++++---
 drivers/thermal/db8500_thermal.c |  5 ++---
 drivers/usb/phy/phy-ab8500-usb.c | 10 ++++++----
 9 files changed, 60 insertions(+), 28 deletions(-)

Quoted 12 lines · Sudeep Holla

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Cc Linus Walleij <[email protected]>

Cc linux-arm-kernel at lists.infradead.org

Signed-off-by Sudeep Holla <[email protected]>

Quoted 11 lines

---
  drivers/mfd/ab8500-core.c        | 11 +++++++++--
  drivers/mfd/ab8500-debugfs.c     |  2 +-
  drivers/mfd/ab8500-gpadc.c       | 15 +++++++++++----
  drivers/mfd/db8500-prcmu.c       | 24 +++++++++++++++++-------
  drivers/power/ab8500_btemp.c     |  6 ++++--
  drivers/power/ab8500_charger.c   |  6 ++++--
  drivers/power/ab8500_fg.c        |  9 ++++++---
  drivers/thermal/db8500_thermal.c |  5 ++---
  drivers/usb/phy/phy-ab8500-usb.c | 10 ++++++----
  9 files changed, 60 insertions(+), 28 deletions(-)

Is there a reason for bundling the changes in all of these subsystems together into a single patch?

Quoted 26 lines · Lee Jones

Quoted 23 lines · Sudeep Holla

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Cc Linus Walleij <[email protected]>

Cc linux-arm-kernel at lists.infradead.org

Signed-off-by Sudeep Holla <[email protected]>

---
 drivers/mfd/ab8500-core.c        | 11 +++++++++--
 drivers/mfd/ab8500-debugfs.c     |  2 +-
 drivers/mfd/ab8500-gpadc.c       | 15 +++++++++++----
 drivers/mfd/db8500-prcmu.c       | 24 +++++++++++++++++-------
 drivers/power/ab8500_btemp.c     |  6 ++++--
 drivers/power/ab8500_charger.c   |  6 ++++--
 drivers/power/ab8500_fg.c        |  9 ++++++---
 drivers/thermal/db8500_thermal.c |  5 ++---
 drivers/usb/phy/phy-ab8500-usb.c | 10 ++++++----
 9 files changed, 60 insertions(+), 28 deletions(-)

Quoted 3 lines

Is there a reason for bundling the changes in all of these subsystems together into a single patch?

No, I understand this needs to be split. Since I was not 100% confident about this change, I wanted LinusW to review and provide feedback and gets things working before I can split this changes. I must have added RFC, but usually it gets ignored :)

Quoted 31 lines · Sudeep Holla

Quoted 29 lines · Lee Jones

Quoted 12 lines · Sudeep Holla

The IRQF_NO_SUSPEND flag is used to identify the interrupts that should be left enabled so as to allow them to work as expected during the suspend-resume cycle, but doesn't guarantee that it will wake the system from a suspended state, enable_irq_wake is recommended to be used for the wakeup.

This patch removes the use of IRQF_NO_SUSPEND flags replacing it with enable_irq_wake instead.

Cc Linus Walleij <[email protected]>

Cc linux-arm-kernel at lists.infradead.org

Signed-off-by Sudeep Holla <[email protected]>

Quoted 11 lines

---
  drivers/mfd/ab8500-core.c        | 11 +++++++++--
  drivers/mfd/ab8500-debugfs.c     |  2 +-
  drivers/mfd/ab8500-gpadc.c       | 15 +++++++++++----
  drivers/mfd/db8500-prcmu.c       | 24 +++++++++++++++++-------
  drivers/power/ab8500_btemp.c     |  6 ++++--
  drivers/power/ab8500_charger.c   |  6 ++++--
  drivers/power/ab8500_fg.c        |  9 ++++++---
  drivers/thermal/db8500_thermal.c |  5 ++---
  drivers/usb/phy/phy-ab8500-usb.c | 10 ++++++----
  9 files changed, 60 insertions(+), 28 deletions(-)

Is there a reason for bundling the changes in all of these subsystems together into a single patch?

Quoted 4 lines

No, I understand this needs to be split. Since I was not 100% confident about this change, I wanted LinusW to review and provide feedback and gets things working before I can split this changes. I must have added RFC, but usually it gets ignored :)

Best if Linus reviews the correct patch-set. Please split and resubmit.