Message ID | 20170921010421.7467-3-f.fainelli@gmail.com |
---|---|
State | Changes Requested, archived |
Headers | show |
Series | pinctrl: Allow indicating loss of state across suspend/resume | expand |
On Thu, Sep 21, 2017 at 3:04 AM, Florian Fainelli <f.fainelli@gmail.com> wrote: > Some platforms (e.g: Broadcom STB: BMIPS_GENERIC/ARCH_BRCMSTB) will lose > their register contents when entering their lower power state. In such a > case, the pinctrl-single driver that is used will not be able to restore > the power states without telling the core about it and having > pinctrl_select_state() check for that. > > This patch adds a new optional boolean property that Device Tree can > define in order to obtain exactly that and having the core pinctrl code > take that into account. > > Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> If we use this per-controller approach rather than the per-state approach I discuss in reply to patch 1/2, we should probably make it a generic property for pin controllers and not just a pinctrl-single business. So patch pinctrl-bindings.txt and put the code somewhere in core. But that is more of a detail, first we need to figure out how to handle this business in general. Yours, Linus Walleij -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 09/22/2017 06:03 AM, Linus Walleij wrote: > On Thu, Sep 21, 2017 at 3:04 AM, Florian Fainelli <f.fainelli@gmail.com> wrote: > >> Some platforms (e.g: Broadcom STB: BMIPS_GENERIC/ARCH_BRCMSTB) will lose >> their register contents when entering their lower power state. In such a >> case, the pinctrl-single driver that is used will not be able to restore >> the power states without telling the core about it and having >> pinctrl_select_state() check for that. >> >> This patch adds a new optional boolean property that Device Tree can >> define in order to obtain exactly that and having the core pinctrl code >> take that into account. >> >> Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> > > If we use this per-controller approach rather than the per-state approach > I discuss in reply to patch 1/2, we should probably make it a generic > property for pin controllers and not just a pinctrl-single business. I suppose it makes sense to make this a generic pinctrl property then. drivers/pinctrl/core.c does not appear to be trying to fetch any properties for a pinctrl device, but that is probably not too hard to add. > > So patch pinctrl-bindings.txt and put the code somewhere in > core. > > But that is more of a detail, first we need to figure out how to > handle this business in general. Fair enough.
diff --git a/Documentation/devicetree/bindings/pinctrl/pinctrl-single.txt b/Documentation/devicetree/bindings/pinctrl/pinctrl-single.txt index e705acd3612c..e71967f6a1a7 100644 --- a/Documentation/devicetree/bindings/pinctrl/pinctrl-single.txt +++ b/Documentation/devicetree/bindings/pinctrl/pinctrl-single.txt @@ -82,6 +82,10 @@ Optional properties: /* pin base, nr pins & gpio function */ pinctrl-single,gpio-range = <&range 0 3 0 &range 3 9 1>; +- pinctrl-single,low-power-state-loss : indicates that the pins lose their + state during low power modes and therefore need to be restored upon + system resumption. + - interrupt-controller : standard interrupt controller binding if using interrupts for wake-up events for example. In this case pinctrl-single is set up as a chained interrupt controller and the wake-up interrupts diff --git a/drivers/pinctrl/pinctrl-single.c b/drivers/pinctrl/pinctrl-single.c index b8b3d932cd73..d69d20b8247a 100644 --- a/drivers/pinctrl/pinctrl-single.c +++ b/drivers/pinctrl/pinctrl-single.c @@ -1749,6 +1749,9 @@ static int pcs_probe(struct platform_device *pdev) goto free; } + if (of_property_read_bool(np, "pinctrl-single,low-power-state-loss")) + pcs->pctl->flags |= PINCTRL_FLG_FORCE_STATE; + ret = pcs_add_gpio_func(np, pcs); if (ret < 0) goto free;
Some platforms (e.g: Broadcom STB: BMIPS_GENERIC/ARCH_BRCMSTB) will lose their register contents when entering their lower power state. In such a case, the pinctrl-single driver that is used will not be able to restore the power states without telling the core about it and having pinctrl_select_state() check for that. This patch adds a new optional boolean property that Device Tree can define in order to obtain exactly that and having the core pinctrl code take that into account. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> --- Documentation/devicetree/bindings/pinctrl/pinctrl-single.txt | 4 ++++ drivers/pinctrl/pinctrl-single.c | 3 +++ 2 files changed, 7 insertions(+)