[PATCH v2 15/16] iio: health: max30102: do not use internal iio_dev lock

Andy Shevchenko andy.shevchenko at gmail.com
Tue Oct 4 07:15:43 PDT 2022


On Tue, Oct 4, 2022 at 4:49 PM Nuno Sá <nuno.sa at analog.com> wrote:
>
> The pattern used in this device does not quite fit in the
> iio_device_claim_direct_mode() typical usage. In this case, we want to
> know if we are in buffered mode or not to know if the device is powered
> (buffer mode) or not. And depending on that max30102_get_temp() will
> power on the device if needed. Hence, in order to keep the same
> functionality, we try to:
>
> 1. Claim Buffered mode;
> 2: If 1) succeeds call max30102_get_temp() without powering on the
>    device;
> 3: Release Buffered mode;
> 4: If 1) fails, Claim Direct mode;
> 5: If 4) succeeds call max30102_get_temp() with powering on the device;
> 6: Release Direct mode;
> 7: If 4) fails, goto to 1) and try again.
>
> This dance between buffered and direct mode is not particularly pretty
> (as well as the loop introduced by the goto statement) but it does allow
> us to get rid of the mlock usage while keeping the same behavior.

...

> +               if (iio_device_claim_buffer_mode(indio_dev)) {

Why is ret not used here?

> +                       /*
> +                        * This one is a *bit* hacky. If we cannot claim buffer
> +                        * mode, then try direct mode so that we make sure
> +                        * things cannot concurrently change. And we just keep
> +                        * trying until we get one of the modes...
> +                        */
> +                       if (iio_device_claim_direct_mode(indio_dev))

...and here?

> +                               goto any_mode_retry;

> +               } else {

> +               }

I.o.w. what error code will be visible to the caller and why.

-- 
With Best Regards,
Andy Shevchenko



More information about the linux-amlogic mailing list