On Mon, 2018-02-12 at 14:16 +0100, Linus Walleij wrote:

As we augmented the regulator core to accept a GPIO descriptor instead
of a GPIO number, we can augment the fixed GPIO regulator to look up
and pass that descriptor directly from device tree or board GPIO
descriptor look up tables.

Some boards just auto-enumerate their fixed regulator platform devices
and I have assumed they get names like "fixed-regulator.0" but it's
pretty hard to guess this. I need some testing from board maintainers
to
be sure. Other boards are straight forward, using just plain
"fixed-regulator" (ID -1) or "fixed-regulator.1" hammering down the
device ID.

The OMAP didn't have proper label names on its GPIO chips so I have
fixed
this with a separate patch to the GPIO tree.

It seems the da9055 and da9211 has never got around to actually
passing
any enable gpio into its platform data (not the in-tree code anyway)
so we
can just decide to simply pass a descriptor instead.

The fixed GPIO-controlled regulator in mach-pxa/ezx.c was confusingly
named
"*_dummy_supply_device" while it is a very real device backed by a
GPIO
line. There is nothing dummy about it at all, so I renamed it with the
infix *_regulator_* as part of this patch set.

For the patch hunk hitting arch/blackfin I would say I do not expect
testing, review or ACKs anymore so if it works, it works.

The hunk hitting the x86 BCM43xx driver is especially tricky as the
number
comes out of SFI which is a mystery to me. I definately need someone
to
look at this. (Hi Andy.)

Hi, Linus!

Nice patch, though I have comments below.

I'm not sure this will be always like this.
We have DEVID_AUTO, which theoretically can be anything.

Okay, it looks like we have only one static regulator for now for Intel
MID. Though it's fragile if anything will change in the future (quite
unlikely).

+	.table	= {

+		/* CHECKME: is this the correct PCI address for the
GPIO controller? */

Yes.

+		GPIO_LOOKUP("0000:00:0c.0", -1, "enable",
GPIO_ACTIVE_LOW),

+		{ },

Terminator should terminate even at compile time, right?

Simple

{}

looks better to me.

+	},
+};
+
 static int __init bcm43xx_regulator_register(void)
 {
+	struct gpiod_lookup_table *table = &bcm43xx_vmmc_gpio_table;
+	struct gpiod_lookup *lookup = table->table;
 	int ret;
 
-	bcm43xx_vmmc.gpio =
get_gpio_by_name(WLAN_SFI_GPIO_ENABLE_NAME);

+	/* FIXME: convert SFI layer to use GPIO descriptors
internally */

We already discussed this few years ago and decide not to do anything
WRT SFI. It should just die.

+	lookup[0].chip_hwnum =
get_gpio_by_name(WLAN_SFI_GPIO_ENABLE_NAME);
+	gpiod_add_lookup_table(table);

 	cfg.ena_gpio_invert = !config->enable_high;

 	if (config->enabled_at_boot) {
 		if (config->enable_high)
-			cfg.ena_gpio_flags |= GPIOF_OUT_INIT_HIGH;
+			gflags = GPIOD_OUT_HIGH;
 		else
-			cfg.ena_gpio_flags |= GPIOF_OUT_INIT_LOW;
+			gflags = GPIOD_OUT_LOW;
 	} else {
 		if (config->enable_high)
-			cfg.ena_gpio_flags |= GPIOF_OUT_INIT_LOW;
+			gflags = GPIOD_OUT_LOW;
 		else
-			cfg.ena_gpio_flags |= GPIOF_OUT_INIT_HIGH;
+			gflags = GPIOD_OUT_HIGH;
 	}

Just a side note: It might make sense to split this to some kind of
generic helper, like:

static inline enum gpiod_flags gpiod_flags_output(bool value, bool
invert)
{
...
}

+	if (config->gpio_is_open_drain) {
+		if (gflags == GPIOD_OUT_HIGH)
+			gflags = GPIOD_OUT_HIGH_OPEN_DRAIN;
+		else
+			gflags = GPIOD_OUT_LOW_OPEN_DRAIN;
+	}
+

+	cfg.ena_gpiod = devm_gpiod_get_optional(&pdev->dev, NULL,
gflags);

Shouldn't we be a little bit more stricter here, i.e. require "enable"
name?

+	if (IS_ERR(cfg.ena_gpiod))
+		return PTR_ERR(cfg.ena_gpiod);

-- 
Andy Shevchenko [off-list ref]
Intel Finland Oy