This fixes an overflow condition that can happen with high max
brightness and period values in compute_duty_cycle. This fixes it by
using a 64 bit variable for computing the duty cycle.
Signed-off-by: Derek Basehore <[email protected]>
---
drivers/video/backlight/pwm_bl.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/drivers/video/backlight/pwm_bl.c b/drivers/video/backlight/pwm_bl.c
index 76311ec5e400..5ccc8e34b7a6 100644
--- a/drivers/video/backlight/pwm_bl.c
+++ b/drivers/video/backlight/pwm_bl.c
@@ -88,14 +88,17 @@ static void pwm_backlight_power_off(struct pwm_bl_data *pb)
static int compute_duty_cycle(struct pwm_bl_data *pb, int brightness)
{
unsigned int lth = pb->lth_brightness;
- int duty_cycle;
+ u64 duty_cycle;
if (pb->levels)
duty_cycle = pb->levels[brightness];
else
duty_cycle = brightness;
- return (duty_cycle * (pb->period - lth) / pb->scale) + lth;
+ duty_cycle *= pb->period - lth;
+ do_div(duty_cycle, pb->scale);
+
+ return duty_cycle + lth;
}
static int pwm_backlight_update_status(struct backlight_device *bl)
--
2.14.1.342.g6490525c54-goog
On Tue, Aug 29, 2017 at 01:34:34PM -0700, Derek Basehore wrote:
> This fixes an overflow condition that can happen with high max
> brightness and period values in compute_duty_cycle. This fixes it by
> using a 64 bit variable for computing the duty cycle.
>
> Signed-off-by: Derek Basehore <[email protected]>
> ---
I believe this is the correct v2 patch, and I agree that Derek is
confused by his mail client :)
Reviewed-by: Brian Norris <[email protected]>
> drivers/video/backlight/pwm_bl.c | 7 +++++--
> 1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/video/backlight/pwm_bl.c b/drivers/video/backlight/pwm_bl.c
> index 76311ec5e400..5ccc8e34b7a6 100644
> --- a/drivers/video/backlight/pwm_bl.c
> +++ b/drivers/video/backlight/pwm_bl.c
> @@ -88,14 +88,17 @@ static void pwm_backlight_power_off(struct pwm_bl_data *pb)
> static int compute_duty_cycle(struct pwm_bl_data *pb, int brightness)
> {
> unsigned int lth = pb->lth_brightness;
> - int duty_cycle;
> + u64 duty_cycle;
>
> if (pb->levels)
> duty_cycle = pb->levels[brightness];
> else
> duty_cycle = brightness;
>
> - return (duty_cycle * (pb->period - lth) / pb->scale) + lth;
> + duty_cycle *= pb->period - lth;
> + do_div(duty_cycle, pb->scale);
> +
> + return duty_cycle + lth;
> }
>
> static int pwm_backlight_update_status(struct backlight_device *bl)
> --
> 2.14.1.342.g6490525c54-goog
>
On Tue, Aug 29, 2017 at 01:34:34PM -0700, Derek Basehore wrote:
> This fixes an overflow condition that can happen with high max
> brightness and period values in compute_duty_cycle. This fixes it by
> using a 64 bit variable for computing the duty cycle.
>
> Signed-off-by: Derek Basehore <[email protected]>
> ---
> drivers/video/backlight/pwm_bl.c | 7 +++++--
> 1 file changed, 5 insertions(+), 2 deletions(-)
I'm still confused by the brightness-levels-scale comment in the other
thread, but this looks like a good fix for a potential overflow so:
Acked-by: Thierry Reding <[email protected]>
On Wed, Aug 30, 2017 at 04:27:41PM +0200, Thierry Reding wrote:
> I'm still confused by the brightness-levels-scale comment in the other
> thread,
I believe Derek was accidentally referring to a change that's only in
our downstream tree.
> but this looks like a good fix for a potential overflow so:
Right, it's still an applicable overflow without our downstream changes.
Brian
> Acked-by: Thierry Reding <[email protected]>
On Tue, 29 Aug 2017, Derek Basehore wrote:
> This fixes an overflow condition that can happen with high max
> brightness and period values in compute_duty_cycle. This fixes it by
> using a 64 bit variable for computing the duty cycle.
>
> Signed-off-by: Derek Basehore <[email protected]>
> ---
> drivers/video/backlight/pwm_bl.c | 7 +++++--
> 1 file changed, 5 insertions(+), 2 deletions(-)
Applied for v4.15, thanks.
--
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog