Skip to contentSkip to conversation

extcon: int3496: process id-pin first so that we start with the right status

PATCH Applied Hans de Goede 2 participants 6 messages Jan 25, 2018 – Feb 6, 2018 External connector subsystem (extcon)

v1 of 4 Applied Latest is v4 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 extcon: int3496: process id-pin first so that we start with the right status

Some other drivers may be waiting for our extcon to show-up (exiting their probe methods with -EPROBE_DEFER until we show up).

These drivers will typically get the cable state directly after getting the extcon, this commit changes the int3496 code to process the id-pin before registering the extcon, so that other drivers see the correct state right away.

---
 drivers/extcon/extcon-intel-int3496.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/extcon/extcon-intel-int3496.c +4 -3
index c8691b5a9cb0..b23ee9d993a3 100644
@@ -131,6 +131,10 @@ static int int3496_probe(struct platform_device *pdev)
 	if (IS_ERR(data->gpio_usb_mux))
 		dev_info(dev, "can't request USB MUX GPIO\n");
 
+	/* process id-pin first so that we start with the right status */
+	queue_delayed_work(system_wq, &data->work, 0);
+	flush_delayed_work(&data->work);
+
 	/* register extcon device */
 	data->edev = devm_extcon_dev_allocate(dev, int3496_cable);
 	if (IS_ERR(data->edev))
@@ -153,9 +157,6 @@ static int int3496_probe(struct platform_device *pdev)
 		return ret;
 	}
 
-	/* queue initial processing of id-pin */
-	queue_delayed_work(system_wq, &data->work, 0);
-
 	platform_set_drvdata(pdev, data);
 
 	return 0;
Signature · 1 line
2.14.3
maintainer Applied
#1 lore raw
Quoted 30 lines · Hans de Goede

Some other drivers may be waiting for our extcon to show-up (exiting their probe methods with -EPROBE_DEFER until we show up).

These drivers will typically get the cable state directly after getting the extcon, this commit changes the int3496 code to process the id-pin before registering the extcon, so that other drivers see the correct state right away.

---
 drivers/extcon/extcon-intel-int3496.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/extcon/extcon-intel-int3496.c +4 -0
index c8691b5a9cb0..b23ee9d993a3 100644
@@ -131,6 +131,10 @@ static int int3496_probe(struct platform_device *pdev)
 	if (IS_ERR(data->gpio_usb_mux))
 		dev_info(dev, "can't request USB MUX GPIO\n");
 
+	/* process id-pin first so that we start with the right status */
+	queue_delayed_work(system_wq, &data->work, 0);
+	flush_delayed_work(&data->work);
+
 	/* register extcon device */
 	data->edev = devm_extcon_dev_allocate(dev, int3496_cable);
 	if (IS_ERR(data->edev))

drivers/extcon/extcon-intel-int3496.c static int int3496_probe(struct platform_device *pdev) · line 162

Quoted 9 lines
drivers/extcon/extcon-intel-int3496.c +0 -3
@@ -153,9 +157,6 @@ static int int3496_probe(struct platform_device *pdev)
 		return ret;
 	}
 
-	/* queue initial processing of id-pin */
-	queue_delayed_work(system_wq, &data->work, 0);
-
 	platform_set_drvdata(pdev, data);
 
 	return 0;

Applied it. Thanks.

Signature · 3 lines
Best Regards,
Chanwoo Choi
Samsung Electronics
maintainer
#2 lore raw
In reply to Hans de Goede
Quoted 30 lines · Hans de Goede

Some other drivers may be waiting for our extcon to show-up (exiting their probe methods with -EPROBE_DEFER until we show up).

These drivers will typically get the cable state directly after getting the extcon, this commit changes the int3496 code to process the id-pin before registering the extcon, so that other drivers see the correct state right away.

---
 drivers/extcon/extcon-intel-int3496.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/extcon/extcon-intel-int3496.c +4 -0
index c8691b5a9cb0..b23ee9d993a3 100644
@@ -131,6 +131,10 @@ static int int3496_probe(struct platform_device *pdev)
 	if (IS_ERR(data->gpio_usb_mux))
 		dev_info(dev, "can't request USB MUX GPIO\n");
 
+	/* process id-pin first so that we start with the right status */
+	queue_delayed_work(system_wq, &data->work, 0);
+	flush_delayed_work(&data->work);
+
 	/* register extcon device */
 	data->edev = devm_extcon_dev_allocate(dev, int3496_cable);
 	if (IS_ERR(data->edev))

drivers/extcon/extcon-intel-int3496.c static int int3496_probe(struct platform_device *pdev) · line 162

Quoted 9 lines
drivers/extcon/extcon-intel-int3496.c +0 -3
@@ -153,9 +157,6 @@ static int int3496_probe(struct platform_device *pdev)
 		return ret;
 	}
 
-	/* queue initial processing of id-pin */
-	queue_delayed_work(system_wq, &data->work, 0);
-
 	platform_set_drvdata(pdev, data);
 
 	return 0;

Looks good to me.
How about adding the 'Fixes' commit id on patch description? I think that your revert patch of axp288 and this patch should be included in extcon-fixes branch.

