The PWM core does not appear to respect the settings in the pwm_lookup table in the ev3dev projects implementation of a board file. Root cause is that pwm_get() does not save the pwm_lookup entry for the best matching device, and therefore ends up using period and polarity from a pointer that's past the end of the table.

These values are often 0, which translates to PWM_POLARITY_NORMAL, but it can be anything really...

First patch ever submitted - fun times finding this one - the bug was in the last place I looked
-----------------------------------------------------------------------

Make the match loop save the best match so that period and polarity can be set correctly

-----------------------------------------------------------------------


diff --git a/drivers/pwm/core.c b/drivers/pwm/core.c
index 4b66bf0..f244e2c 100644
--- a/drivers/pwm/core.c
+++ b/drivers/pwm/core.c
@@ -605,6 +605,7 @@ struct pwm_device *pwm_get(struct device *dev, const char *con_id)
        unsigned int index = 0;
        unsigned int best = 0;
        struct pwm_lookup *p;
+       struct pwm_lookup *pl = NULL;
        unsigned int match;

        /* look up via DT first */
@@ -651,6 +652,7 @@ struct pwm_device *pwm_get(struct device *dev, const char *con_id)
                }

                if (match > best) {
+                       pl = p;
                        chip = pwmchip_find_by_name(p->provider);
                        index = p->index;

@@ -668,8 +670,8 @@ struct pwm_device *pwm_get(struct device *dev, const char *con_id)
        if (IS_ERR(pwm))
                return pwm;

-       pwm_set_period(pwm, p->period);
-       pwm_set_polarity(pwm, p->polarity);
+       pwm_set_period(pwm, pl->period);
+       pwm_set_polarity(pwm, pl->polarity);


        return pwm;

Ralph Hempel (rhem...@hempeldesigngroup.com)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Reply via email to