Hi Pali,
On Tue, Sep 06, 2022 at 11:16:28PM +0200, Pali Rohár wrote:
> Hello!
>
> On Tuesday 06 September 2022 13:43:01 Dmitry Torokhov wrote:
> > This patch switches the driver away from legacy gpio/of_gpio API to
> > gpiod API, and removes use of of_get_named_gpio_flags() which I want to
> > make private to gpiolib.
>
> There are many pending pci-mvebu.c patches waiting for review and merge,
> so I would suggest to wait until all other mvebu patches are processed
> and then process this one... longer waiting period :-(
OK, it is not super urgent. OTOH it is a very simple patch :)
>
> > Signed-off-by: Dmitry Torokhov <dmitry.torokhov@xxxxxxxxx>
> > ---
> > drivers/pci/controller/pci-mvebu.c | 48 +++++++++---------------------
> > 1 file changed, 14 insertions(+), 34 deletions(-)
> >
> > diff --git a/drivers/pci/controller/pci-mvebu.c b/drivers/pci/controller/pci-mvebu.c
> > index 1ced73726a26..a54beb8f611c 100644
> > --- a/drivers/pci/controller/pci-mvebu.c
> > +++ b/drivers/pci/controller/pci-mvebu.c
> > @@ -11,14 +11,13 @@
> > #include <linux/bitfield.h>
> > #include <linux/clk.h>
> > #include <linux/delay.h>
> > -#include <linux/gpio.h>
> > +#include <linux/gpio/consumer.h>
> > #include <linux/init.h>
> > #include <linux/mbus.h>
> > #include <linux/slab.h>
> > #include <linux/platform_device.h>
> > #include <linux/of_address.h>
> > #include <linux/of_irq.h>
> > -#include <linux/of_gpio.h>
> > #include <linux/of_pci.h>
> > #include <linux/of_platform.h>
> >
> > @@ -1261,9 +1260,8 @@ static int mvebu_pcie_parse_port(struct mvebu_pcie *pcie,
> > struct mvebu_pcie_port *port, struct device_node *child)
> > {
> > struct device *dev = &pcie->pdev->dev;
> > - enum of_gpio_flags flags;
> > u32 slot_power_limit;
> > - int reset_gpio, ret;
> > + int ret;
> > u32 num_lanes;
> >
> > port->pcie = pcie;
> > @@ -1327,40 +1325,22 @@ static int mvebu_pcie_parse_port(struct mvebu_pcie *pcie,
> > port->name, child);
> > }
> >
> > - reset_gpio = of_get_named_gpio_flags(child, "reset-gpios", 0, &flags);
> > - if (reset_gpio == -EPROBE_DEFER) {
> > - ret = reset_gpio;
> > + port->reset_name = devm_kasprintf(dev, GFP_KERNEL, "%s-reset",
> > + port->name);
> > + if (!port->reset_name) {
> > + ret = -ENOMEM;
> > goto err;
> > }
> >
> > - if (gpio_is_valid(reset_gpio)) {
> > - unsigned long gpio_flags;
> > -
> > - port->reset_name = devm_kasprintf(dev, GFP_KERNEL, "%s-reset",
> > - port->name);
> > - if (!port->reset_name) {
> > - ret = -ENOMEM;
> > + port->reset_gpio = devm_fwnode_gpiod_get(dev, of_fwnode_handle(child),
> > + "reset", GPIOD_OUT_HIGH,
>
> What does it mean that there is a new GPIOD_OUT_HIGH flag passed to the
> devm_fwnode_gpiod_get() function?
This means that we drive the line as "active" as soon as we successfully
grab GPIO. This is the same as we had with devm_gpio_request_one(), but
we do not need to figure out actual polarity.
>
> > + port->name);
> > + ret = PTR_ERR_OR_ZERO(port->reset_gpio);
> > + if (ret) {
> > + if (ret != -ENOENT)
> > goto err;
> > - }
> > -
> > - if (flags & OF_GPIO_ACTIVE_LOW) {
> > - dev_info(dev, "%pOF: reset gpio is active low\n",
> > - child);
> > - gpio_flags = GPIOF_ACTIVE_LOW |
> > - GPIOF_OUT_INIT_LOW;
> > - } else {
> > - gpio_flags = GPIOF_OUT_INIT_HIGH;
> > - }
> > -
> > - ret = devm_gpio_request_one(dev, reset_gpio, gpio_flags,
> > - port->reset_name);
> > - if (ret) {
> > - if (ret == -EPROBE_DEFER)
> > - goto err;
> > - goto skip;
> > - }
> > -
> > - port->reset_gpio = gpio_to_desc(reset_gpio);
> > + /* reset gpio is optional */
> > + port->reset_gpio = NULL;
>
> Maybe you can also release port->reset_name as it is not used at this
> stage?
OK, I figured it was just a few bytes, but sure, I'll add devm_kfree().
Thanks for the review.
--
Dmitry
