diff mbox

ARM: SAMSUNG: Gracefully exit on suspend failure

Message ID 1357729952-14881-1-git-send-email-a.kesavan@samsung.com (mailing list archive)
State New, archived
Headers show

Commit Message

Abhilash Kesavan Jan. 9, 2013, 11:12 a.m. UTC
As per the Exynos5250 User Manual:

When there are pending interrupt events, WFI/WFE instruction are
ignored. To cancel the power-down sequence follow these steps:
1) Disable system power-down through CENTRAL_SEQ_CONFIGURATION
and CENTRAL_SEQ_CONFIGURATION_DMC registers
2) Clear WAKEUP_STAT register
3) Enable interrupt service routine for CPU

Code for early wakeup for exynos already exists; remove the panic
on suspend failure, clear the wakeup state register and return
error from cpu_suspend to indicate a failed suspend (to a user
daemon).

Older Samsung SoCs have similar panics and I have removed them all.
Haven't touched the S3C2410 sleep code.

Signed-off-by: Abhilash Kesavan <a.kesavan@samsung.com>
---
 arch/arm/mach-exynos/pm.c          |    6 ++++--
 arch/arm/mach-s3c24xx/pm-s3c2412.c |    3 ++-
 arch/arm/mach-s3c24xx/pm-s3c2416.c |    3 ++-
 arch/arm/mach-s3c64xx/pm.c         |    3 ++-
 arch/arm/mach-s5p64x0/pm.c         |    4 ++--
 arch/arm/mach-s5pv210/pm.c         |    4 ++--
 arch/arm/plat-samsung/pm.c         |    5 ++++-
 7 files changed, 18 insertions(+), 10 deletions(-)

Comments

Kyungmin Park Jan. 9, 2013, 7:46 p.m. UTC | #1
Hi,
I wodner do you see the real case it's happened?
It it's occured, does it better to find out the root cause?

