On 2/5/19 7:30 PM, Tomasz Duszynski wrote: > On Tue, Feb 05, 2019 at 01:40:27PM +0100, Fabrice Gasnier wrote: >> Add suspend/resume PM sleep ops. When going to low power, disable >> active PWM channel. Active PWM channel is resumed, by calling >> pwm_apply_state(). This is inspired by Thierry's comment in [1]. >> Don't touch inactive channels, as it may be used by other LPTimer MFD >> child driver. >> [1]https://lkml.org/lkml/2017/12/5/175 >> >> Signed-off-by: Fabrice Gasnier <fabrice.gasnier@xxxxxx> >> --- >> drivers/pwm/pwm-stm32-lp.c | 38 ++++++++++++++++++++++++++++++++++++++ >> 1 file changed, 38 insertions(+) >> >> diff --git a/drivers/pwm/pwm-stm32-lp.c b/drivers/pwm/pwm-stm32-lp.c >> index 0059b24c..0c40d48 100644 >> --- a/drivers/pwm/pwm-stm32-lp.c >> +++ b/drivers/pwm/pwm-stm32-lp.c >> @@ -13,6 +13,7 @@ >> #include <linux/mfd/stm32-lptimer.h> >> #include <linux/module.h> >> #include <linux/of.h> >> +#include <linux/pinctrl/consumer.h> >> #include <linux/platform_device.h> >> #include <linux/pwm.h> >> >> @@ -20,6 +21,8 @@ struct stm32_pwm_lp { >> struct pwm_chip chip; >> struct clk *clk; >> struct regmap *regmap; >> + struct pwm_state suspend; >> + bool suspended; >> }; >> >> static inline struct stm32_pwm_lp *to_stm32_pwm_lp(struct pwm_chip *chip) >> @@ -223,6 +226,40 @@ static int stm32_pwm_lp_remove(struct platform_device *pdev) >> return pwmchip_remove(&priv->chip); >> } >> >> +#ifdef CONFIG_PM_SLEEP > > You might consider dropping ifdefs and marking pm functions with > __maybe_unused instead. In case CONFIG_PM_SLEEP=n then these two guys > will be removed and pm ops structure will be empty. Hi Tomasz, Thanks for this suggestion. I can do this change. I'll wait for more feedback from Uwe and Thierry before sending a v2 with that. > >> +static int stm32_pwm_lp_suspend(struct device *dev) >> +{ >> + struct stm32_pwm_lp *priv = dev_get_drvdata(dev); >> + > > I guess you first need to get platform_device from dev and eventually > stm32_pwm_lp. Wondering how this works now. This should be safe for now. This works because the probe routine calls: platform_set_drvdata(pdev, priv); And the underlying call is dev_set_drvdata() static inline void platform_set_drvdata(struct platform_device *pdev, void *data) { dev_set_drvdata(&pdev->dev, data); } > >> + pwm_get_state(&priv->chip.pwms[0], &priv->suspend); >> + priv->suspended = priv->suspend.enabled; >> + >> + /* safe to call pwm_disable() for already disabled pwm */ >> + pwm_disable(&priv->chip.pwms[0]); >> + >> + return pinctrl_pm_select_sleep_state(dev); >> +} >> + >> +static int stm32_pwm_lp_resume(struct device *dev) >> +{ >> + struct stm32_pwm_lp *priv = dev_get_drvdata(dev); >> + int ret; >> + >> + ret = pinctrl_pm_select_default_state(dev); >> + if (ret) >> + return ret; >> + >> + /* Only restore suspended pwm, not to disrupt other MFD child */ >> + if (!priv->suspended) >> + return 0; > > Would it make sense to use suspend.enabled directly? I propose to keep priv->suspended. Using 'suspend.enabled' directly would simply not work as the pwm_disable() call in stm32_pwm_lp_suspend() routine marks the 'suspend' state.enabled = false. That's why it's saved in the suspend routine, to be restored upon resume. Thanks for reviewing, Best regards, Fabrice > >> + >> + return pwm_apply_state(&priv->chip.pwms[0], &priv->suspend); >> +} >> +#endif >> + >> +static SIMPLE_DEV_PM_OPS(stm32_pwm_lp_pm_ops, stm32_pwm_lp_suspend, >> + stm32_pwm_lp_resume); >> + >> static const struct of_device_id stm32_pwm_lp_of_match[] = { >> { .compatible = "st,stm32-pwm-lp", }, >> {}, >> @@ -235,6 +272,7 @@ static int stm32_pwm_lp_remove(struct platform_device *pdev) >> .driver = { >> .name = "stm32-pwm-lp", >> .of_match_table = of_match_ptr(stm32_pwm_lp_of_match), >> + .pm = &stm32_pwm_lp_pm_ops, >> }, >> }; >> module_platform_driver(stm32_pwm_lp_driver); >> -- >> 1.9.1 >>
