[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CALvyBcV8b_AiKK-QBRU1sphrWjgATecTyd7c27Y-gNtJv1hV8A@mail.gmail.com>
Date: Thu, 4 Jul 2024 13:52:54 -0400
From: Jack Chen <zenghuchen@...gle.com>
To: Lee Jones <lee@...nel.org>
Cc: Pavel Machek <pavel@....cz>, Mark Brown <broonie@...nel.org>,
Vadim Pasternak <vadimp@...dia.com>, Randy Dunlap <rdunlap@...radead.org>, linux-kernel@...r.kernel.org,
linux-leds@...r.kernel.org
Subject: Re: [PATCH] leds:lm3601x:calculate max_brightness and brightness properly
Many thanks to Lee for the quick response. I will submit a V2 patch later.
> > - ret = regmap_write(led->regmap, LM3601X_LED_TORCH_REG, brightness);
> > + ret = regmap_write(led->regmap, LM3601X_LED_TORCH_REG, brightness - 1);
> Why is there now a need to start adding/subtracting 1s to make the maths work?
This is because lm3601x torch brightness register takes 0 as the
minimum output current (2.4mA). But led_brightness uses 0 as LED_OFF.
So we need to add -1 offset to avoid blocking users who want to set
the torch output minimum.
On Thu, Jul 4, 2024 at 12:50 PM Lee Jones <lee@...nel.org> wrote:
>
> Subject line needs fixing.
>
> `git log --oneline -- drivers/<subsystem>` is your friend.
>
> > 1) check the range of torch_current_max,
> > 2) calcualtes max_brightness precisely,
> > 3) lm3601x torch brightness register starts from 0 (2.4 mA).
>
> Please write this out as a proper paragraph.
>
> This isn't really a numbered list type situation.
>
> > Tested: tested with a lm36011 and it can set its brightness to lowest
> > value (2.4 mA)
>
> What is the Tested: trailer? Again, please write this out properly.
>
> > Signed-off-by: Jack Chen <zenghuchen@...gle.com>
> > ---
> > drivers/leds/flash/leds-lm3601x.c | 15 ++++++++++++---
> > 1 file changed, 12 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/leds/flash/leds-lm3601x.c b/drivers/leds/flash/leds-lm3601x.c
> > index 7e93c447fec5..fc4df904ea90 100644
> > --- a/drivers/leds/flash/leds-lm3601x.c
> > +++ b/drivers/leds/flash/leds-lm3601x.c
> > @@ -190,7 +190,7 @@ static int lm3601x_brightness_set(struct led_classdev *cdev,
> > goto out;
> > }
> >
> > - ret = regmap_write(led->regmap, LM3601X_LED_TORCH_REG, brightness);
> > + ret = regmap_write(led->regmap, LM3601X_LED_TORCH_REG, brightness - 1);
>
> Why is there now a need to start adding/subtracting 1s to make the maths work?
>
> > if (ret < 0)
> > goto out;
> >
> > @@ -341,8 +341,9 @@ static int lm3601x_register_leds(struct lm3601x_led *led,
> >
> > led_cdev = &led->fled_cdev.led_cdev;
> > led_cdev->brightness_set_blocking = lm3601x_brightness_set;
> > - led_cdev->max_brightness = DIV_ROUND_UP(led->torch_current_max,
> > - LM3601X_TORCH_REG_DIV);
> > + led_cdev->max_brightness = DIV_ROUND_UP(
>
> This is no place to break a line.
>
> Break after the '=' and wrap at 100-chars instead.
>
> > + led->torch_current_max - LM3601X_MIN_TORCH_I_UA + 1,
> > + LM3601X_TORCH_REG_DIV);
> > led_cdev->flags |= LED_DEV_CAP_FLASH;
> >
> > init_data.fwnode = fwnode;
> > @@ -386,6 +387,14 @@ static int lm3601x_parse_node(struct lm3601x_led *led,
> > goto out_err;
> > }
> >
> > + if (led->torch_current_max > LM3601X_MAX_TORCH_I_UA) {
> > + dev_warn(&led->client->dev,
> > + "led-max-microamp cannot be higher than %d\n",
> > + LM3601X_MAX_TORCH_I_UA);
>
> "Max torch current set too high (%d vs %d)"
>
> > + led->torch_current_max = LM3601X_MAX_TORCH_I_UA;
> > + }
> > +
> > +
>
> 2 lines?
>
> > ret = fwnode_property_read_u32(child, "flash-max-microamp",
> > &led->flash_current_max);
> > if (ret) {
> > --
> > 2.45.2.803.g4e1b14247a-goog
> >
>
> --
> Lee Jones [李琼斯]
Powered by blists - more mailing lists