Thank you,
Kyungmin Park

On Wed, Jan 9, 2013 at 8:12 PM, Abhilash Kesavan <a.kesavan@samsung.com> wrote:
> As per the Exynos5250 User Manual:
>
> When there are pending interrupt events, WFI/WFE instruction are
> ignored. To cancel the power-down sequence follow these steps:
> 1) Disable system power-down through CENTRAL_SEQ_CONFIGURATION
> and CENTRAL_SEQ_CONFIGURATION_DMC registers
> 2) Clear WAKEUP_STAT register
> 3) Enable interrupt service routine for CPU
>
> Code for early wakeup for exynos already exists; remove the panic
> on suspend failure, clear the wakeup state register and return
> error from cpu_suspend to indicate a failed suspend (to a user
> daemon).
>
> Older Samsung SoCs have similar panics and I have removed them all.
> Haven't touched the S3C2410 sleep code.
>
> Signed-off-by: Abhilash Kesavan <a.kesavan@samsung.com>
> ---
>  arch/arm/mach-exynos/pm.c          |    6 ++++--
>  arch/arm/mach-s3c24xx/pm-s3c2412.c |    3 ++-
>  arch/arm/mach-s3c24xx/pm-s3c2416.c |    3 ++-
>  arch/arm/mach-s3c64xx/pm.c         |    3 ++-
>  arch/arm/mach-s5p64x0/pm.c         |    4 ++--
>  arch/arm/mach-s5pv210/pm.c         |    4 ++--
>  arch/arm/plat-samsung/pm.c         |    5 ++++-
>  7 files changed, 18 insertions(+), 10 deletions(-)
>
> diff --git a/arch/arm/mach-exynos/pm.c b/arch/arm/mach-exynos/pm.c
> index f459afd..55df38cd 100644
> --- a/arch/arm/mach-exynos/pm.c
> +++ b/arch/arm/mach-exynos/pm.c
> @@ -92,8 +92,8 @@ static int exynos_cpu_suspend(unsigned long arg)
>         /* issue the standby signal into the pm unit. */
>         cpu_do_idle();
>
> -       /* we should never get past here */
> -       panic("sleep resumed to originator?");
> +       pr_err("Failed to suspend the system\n");
> +       return -EBUSY;
>  }
>
>  static void exynos_pm_prepare(void)
> @@ -283,6 +283,8 @@ static void exynos_pm_resume(void)
>         if (!(tmp & S5P_CENTRAL_LOWPWR_CFG)) {
>                 tmp |= S5P_CENTRAL_LOWPWR_CFG;
>                 __raw_writel(tmp, S5P_CENTRAL_SEQ_CONFIGURATION);
> +               /* clear the wakeup state register */
> +               __raw_writel(0x0, S5P_WAKEUP_STAT);
>                 /* No need to perform below restore code */
>                 goto early_wakeup;
>         }
> diff --git a/arch/arm/mach-s3c24xx/pm-s3c2412.c b/arch/arm/mach-s3c24xx/pm-s3c2412.c
> index 206765c..a7bc30d 100644
> --- a/arch/arm/mach-s3c24xx/pm-s3c2412.c
> +++ b/arch/arm/mach-s3c24xx/pm-s3c2412.c
> @@ -48,7 +48,8 @@ static int s3c2412_cpu_suspend(unsigned long arg)
>
>         s3c2412_sleep_enter();
>
> -       panic("sleep resumed to originator?");
> +       pr_err("Failed to suspend the system\n");
> +       return -EBUSY;
>  }
>
>  static void s3c2412_pm_prepare(void)
> diff --git a/arch/arm/mach-s3c24xx/pm-s3c2416.c b/arch/arm/mach-s3c24xx/pm-s3c2416.c
> index 1bd4817..97a8d70 100644
> --- a/arch/arm/mach-s3c24xx/pm-s3c2416.c
> +++ b/arch/arm/mach-s3c24xx/pm-s3c2416.c
> @@ -34,7 +34,8 @@ static int s3c2416_cpu_suspend(unsigned long arg)
>
>         s3c2412_sleep_enter();
>
> -       panic("sleep resumed to originator?");
> +       pr_err("Failed to suspend the system\n");
> +       return -EBUSY;
>  }
>
>  static void s3c2416_pm_prepare(void)
> diff --git a/arch/arm/mach-s3c64xx/pm.c b/arch/arm/mach-s3c64xx/pm.c
> index f6b13a2..4d1accc 100644
> --- a/arch/arm/mach-s3c64xx/pm.c
> +++ b/arch/arm/mach-s3c64xx/pm.c
> @@ -297,7 +297,8 @@ static int s3c64xx_cpu_suspend(unsigned long arg)
>
>         /* we should never get past here */
>
> -       panic("sleep resumed to originator?");
> +       pr_err("Failed to suspend the system\n");
> +       return -EBUSY;
>  }
>
>  /* mapping of interrupts to parts of the wakeup mask */
> diff --git a/arch/arm/mach-s5p64x0/pm.c b/arch/arm/mach-s5p64x0/pm.c
> index 9cba18b..710651d 100644
> --- a/arch/arm/mach-s5p64x0/pm.c
> +++ b/arch/arm/mach-s5p64x0/pm.c
> @@ -103,8 +103,8 @@ static int s5p64x0_cpu_suspend(unsigned long arg)
>             "mcr p15, 0, %0, c7, c10, 4\n\t"
>             "mcr p15, 0, %0, c7, c0, 4" : : "r" (tmp));
>
> -       /* we should never get past here */
> -       panic("sleep resumed to originator?");
> +       pr_err("Failed to suspend the system\n");
> +       return -EBUSY;
>  }
>
>  /* mapping of interrupts to parts of the wakeup mask */
> diff --git a/arch/arm/mach-s5pv210/pm.c b/arch/arm/mach-s5pv210/pm.c
> index 736bfb1..c584f49 100644
> --- a/arch/arm/mach-s5pv210/pm.c
> +++ b/arch/arm/mach-s5pv210/pm.c
> @@ -104,8 +104,8 @@ static int s5pv210_cpu_suspend(unsigned long arg)
>             "mcr p15, 0, %0, c7, c10, 4\n\t"
>             "wfi" : : "r" (tmp));
>
> -       /* we should never get past here */
> -       panic("sleep resumed to originator?");
> +       pr_err("Failed to suspend the system\n");
> +       return -EBUSY;
>  }
>
>  static void s5pv210_pm_prepare(void)
> diff --git a/arch/arm/plat-samsung/pm.c b/arch/arm/plat-samsung/pm.c
> index 1507028..aa4b06d 100644
> --- a/arch/arm/plat-samsung/pm.c
> +++ b/arch/arm/plat-samsung/pm.c
> @@ -243,6 +243,7 @@ int (*pm_cpu_sleep)(unsigned long);
>
>  static int s3c_pm_enter(suspend_state_t state)
>  {
> +       int ret;
>         /* ensure the debug is initialised (if enabled) */
>
>         s3c_pm_debug_init();
> @@ -300,7 +301,9 @@ static int s3c_pm_enter(suspend_state_t state)
>          * we resume as it saves its own register state and restores it
>          * during the resume.  */
>
> -       cpu_suspend(0, pm_cpu_sleep);
> +       ret = cpu_suspend(0, pm_cpu_sleep);
> +       if (ret)
> +               return ret;
>
>         /* restore the system state */
>
> --
> 1.7.9.5
>
>
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Russell King - ARM Linux Jan. 9, 2013, 7:52 p.m. UTC | #2
On Thu, Jan 10, 2013 at 04:46:45AM +0900, Kyungmin Park wrote:
> Hi,
> I wodner do you see the real case it's happened?
> It it's occured, does it better to find out the root cause?

