[PATCH 1/6] ARM: at91: pm: rework cpu detection

[Boris Brezillon]

Hi Jean-Christophe,

On Wed, 14 Jan 2015 20:14:12 +0100
Jean-Christophe PLAGNIOL-VILLARD <plagnioj at jcrosoft.com> wrote:

> On 22:23 Mon 12 Jan     , Alexandre Belloni wrote:
> > Store SoC differences in a struct to remove cpu_is_* usage.
> > 
> > Signed-off-by: Alexandre Belloni <alexandre.belloni at free-electrons.com>
> > ---
> >  arch/arm/mach-at91/pm.c | 54 ++++++++++++++++++++++++++++++-------------------
> >  1 file changed, 33 insertions(+), 21 deletions(-)
> > 
> > diff --git a/arch/arm/mach-at91/pm.c b/arch/arm/mach-at91/pm.c
> > index 9b15169a1c62..79aa793d1f00 100644
> > --- a/arch/arm/mach-at91/pm.c
> > +++ b/arch/arm/mach-at91/pm.c
> > @@ -17,6 +17,7 @@
> >  #include <linux/interrupt.h>
> >  #include <linux/sysfs.h>
> >  #include <linux/module.h>
> > +#include <linux/of.h>
> >  #include <linux/platform_device.h>
> >  #include <linux/io.h>
> >  #include <linux/clk/at91_pmc.h>
> > @@ -32,6 +33,11 @@
> >  #include "generic.h"
> >  #include "pm.h"
> >  
> > +static struct {
> > +	unsigned long uhp_udp_mask;
> > +	int memctrl;
> > +} at91_pm_data;
> > +
> >  static void (*at91_pm_standby)(void);
> >  
> >  static int at91_pm_valid_state(suspend_state_t state)
> > @@ -71,17 +77,9 @@ static int at91_pm_verify_clocks(void)
> >  	scsr = at91_pmc_read(AT91_PMC_SCSR);
> >  
> >  	/* USB must not be using PLLB */
> > -	if (cpu_is_at91rm9200()) {
> > -		if ((scsr & (AT91RM9200_PMC_UHP | AT91RM9200_PMC_UDP)) != 0) {
> > -			pr_err("AT91: PM - Suspend-to-RAM with USB still active\n");
> > -			return 0;
> > -		}
> > -	} else if (cpu_is_at91sam9260() || cpu_is_at91sam9261() || cpu_is_at91sam9263()
> > -			|| cpu_is_at91sam9g20() || cpu_is_at91sam9g10()) {
> > -		if ((scsr & (AT91SAM926x_PMC_UHP | AT91SAM926x_PMC_UDP)) != 0) {
> > -			pr_err("AT91: PM - Suspend-to-RAM with USB still active\n");
> > -			return 0;
> > -		}
> > +	if ((scsr & at91_pm_data.uhp_udp_mask) != 0) {
> > +		pr_err("AT91: PM - Suspend-to-RAM with USB still active\n");
> > +		return 0;
> >  	}
> >  
> >  	/* PCK0..PCK3 must be disabled, or configured to use clk32k */
> > @@ -149,18 +147,13 @@ static int at91_pm_enter(suspend_state_t state)
> >  			 * turning off the main oscillator; reverse on wakeup.
> >  			 */
> >  			if (slow_clock) {
> > -				int memctrl = AT91_MEMCTRL_SDRAMC;
> > -
> > -				if (cpu_is_at91rm9200())
> > -					memctrl = AT91_MEMCTRL_MC;
> > -				else if (cpu_is_at91sam9g45())
> > -					memctrl = AT91_MEMCTRL_DDRSDR;
> >  #ifdef CONFIG_AT91_SLOW_CLOCK
> >  				/* copy slow_clock handler to SRAM, and call it */
> >  				memcpy(slow_clock, at91_slow_clock, at91_slow_clock_sz);
> >  #endif
> >  				slow_clock(at91_pmc_base, at91_ramc_base[0],
> > -					   at91_ramc_base[1], memctrl);
> > +					   at91_ramc_base[1],
> > +					   at91_pm_data.memctrl);
> >  				break;
> >  			} else {
> >  				pr_info("AT91: PM - no slow clock mode enabled ...\n");
> > @@ -237,10 +230,29 @@ static int __init at91_pm_init(void)
> >  
> >  	pr_info("AT91: Power Management%s\n", (slow_clock ? " (with slow clock mode)" : ""));
> >  
> > -	/* AT91RM9200 SDRAM low-power mode cannot be used with self-refresh. */
> > -	if (cpu_is_at91rm9200())
> > +	at91_pm_data.memctrl = AT91_MEMCTRL_SDRAMC;
> > +
> > +	if (of_machine_is_compatible("atmel,at91rm9200")) {
> > +		/*
> > +		 * AT91RM9200 SDRAM low-power mode cannot be used with
> > +		 * self-refresh.
> > +		 */
> >  		at91_ramc_write(0, AT91RM9200_SDRAMC_LPR, 0);
> > -	
> > +
> > +		at91_pm_data.uhp_udp_mask = AT91RM9200_PMC_UHP |
> > +					    AT91RM9200_PMC_UDP;
> > +		at91_pm_data.memctrl = AT91_MEMCTRL_MC;
> > +	} else if (of_machine_is_compatible("atmel,at91sam9260") ||
> > +		   of_machine_is_compatible("atmel,at91sam9g20") ||
> > +		   of_machine_is_compatible("atmel,at91sam9261") ||
> > +		   of_machine_is_compatible("atmel,at91sam9g10") ||
> > +		   of_machine_is_compatible("atmel,at91sam9263")) {
> nack here
> 
> you switch for runtime information from the SOC register store in ram to DT
> 
> DT is great but I do prefer to rely on the SoC register here as the whole was
> also to check that the is correct

Well, cpu_is_xxx macros are defined in a mach specific header, and we're
currently trying to enable multi platform support. 

> 
> wihich is way slower

Hmm, this SoC detection has been move from the suspend path (where, as
you stated, speed is a real concern) to the pm init function (which is
only called once at startup), and I doubt the boot time penalty will
even be noticeable...

Best Regards,

Boris


-- 
Boris Brezillon, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com