The KingFisher board has regulators. They just need to be en-/disabled,
so we can leave the handling to devm.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
---

Changes since RFC:
* add 12v regulator
* add comment about the order of enabling the regulators
* use a for-loop to iterate over the regulators

Sidenote: I tried to introduce 'devm_regulator_bulk_get_enable_optional'
to avoid the for-loop but that was a too intrusive change because all of
the regulator_bulk logic is designed to fail if something bad happens
somewhere.

 drivers/pci/controller/pcie-rcar-host.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/controller/pcie-rcar-host.c b/drivers/pci/controller/pcie-rcar-host.c
index e80e56b2a842..e86bf0f7729b 100644
--- a/drivers/pci/controller/pcie-rcar-host.c
+++ b/drivers/pci/controller/pcie-rcar-host.c
@@ -29,6 +29,7 @@
 #include <linux/phy/phy.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
+#include <linux/regulator/consumer.h>
 
 #include "pcie-rcar.h"
 
@@ -974,13 +975,18 @@ static const struct of_device_id rcar_pcie_of_match[] = {
 	{},
 };
 
+/* Design note 346 from Linear Technology says order is not important */
+static const char * const rcar_pcie_supplies[] = {
+	"vpcie12v", "vpcie3v3", "vpcie1v5"
+};
+
 static int rcar_pcie_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct rcar_pcie_host *host;
 	struct rcar_pcie *pcie;
 	u32 data;
-	int err;
+	int i, err;
 	struct pci_host_bridge *bridge;
 
 	bridge = devm_pci_alloc_host_bridge(dev, sizeof(*host));
@@ -992,6 +998,13 @@ static int rcar_pcie_probe(struct platform_device *pdev)
 	pcie->dev = dev;
 	platform_set_drvdata(pdev, host);
 
+	for (i = 0; i < ARRAY_SIZE(rcar_pcie_supplies); i++) {
+		err = devm_regulator_get_enable_optional(dev, rcar_pcie_supplies[i]);
+		if (err < 0 && err != -ENODEV)
+			dev_err_probe(dev, err, "error enabling regulator %s\n",
+				      rcar_pcie_supplies[i]);
+	}
+
 	pm_runtime_enable(pcie->dev);
 	err = pm_runtime_get_sync(pcie->dev);
 	if (err < 0) {
-- 
2.30.2

Hi Wolfram,

On Wed, May 10, 2023 at 8:59 AM Wolfram Sang
<wsa+renesas@sang-engineering.com> wrote:
> The KingFisher board has regulators. They just need to be en-/disabled,
> so we can leave the handling to devm.
>
> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
> ---
>
> Changes since RFC:
> * add 12v regulator
> * add comment about the order of enabling the regulators
> * use a for-loop to iterate over the regulators

Thanks for the update!

> --- a/drivers/pci/controller/pcie-rcar-host.c
> +++ b/drivers/pci/controller/pcie-rcar-host.c
> @@ -29,6 +29,7 @@
>  #include <linux/phy/phy.h>
>  #include <linux/platform_device.h>
>  #include <linux/pm_runtime.h>
> +#include <linux/regulator/consumer.h>
>
>  #include "pcie-rcar.h"
>
> @@ -974,13 +975,18 @@ static const struct of_device_id rcar_pcie_of_match[] = {
>         {},
>  };
>
> +/* Design note 346 from Linear Technology says order is not important */
> +static const char * const rcar_pcie_supplies[] = {
> +       "vpcie12v", "vpcie3v3", "vpcie1v5"
> +};
> +
>  static int rcar_pcie_probe(struct platform_device *pdev)
>  {
>         struct device *dev = &pdev->dev;
>         struct rcar_pcie_host *host;
>         struct rcar_pcie *pcie;
>         u32 data;
> -       int err;
> +       int i, err;

unsigned int i?

>         struct pci_host_bridge *bridge;

The (lack of) reverse-Xmas-tree ordering is hurting my OCD, but that's
not your fault...

> @@ -992,6 +998,13 @@ static int rcar_pcie_probe(struct platform_device *pdev)
>         pcie->dev = dev;
>         platform_set_drvdata(pdev, host);
>
> +       for (i = 0; i < ARRAY_SIZE(rcar_pcie_supplies); i++) {
> +               err = devm_regulator_get_enable_optional(dev, rcar_pcie_supplies[i]);
> +               if (err < 0 && err != -ENODEV)
> +                       dev_err_probe(dev, err, "error enabling regulator %s\n",
> +                                     rcar_pcie_supplies[i]);

Shouldn't this return, and propagate the error code upstream?

> +       }
> +
>         pm_runtime_enable(pcie->dev);
>         err = pm_runtime_get_sync(pcie->dev);
>         if (err < 0) {

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

On Wed, May 10, 2023 at 09:27:46AM +0200, Geert Uytterhoeven wrote:
> On Wed, May 10, 2023 at 8:59 AM Wolfram Sang
> <wsa+renesas@sang-engineering.com> wrote:
> ...

> >  static int rcar_pcie_probe(struct platform_device *pdev)
> >  {
> >         struct device *dev = &pdev->dev;
> >         struct rcar_pcie_host *host;
> >         struct rcar_pcie *pcie;
> >         u32 data;
> > -       int err;
> > +       int i, err;
> 
> unsigned int i?
> 
> >         struct pci_host_bridge *bridge;
> 
> The (lack of) reverse-Xmas-tree ordering is hurting my OCD, but that's
> not your fault...

I usually put things in order of use, with initializations from
parameters first.  Happily, that is often a pretty good approximation
of reverse-Xmas-tree, as it is here, so I'm all in favor of moving
"struct pci_host_bridge" up there :)

Bjorn

> > +       int i, err;
> 
> unsigned int i?

OK.

> 
> >         struct pci_host_bridge *bridge;
> 
> The (lack of) reverse-Xmas-tree ordering is hurting my OCD, but that's
> not your fault...

Ack :) I can change it, though.

> > +                       dev_err_probe(dev, err, "error enabling regulator %s\n",
> > +                                     rcar_pcie_supplies[i]);
> 
> Shouldn't this return, and propagate the error code upstream?

Ouch, brown paper bag, please.