No.  It's better that if it does happen, the system remains functional
so you _can_ debug why it happened.  If you cause the system to panic
then you have to find some other way to get information out of it.
Kim Kukjin Jan. 10, 2013, 12:32 a.m. UTC | #3
Abhilash Kesavan wrote:
> 
> As per the Exynos5250 User Manual:
> 
> When there are pending interrupt events, WFI/WFE instruction are
> ignored. To cancel the power-down sequence follow these steps:
> 1) Disable system power-down through CENTRAL_SEQ_CONFIGURATION
> and CENTRAL_SEQ_CONFIGURATION_DMC registers
> 2) Clear WAKEUP_STAT register
> 3) Enable interrupt service routine for CPU
> 
> Code for early wakeup for exynos already exists; remove the panic
> on suspend failure, clear the wakeup state register and return
> error from cpu_suspend to indicate a failed suspend (to a user
> daemon).
> 
> Older Samsung SoCs have similar panics and I have removed them all.
> Haven't touched the S3C2410 sleep code.
> 
> Signed-off-by: Abhilash Kesavan <a.kesavan@samsung.com>
> ---
>  arch/arm/mach-exynos/pm.c          |    6 ++++--
>  arch/arm/mach-s3c24xx/pm-s3c2412.c |    3 ++-
>  arch/arm/mach-s3c24xx/pm-s3c2416.c |    3 ++-
>  arch/arm/mach-s3c64xx/pm.c         |    3 ++-
>  arch/arm/mach-s5p64x0/pm.c         |    4 ++--
>  arch/arm/mach-s5pv210/pm.c         |    4 ++--
>  arch/arm/plat-samsung/pm.c         |    5 ++++-
>  7 files changed, 18 insertions(+), 10 deletions(-)
> 
> diff --git a/arch/arm/mach-exynos/pm.c b/arch/arm/mach-exynos/pm.c
> index f459afd..55df38cd 100644
> --- a/arch/arm/mach-exynos/pm.c
> +++ b/arch/arm/mach-exynos/pm.c
> @@ -92,8 +92,8 @@ static int exynos_cpu_suspend(unsigned long arg)
>  	/* issue the standby signal into the pm unit. */
>  	cpu_do_idle();
> 
> -	/* we should never get past here */
> -	panic("sleep resumed to originator?");
> +	pr_err("Failed to suspend the system\n");
> +	return -EBUSY;

This can be happened under non-error situation, so this should be:

+	pr_info("sleep resumed to originator?");
+	return 1; /* abort suspend */

[...]

- Kukjin
Abhilash Kesavan Jan. 18, 2013, 1:10 p.m. UTC | #4
Hi,

On Thu, Jan 10, 2013 at 1:16 AM, Kyungmin Park <kmpark@infradead.org> wrote:
> Hi,
> I wodner do you see the real case it's happened?
> It it's occured, does it better to find out the root cause?
>
We have seen infrequent suspend failures on the Exynos5. This patch
helps us obtain
failure statistics over a period of time and continue debugging the issue.
> Thank you,
> Kyungmin Park

Abhilash
Abhilash Kesavan Jan. 18, 2013, 1:10 p.m. UTC | #5
Hi,

>> -     panic("sleep resumed to originator?");
>> +     pr_err("Failed to suspend the system\n");
>> +     return -EBUSY;
>
> This can be happened under non-error situation, so this should be:
>
> +       pr_info("sleep resumed to originator?");
> +       return 1; /* abort suspend */
>
Will re-post with suggested changes.

Abhilash
diff mbox

Patch

diff --git a/arch/arm/mach-exynos/pm.c b/arch/arm/mach-exynos/pm.c
index f459afd..55df38cd 100644
--- a/arch/arm/mach-exynos/pm.c
+++ b/arch/arm/mach-exynos/pm.c
@@ -92,8 +92,8 @@  static int exynos_cpu_suspend(unsigned long arg)
 	/* issue the standby signal into the pm unit. */
 	cpu_do_idle();
 
