Re: [PATCH] Input: s6sy761 - fix error handling in power management paths

[David Heidelberg]

On 30/09/2026 07:18, Dmitry Torokhov wrote:
> In s6sy761_power_on(), if any of the I2C operations or status checks
> fail after enabling the regulators, the function returns an error while
> leaving the regulators enabled. Conversely, in s6sy761_resume(), if
> powering on the controller fails, enable_irq() is skipped while the PM
> core still marks the device as resumed. A subsequent system suspend or
> driver unbind then invokes s6sy761_power_off() again, attempting to
> disable already disabled regulators and incrementing the IRQ disable
> depth a second time, which leaves the interrupt permanently disabled on
> later resumes. Additionally, if restoring sensing in s6sy761_resume()
> fails, the controller is left powered on.
>
> Track whether the controller is powered up with a powered_up flag,
> unwind regulator enablement on error in s6sy761_power_on(), move
> disable_irq() to s6sy761_suspend() so it is only called during system
> sleep when the controller is powered up, and power off the controller
> if restoring sensing fails during resume.
>
> Reported-by: sashiko-bot@xxxxxxxxxx
> Assisted-by: LLM
> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@xxxxxxxxx>
> ---
>   drivers/input/touchscreen/s6sy761.c | 59 +++++++++++++++++++++--------
>   1 file changed, 43 insertions(+), 16 deletions(-)

+<<<<<<< HEAD
        return s6sy761_power_on(sdata);
+=======
+       return 0;
+
+err_power_off:
+       s6sy761_power_off(sdata);
+       return err;
+>>>>>>> Input: s6sy761 - fix error handling in power management paths

doesn't apply cleanly on linux-next, against which branch was it build? (I don't 
see anything touch it in input)

Thanks
David