[PATCH v2] OMAP2+: disable idle early in the suspend sequence

[PATCH v2] OMAP2+: disable idle early in the suspend sequence

[jean.pihet]

From: Jean Pihet <j-pihet@ti.com>

Some bad interaction between the idle and the suspend paths has been
noticed: the idle code is called during the suspend enter and exit
sequences. This could cause corruption or lock-up of resources.

The solution is to move the calls to disable_hlt at the very beginning
of the suspend sequence (ex. in omap3_pm_begin instead of
omap3_pm_prepare), and the call to enable_hlt at the very end of
the suspend sequence (ex. in omap3_pm_end instead of omap3_pm_finish).

Tested with RET and OFF on Beagle and OMAP3EVM.

Signed-off-by: Jean Pihet <j-pihet@ti.com>
Cc: Kevin Hilman <khilman@deeprootsystems.com>
---
v1: support for OMAP3 only.
 Tested on board.

v2: Added support for OMAP2 and OMAP4.
 Compile tested only for OMAP2 and OMAP4.
 Tested on board for OMAP3.

 arch/arm/mach-omap2/pm24xx.c |   16 ++++++++++++++--
 arch/arm/mach-omap2/pm34xx.c |    4 ++--
 arch/arm/mach-omap2/pm44xx.c |    4 ++--
 3 files changed, 18 insertions(+), 6 deletions(-)

diff --git a/arch/arm/mach-omap2/pm24xx.c b/arch/arm/mach-omap2/pm24xx.c
index c85923e..3820179 100644
--- a/arch/arm/mach-omap2/pm24xx.c
+++ b/arch/arm/mach-omap2/pm24xx.c
@@ -286,8 +286,6 @@ out:
 
 static int omap2_pm_prepare(void)
 {
-	/* We cannot sleep in idle until we have resumed */
-	disable_hlt();
 	return 0;
 }
 
@@ -330,10 +328,24 @@ static int omap2_pm_enter(suspend_state_t state)
 
 static void omap2_pm_finish(void)
 {
+}
+
+static int omap2_pm_begin(suspend_state_t state)
+{
+	/* We cannot sleep in idle until we have resumed */
+	disable_hlt();
+	return 0;
+}
+
+static void omap2_pm_end(void)
+{
 	enable_hlt();
+	return;
 }
 
 static struct platform_suspend_ops omap_pm_ops = {
+	.begin		= omap2_pm_begin,
+	.end		= omap2_pm_end,
 	.prepare	= omap2_pm_prepare,
 	.enter		= omap2_pm_enter,
 	.finish		= omap2_pm_finish,
diff --git a/arch/arm/mach-omap2/pm34xx.c b/arch/arm/mach-omap2/pm34xx.c
index 0ec8a04..ec80d38 100644
--- a/arch/arm/mach-omap2/pm34xx.c
+++ b/arch/arm/mach-omap2/pm34xx.c
@@ -518,7 +518,6 @@ static suspend_state_t suspend_state;
 
 static int omap3_pm_prepare(void)
 {
-	disable_hlt();
 	return 0;
 }
 
@@ -586,12 +585,12 @@ static int omap3_pm_enter(suspend_state_t unused)
 
 static void omap3_pm_finish(void)
 {
-	enable_hlt();
 }
 
 /* Hooks to enable / disable UART interrupts during suspend */
 static int omap3_pm_begin(suspend_state_t state)
 {
+	disable_hlt();
 	suspend_state = state;
 	omap_uart_enable_irqs(0);
 	return 0;
@@ -601,6 +600,7 @@ static void omap3_pm_end(void)
 {
 	suspend_state = PM_SUSPEND_ON;
 	omap_uart_enable_irqs(1);
+	enable_hlt();
 	return;
 }
 
diff --git a/arch/arm/mach-omap2/pm44xx.c b/arch/arm/mach-omap2/pm44xx.c
index 54544b4..f4c4fab 100644
--- a/arch/arm/mach-omap2/pm44xx.c
+++ b/arch/arm/mach-omap2/pm44xx.c
@@ -33,7 +33,6 @@ static LIST_HEAD(pwrst_list);
 #ifdef CONFIG_SUSPEND
 static int omap4_pm_prepare(void)
 {
-	disable_hlt();
 	return 0;
 }
 
@@ -61,17 +60,18 @@ static int omap4_pm_enter(suspend_state_t suspend_state)
 
 static void omap4_pm_finish(void)
 {
-	enable_hlt();
 	return;
 }
 
 static int omap4_pm_begin(suspend_state_t state)
 {
+	disable_hlt();
 	return 0;
 }
 
 static void omap4_pm_end(void)
 {
+	enable_hlt();
 	return;
 }
 
-- 
1.7.1

Re: [PATCH v2] OMAP2+: disable idle early in the suspend sequence

[Kevin Hilman]

jean.pihet@newoldbits.com writes:

> From: Jean Pihet <j-pihet@ti.com>
>
> Some bad interaction between the idle and the suspend paths has been
> noticed: the idle code is called during the suspend enter and exit
> sequences. This could cause corruption or lock-up of resources.
>
> The solution is to move the calls to disable_hlt at the very beginning
> of the suspend sequence (ex. in omap3_pm_begin instead of
> omap3_pm_prepare), and the call to enable_hlt at the very end of
> the suspend sequence (ex. in omap3_pm_end instead of omap3_pm_finish).

In an earlier version, I thought we agreed to just remove the empty
prepare/finish calls.  Can you do that?