-	/* we should never get past here */
-	panic("sleep resumed to originator?");
+	pr_err("Failed to suspend the system\n");
+	return -EBUSY;
 }
 
 static void exynos_pm_prepare(void)
@@ -283,6 +283,8 @@  static void exynos_pm_resume(void)
 	if (!(tmp & S5P_CENTRAL_LOWPWR_CFG)) {
 		tmp |= S5P_CENTRAL_LOWPWR_CFG;
 		__raw_writel(tmp, S5P_CENTRAL_SEQ_CONFIGURATION);
+		/* clear the wakeup state register */
+		__raw_writel(0x0, S5P_WAKEUP_STAT);
 		/* No need to perform below restore code */
 		goto early_wakeup;
 	}
diff --git a/arch/arm/mach-s3c24xx/pm-s3c2412.c b/arch/arm/mach-s3c24xx/pm-s3c2412.c
index 206765c..a7bc30d 100644
--- a/arch/arm/mach-s3c24xx/pm-s3c2412.c
+++ b/arch/arm/mach-s3c24xx/pm-s3c2412.c
@@ -48,7 +48,8 @@  static int s3c2412_cpu_suspend(unsigned long arg)
 
 	s3c2412_sleep_enter();
 
-	panic("sleep resumed to originator?");
+	pr_err("Failed to suspend the system\n");
+	return -EBUSY;
 }
 
 static void s3c2412_pm_prepare(void)
diff --git a/arch/arm/mach-s3c24xx/pm-s3c2416.c b/arch/arm/mach-s3c24xx/pm-s3c2416.c
index 1bd4817..97a8d70 100644
--- a/arch/arm/mach-s3c24xx/pm-s3c2416.c
+++ b/arch/arm/mach-s3c24xx/pm-s3c2416.c
@@ -34,7 +34,8 @@  static int s3c2416_cpu_suspend(unsigned long arg)
 
 	s3c2412_sleep_enter();
 
-	panic("sleep resumed to originator?");
+	pr_err("Failed to suspend the system\n");
+	return -EBUSY;
 }
 
 static void s3c2416_pm_prepare(void)
diff --git a/arch/arm/mach-s3c64xx/pm.c b/arch/arm/mach-s3c64xx/pm.c
index f6b13a2..4d1accc 100644
--- a/arch/arm/mach-s3c64xx/pm.c
+++ b/arch/arm/mach-s3c64xx/pm.c
@@ -297,7 +297,8 @@  static int s3c64xx_cpu_suspend(unsigned long arg)
 
 	/* we should never get past here */
 
-	panic("sleep resumed to originator?");
+	pr_err("Failed to suspend the system\n");
+	return -EBUSY;
 }
 
 /* mapping of interrupts to parts of the wakeup mask */
diff --git a/arch/arm/mach-s5p64x0/pm.c b/arch/arm/mach-s5p64x0/pm.c
index 9cba18b..710651d 100644
--- a/arch/arm/mach-s5p64x0/pm.c
+++ b/arch/arm/mach-s5p64x0/pm.c
@@ -103,8 +103,8 @@  static int s5p64x0_cpu_suspend(unsigned long arg)
 	    "mcr p15, 0, %0, c7, c10, 4\n\t"
 	    "mcr p15, 0, %0, c7, c0, 4" : : "r" (tmp));
 
-	/* we should never get past here */
-	panic("sleep resumed to originator?");
+	pr_err("Failed to suspend the system\n");
+	return -EBUSY;
 }
 
 /* mapping of interrupts to parts of the wakeup mask */
diff --git a/arch/arm/mach-s5pv210/pm.c b/arch/arm/mach-s5pv210/pm.c
index 736bfb1..c584f49 100644
--- a/arch/arm/mach-s5pv210/pm.c
+++ b/arch/arm/mach-s5pv210/pm.c
@@ -104,8 +104,8 @@  static int s5pv210_cpu_suspend(unsigned long arg)
 	    "mcr p15, 0, %0, c7, c10, 4\n\t"
 	    "wfi" : : "r" (tmp));
 
-	/* we should never get past here */
-	panic("sleep resumed to originator?");
+	pr_err("Failed to suspend the system\n");
+	return -EBUSY;
 }
 
 static void s5pv210_pm_prepare(void)
diff --git a/arch/arm/plat-samsung/pm.c b/arch/arm/plat-samsung/pm.c
index 1507028..aa4b06d 100644
--- a/arch/arm/plat-samsung/pm.c
+++ b/arch/arm/plat-samsung/pm.c
@@ -243,6 +243,7 @@  int (*pm_cpu_sleep)(unsigned long);
 
 static int s3c_pm_enter(suspend_state_t state)
 {
+	int ret;
 	/* ensure the debug is initialised (if enabled) */
 
 	s3c_pm_debug_init();
@@ -300,7 +301,9 @@  static int s3c_pm_enter(suspend_state_t state)
 	 * we resume as it saves its own register state and restores it
 	 * during the resume.  */
 
-	cpu_suspend(0, pm_cpu_sleep);
+	ret = cpu_suspend(0, pm_cpu_sleep);
+	if (ret)
+		return ret;
 
 	/* restore the system state */