| Message ID | 20200911064924.26281-1-andreas@kemnade.info (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | omap3: enable off mode automatically | expand |
* Andreas Kemnade <andreas@kemnade.info> [200911 09:50]: > --- a/arch/arm/mach-omap2/pm.h > +++ b/arch/arm/mach-omap2/pm.h > @@ -49,11 +49,7 @@ static inline int omap4_opp_init(void) > extern int omap3_pm_get_suspend_state(struct powerdomain *pwrdm); > extern int omap3_pm_set_suspend_state(struct powerdomain *pwrdm, int state); > > -#ifdef CONFIG_PM_DEBUG > extern u32 enable_off_mode; > -#else > -#define enable_off_mode 0 > -#endif Hmm isn't the above still needed for the other SoCs? Or is omap3 the only user? Otherwise looks good to me :) Tony
On Fri, 11 Sep 2020 13:33:37 +0300 Tony Lindgren <tony@atomide.com> wrote: > * Andreas Kemnade <andreas@kemnade.info> [200911 09:50]: > > --- a/arch/arm/mach-omap2/pm.h > > +++ b/arch/arm/mach-omap2/pm.h > > @@ -49,11 +49,7 @@ static inline int omap4_opp_init(void) > > extern int omap3_pm_get_suspend_state(struct powerdomain *pwrdm); > > extern int omap3_pm_set_suspend_state(struct powerdomain *pwrdm, int state); > > > > -#ifdef CONFIG_PM_DEBUG > > extern u32 enable_off_mode; > > -#else > > -#define enable_off_mode 0 > > -#endif > > Hmm isn't the above still needed for the other SoCs? Or is omap3 the only > user? > well, the linker moans about undefined symbols if CONFIG_ARCH_OMAP3 is not enabled. I will send a v2. Regards, Andreas
diff --git a/arch/arm/mach-omap2/pm.h b/arch/arm/mach-omap2/pm.h index 2a883a0c1fcd..80e84ae66aee 100644 --- a/arch/arm/mach-omap2/pm.h +++ b/arch/arm/mach-omap2/pm.h @@ -49,11 +49,7 @@ static inline int omap4_opp_init(void) extern int omap3_pm_get_suspend_state(struct powerdomain *pwrdm); extern int omap3_pm_set_suspend_state(struct powerdomain *pwrdm, int state); -#ifdef CONFIG_PM_DEBUG extern u32 enable_off_mode; -#else -#define enable_off_mode 0 -#endif #if defined(CONFIG_PM_DEBUG) && defined(CONFIG_DEBUG_FS) extern void pm_dbg_update_time(struct powerdomain *pwrdm, int prev); diff --git a/arch/arm/mach-omap2/pm34xx.c b/arch/arm/mach-omap2/pm34xx.c index f5dfddf492e2..d069d581c372 100644 --- a/arch/arm/mach-omap2/pm34xx.c +++ b/arch/arm/mach-omap2/pm34xx.c @@ -25,6 +25,7 @@ #include <linux/clk.h> #include <linux/delay.h> #include <linux/slab.h> +#include <linux/of.h> #include <linux/omap-gpmc.h> #include <trace/events/power.h> @@ -51,6 +52,10 @@ /* pm34xx errata defined in pm.h */ u16 pm34xx_errata; +#ifndef CONFIG_PM_DEBUG +u32 enable_off_mode; +#endif + struct power_state { struct powerdomain *pwrdm; u32 next_state; @@ -410,7 +415,12 @@ static int __init pwrdms_setup(struct powerdomain *pwrdm, void *unused) if (!pwrst) return -ENOMEM; pwrst->pwrdm = pwrdm; - pwrst->next_state = PWRDM_POWER_RET; + + if (enable_off_mode) + pwrst->next_state = PWRDM_POWER_OFF; + else + pwrst->next_state = PWRDM_POWER_RET; + list_add(&pwrst->node, &pwrst_list); if (pwrdm_has_hdwr_sar(pwrdm)) @@ -444,6 +454,21 @@ static void __init pm_errata_configure(void) } } +static void __init omap3_pm_check_pmic(void) +{ + struct device_node *np; + + np = of_find_compatible_node(NULL, NULL, "ti,twl4030-power-idle"); + if (!np) + np = of_find_compatible_node(NULL, NULL, "ti,twl4030-power-idle-osc-off"); + + if (np) { + of_node_put(np); + enable_off_mode = 1; + } else + enable_off_mode = 0; +} + int __init omap3_pm_init(void) { struct power_state *pwrst, *tmp; @@ -477,6 +502,8 @@ int __init omap3_pm_init(void) goto err2; } + omap3_pm_check_pmic(); + ret = pwrdm_for_each(pwrdms_setup, NULL); if (ret) { pr_err("Failed to setup powerdomains\n");
Enabling off mode was only reachable deeply hidden in the debugfs. As powersaving is an important feature, move the option out of its shady place. The debugfs file can still be used to override the default. Use the presence of a device compatible to ti,twl4030-idle or ti,twl4030-idle-osc-off as an indicator that the board is wired correctly for off mode. Signed-off-by: Andreas Kemnade <andreas@kemnade.info> --- An earlier version of this patch was here: https://patchwork.kernel.org/patch/10794121/ A config option was used instead of the suggested devicetree check. arch/arm/mach-omap2/pm.h | 4 ---- arch/arm/mach-omap2/pm34xx.c | 29 ++++++++++++++++++++++++++++- 2 files changed, 28 insertions(+), 5 deletions(-)