Also, can you base this on top of my recently posted patch:

[PATCH v3] OMAP2+: PM/serial: fix console semaphore acquire during suspend

since it touches some of the same code for pm24xx.c

Thanks,

Kevin

> Tested with RET and OFF on Beagle and OMAP3EVM.
>
> Signed-off-by: Jean Pihet <j-pihet@ti.com>
> Cc: Kevin Hilman <khilman@deeprootsystems.com>
> ---
> v1: support for OMAP3 only.
>  Tested on board.
>
> v2: Added support for OMAP2 and OMAP4.
>  Compile tested only for OMAP2 and OMAP4.
>  Tested on board for OMAP3.
>
>  arch/arm/mach-omap2/pm24xx.c |   16 ++++++++++++++--
>  arch/arm/mach-omap2/pm34xx.c |    4 ++--
>  arch/arm/mach-omap2/pm44xx.c |    4 ++--
>  3 files changed, 18 insertions(+), 6 deletions(-)
>
> diff --git a/arch/arm/mach-omap2/pm24xx.c b/arch/arm/mach-omap2/pm24xx.c
> index c85923e..3820179 100644
> --- a/arch/arm/mach-omap2/pm24xx.c
> +++ b/arch/arm/mach-omap2/pm24xx.c
> @@ -286,8 +286,6 @@ out:
>  
>  static int omap2_pm_prepare(void)
>  {
> -	/* We cannot sleep in idle until we have resumed */
> -	disable_hlt();
>  	return 0;
>  }
>
> @@ -330,10 +328,24 @@ static int omap2_pm_enter(suspend_state_t state)
>  
>  static void omap2_pm_finish(void)
>  {
> +}
>
> +
> +static int omap2_pm_begin(suspend_state_t state)
> +{
> +	/* We cannot sleep in idle until we have resumed */
> +	disable_hlt();
> +	return 0;
> +}
> +
> +static void omap2_pm_end(void)
> +{
>  	enable_hlt();
> +	return;
>  }
>  
>  static struct platform_suspend_ops omap_pm_ops = {
> +	.begin		= omap2_pm_begin,
> +	.end		= omap2_pm_end,
>  	.prepare	= omap2_pm_prepare,
>  	.enter		= omap2_pm_enter,
>  	.finish		= omap2_pm_finish,
> diff --git a/arch/arm/mach-omap2/pm34xx.c b/arch/arm/mach-omap2/pm34xx.c
> index 0ec8a04..ec80d38 100644
> --- a/arch/arm/mach-omap2/pm34xx.c
> +++ b/arch/arm/mach-omap2/pm34xx.c
> @@ -518,7 +518,6 @@ static suspend_state_t suspend_state;
>  
>  static int omap3_pm_prepare(void)
>  {
> -	disable_hlt();
>  	return 0;
>  }
>  
> @@ -586,12 +585,12 @@ static int omap3_pm_enter(suspend_state_t unused)
>  
>  static void omap3_pm_finish(void)
>  {
> -	enable_hlt();
>  }
>  
>  /* Hooks to enable / disable UART interrupts during suspend */
>  static int omap3_pm_begin(suspend_state_t state)
>  {
> +	disable_hlt();
>  	suspend_state = state;
>  	omap_uart_enable_irqs(0);
>  	return 0;
> @@ -601,6 +600,7 @@ static void omap3_pm_end(void)
>  {
>  	suspend_state = PM_SUSPEND_ON;
>  	omap_uart_enable_irqs(1);
> +	enable_hlt();
>  	return;
>  }
>  
> diff --git a/arch/arm/mach-omap2/pm44xx.c b/arch/arm/mach-omap2/pm44xx.c
> index 54544b4..f4c4fab 100644
> --- a/arch/arm/mach-omap2/pm44xx.c
> +++ b/arch/arm/mach-omap2/pm44xx.c
> @@ -33,7 +33,6 @@ static LIST_HEAD(pwrst_list);
>  #ifdef CONFIG_SUSPEND
>  static int omap4_pm_prepare(void)
>  {
> -	disable_hlt();
>  	return 0;
>  }
>  
> @@ -61,17 +60,18 @@ static int omap4_pm_enter(suspend_state_t suspend_state)
>  
>  static void omap4_pm_finish(void)
>  {
> -	enable_hlt();
>  	return;
>  }
>  
>  static int omap4_pm_begin(suspend_state_t state)
>  {
> +	disable_hlt();
>  	return 0;
>  }
>  
>  static void omap4_pm_end(void)
>  {
> +	enable_hlt();
>  	return;
>  }

Re: [PATCH v2] OMAP2+: disable idle early in the suspend sequence

[Jean Pihet]

Kevin,

On Thu, Dec 9, 2010 at 12:01 AM, Kevin Hilman
<khilman@deeprootsystems.com> wrote:
> jean.pihet@newoldbits.com writes:
>
>> From: Jean Pihet <j-pihet@ti.com>
>>
>> Some bad interaction between the idle and the suspend paths has been
>> noticed: the idle code is called during the suspend enter and exit
>> sequences. This could cause corruption or lock-up of resources.
>>
>> The solution is to move the calls to disable_hlt at the very beginning
>> of the suspend sequence (ex. in omap3_pm_begin instead of
>> omap3_pm_prepare), and the call to enable_hlt at the very end of
>> the suspend sequence (ex. in omap3_pm_end instead of omap3_pm_finish).
>
> In an earlier version, I thought we agreed to just remove the empty
> prepare/finish calls.  Can you do that?
>
> Also, can you base this on top of my recently posted patch:
>
> [PATCH v3] OMAP2+: PM/serial: fix console semaphore acquire during suspend
>
> since it touches some of the same code for pm24xx.c

Ok sent as '[PATCH v3] OMAP2+: disable idle early in the suspend sequence'

Thanks,
Jean

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