On 2018-08-27 20:24, Bjorn Andersson wrote:

On Fri 24 Aug 03:27 PDT 2018, Kiran Gunda wrote:

quoted

diff --git a/drivers/video/backlight/qcom-wled.c 

b/drivers/video/backlight/qcom-wled.c

[..]

quoted

 struct wled {

@@ -135,16 +146,22 @@ struct wled {
 	struct regmap *regmap;
 	struct mutex lock;	/* Lock to avoid race from thread irq handler */
 	ktime_t last_short_event;
+	ktime_t start_ovp_fault_time;
 	u16 ctrl_addr;
 	u16 sink_addr;
 	u16 max_string_count;
+	u16 auto_detection_ovp_count;
 	u32 brightness;
 	u32 max_brightness;
 	u32 short_count;
+	u32 auto_detect_count;
 	bool disabled_by_short;
 	bool has_short_detect;
+	int ovp_irq;
+	bool ovp_irq_disabled;

ovp_irq_disabled is now only assigned, never read. So you should be 
able
to drop it.

Ok. will address in next series.

[..]

quoted

@@ -200,7 +226,18 @@ static int wled_module_enable(struct wled *wled, 

int val)
 				WLED_CTRL_REG_MOD_EN,
 				WLED_CTRL_REG_MOD_EN_MASK,
 				val << WLED_CTRL_REG_MOD_EN_SHIFT);
-	return rc;
+	if (rc < 0)
+		return rc;
+
+	if (val) {
+		schedule_delayed_work(&wled->ovp_work, HZ / 100);

wled_ovp_work() is a nop when ovp_irq = 0, so wrap the entire if/else
in a check for ovp_irq > 0 (rather than include it in the else) and 
drop
it from the worker.

Will address in next series.

quoted

+	} else {
+		if (!cancel_delayed_work_sync(&wled->ovp_work) &&
+		    wled->ovp_irq > 0)
+			disable_irq(wled->ovp_irq);
+	}
+
+	return 0;

Also, if a user calls rmmod within 10ms of enabling module we will
schedule work of code that's no longer available in the kernel. So you
need to also cancel_delayed_work_sync() in the remove function of the
driver.

Will address in next series.

Apart from this I think this patch looks good now!

Thank you

Regards,
Bjorn