From mboxrd@z Thu Jan  1 00:00:00 1970
From: marc_gonzalez@sigmadesigns.com (Marc Gonzalez)
Date: Mon, 4 Jul 2016 11:01:44 +0200
Subject: [RFC PATCH v2] ARM: tango: add Suspend-to-RAM support
In-Reply-To: <5774DBDE.1070605@sigmadesigns.com>
References: <5773D3FB.4030101@sigmadesigns.com> <577409E2.7020106@arm.com>
 <5774DBDE.1070605@sigmadesigns.com>
Message-ID: <577A25F8.80702@sigmadesigns.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On 30/06/2016 10:44, Marc Gonzalez wrote:

> Ask firmware to put RAM in self-refresh mode and power system down.
> 
>     echo mem > /sys/power/state
> 
> See Documentation/power/states.txt
> 
> Signed-off-by: Marc Gonzalez <marc_gonzalez@sigmadesigns.com>
> ---
> 
> This is a RFC because I have a few doubts:
> 
> 1) return value when tango_suspend() fails and unexpectedly returns (I used -EIO)
> Is there a better error to return?
> 
> 2) conditionally building pm.c (using CONFIG_SUSPEND? CONFIG_PM_SLEEP? other?)
> Is there a better config knob to use?
> 
> 3) .valid is required, right?
> AFAICS, the answer is "yes", .valid is indeed required:
> ("no valid callback implies that none are valid")
> 
> static bool valid_state(suspend_state_t state)
> {
> 	/*
> 	 * PM_SUSPEND_STANDBY and PM_SUSPEND_MEM states need low level
> 	 * support and need to be valid to the low level
> 	 * implementation, no valid callback implies that none are valid.
> 	 */
> 	return suspend_ops && suspend_ops->valid && suspend_ops->valid(state);
> }
> 
> ---
>  arch/arm/mach-tango/Makefile |  1 +
>  arch/arm/mach-tango/pm.c     | 32 ++++++++++++++++++++++++++++++++
>  arch/arm/mach-tango/smc.h    |  1 +
>  3 files changed, 34 insertions(+)
>  create mode 100644 arch/arm/mach-tango/pm.c
> 
> diff --git a/arch/arm/mach-tango/Makefile b/arch/arm/mach-tango/Makefile
> index 7ca753aa9fb6..822b7ca803c6 100644
> --- a/arch/arm/mach-tango/Makefile
> +++ b/arch/arm/mach-tango/Makefile
> @@ -3,3 +3,4 @@ AFLAGS_smc.o := -Wa,-march=armv7-a$(plus_sec)
>  
>  obj-y += setup.o smc.o xenv.o
>  obj-$(CONFIG_SMP) += platsmp.o
> +obj-$(CONFIG_SUSPEND) += pm.o
> diff --git a/arch/arm/mach-tango/pm.c b/arch/arm/mach-tango/pm.c
> new file mode 100644
> index 000000000000..b05c6d6f99d0
> --- /dev/null
> +++ b/arch/arm/mach-tango/pm.c
> @@ -0,0 +1,32 @@
> +#include <linux/init.h>
> +#include <linux/suspend.h>
> +#include <asm/suspend.h>
> +#include "smc.h"
> +
> +static int tango_pm_powerdown(unsigned long arg)
> +{
> +	tango_suspend(virt_to_phys(cpu_resume));
> +
> +	return -EIO; /* tango_suspend has failed */
> +}
> +
> +static int tango_pm_enter(suspend_state_t state)
> +{
> +	if (state == PM_SUSPEND_MEM)
> +		return cpu_suspend(0, tango_pm_powerdown);
> +
> +	return -EINVAL;
> +}
> +
> +static const struct platform_suspend_ops tango_pm_ops = {
> +	.enter = tango_pm_enter,
> +	.valid = suspend_valid_only_mem,
> +};
> +
> +static int __init tango_pm_init(void)
> +{
> +	suspend_set_ops(&tango_pm_ops);
> +	return 0;
> +}
> +
> +late_initcall(tango_pm_init);
> diff --git a/arch/arm/mach-tango/smc.h b/arch/arm/mach-tango/smc.h
> index 3d31f984f44c..57919539da1b 100644
> --- a/arch/arm/mach-tango/smc.h
> +++ b/arch/arm/mach-tango/smc.h
> @@ -3,5 +3,6 @@ extern int tango_smc(unsigned int val, unsigned int service);
>  #define tango_set_l2_control(val)	tango_smc(val, 0x102)
>  #define tango_start_aux_core(val)	tango_smc(val, 0x104)
>  #define tango_set_aux_boot_addr(val)	tango_smc(val, 0x105)
> +#define tango_suspend(val)		tango_smc(val, 0x120)
>  #define tango_aux_core_die(val)		tango_smc(val, 0x121)
>  #define tango_aux_core_kill(val)	tango_smc(val, 0x122)
> 

Hello everyone,

I'll wait a few more days for comments, then I'll submit the patch
to arm-soc for inclusion in v4.8 (if no issues are raised).

(I suppose the return value for tango_pm_powerdown() is not that
big a deal.)

Regards.