Signature · 3 lines
Best Regards,
Chanwoo Choi
Samsung Electronics
#3 lore raw

Hi,

Quoted 38 lines · Chanwoo Choi
Quoted 37 lines · Hans de Goede

Some other drivers may be waiting for our extcon to show-up (exiting their probe methods with -EPROBE_DEFER until we show up).

These drivers will typically get the cable state directly after getting the extcon, this commit changes the int3496 code to process the id-pin before registering the extcon, so that other drivers see the correct state right away.

---
  drivers/extcon/extcon-intel-int3496.c | 7 ++++---
  1 file changed, 4 insertions(+), 3 deletions(-)
drivers/extcon/extcon-intel-int3496.c +4 -3
index c8691b5a9cb0..b23ee9d993a3 100644
@@ -131,6 +131,10 @@ static int int3496_probe(struct platform_device *pdev)
  	if (IS_ERR(data->gpio_usb_mux))
  		dev_info(dev, "can't request USB MUX GPIO\n");
 
+	/* process id-pin first so that we start with the right status */
+	queue_delayed_work(system_wq, &data->work, 0);
+	flush_delayed_work(&data->work);
+
  	/* register extcon device */
  	data->edev = devm_extcon_dev_allocate(dev, int3496_cable);
  	if (IS_ERR(data->edev))
@@ -153,9 +157,6 @@ static int int3496_probe(struct platform_device *pdev)
  		return ret;
  	}
 
-	/* queue initial processing of id-pin */
-	queue_delayed_work(system_wq, &data->work, 0);
-
  	platform_set_drvdata(pdev, data);
 
Quoted 5 lines

drivers/extcon/extcon-intel-int3496.c static int int3496_probe(struct platform_device *pdev) · line 162

Quoted 2 lines
drivers/extcon/extcon-intel-int3496.c +0 -0
  	return 0;

Looks good to me.
How about adding the 'Fixes' commit id on patch description?

Right, this could use a:

Fixes 2f556bdb9f2e ("extcon: int3496: Add Intel INT3496 ACPI device extcon driver")

tag.

Quoted 2 lines

I think that your revert patch of axp288 and this patch should be included in extcon-fixes branch.

Ack.

Regards,

Hans

maintainer
#4 lore raw
Quoted 44 lines · Hans de Goede

Hi,

Quoted 32 lines · Chanwoo Choi
Quoted 31 lines · Hans de Goede

Some other drivers may be waiting for our extcon to show-up (exiting their probe methods with -EPROBE_DEFER until we show up).

These drivers will typically get the cable state directly after getting the extcon, this commit changes the int3496 code to process the id-pin before registering the extcon, so that other drivers see the correct state right away.

---
  drivers/extcon/extcon-intel-int3496.c | 7 ++++---
  1 file changed, 4 insertions(+), 3 deletions(-)
drivers/extcon/extcon-intel-int3496.c +3 -0
index c8691b5a9cb0..b23ee9d993a3 100644
@@ -131,6 +131,10 @@ static int int3496_probe(struct platform_device *pdev)
      if (IS_ERR(data->gpio_usb_mux))
          dev_info(dev, "can't request USB MUX GPIO\n");
  +    /* process id-pin first so that we start with the right status */
+    queue_delayed_work(system_wq, &data->work, 0);
+    flush_delayed_work(&data->work);
+
      /* register extcon device */
      data->edev = devm_extcon_dev_allocate(dev, int3496_cable);
      if (IS_ERR(data->edev))
@@ -153,9 +157,6 @@ static int int3496_probe(struct platform_device *pdev)
          return ret;
      }
  -    /* queue initial processing of id-pin */
Quoted 8 lines

drivers/extcon/extcon-intel-int3496.c static int int3496_probe(struct platform_device *pdev)

Quoted 5 lines
drivers/extcon/extcon-intel-int3496.c +0 -2
-    queue_delayed_work(system_wq, &data->work, 0);
-
      platform_set_drvdata(pdev, data);
        return 0;
 

Looks good to me.
How about adding the 'Fixes' commit id on patch description?

Quoted 4 lines

Right, this could use a:

Fixes 2f556bdb9f2e ("extcon: int3496: Add Intel INT3496 ACPI device extcon driver")

Please resend this patch Because you need to send the Linux stable mailing list.

Signature · 3 lines
Best Regards,
Chanwoo Choi
Samsung Electronics
#5 lore raw

HI,

Quoted 45 lines · Chanwoo Choi
Quoted 44 lines · Hans de Goede

Hi,

Quoted 40 lines · Chanwoo Choi
Quoted 28 lines · Hans de Goede

Some other drivers may be waiting for our extcon to show-up (exiting their probe methods with -EPROBE_DEFER until we show up).

These drivers will typically get the cable state directly after getting the extcon, this commit changes the int3496 code to process the id-pin before registering the extcon, so that other drivers see the correct state right away.

---
   drivers/extcon/extcon-intel-int3496.c | 7 ++++---
   1 file changed, 4 insertions(+), 3 deletions(-)
drivers/extcon/extcon-intel-int3496.c +3 -0
index c8691b5a9cb0..b23ee9d993a3 100644
@@ -131,6 +131,10 @@ static int int3496_probe(struct platform_device *pdev)
       if (IS_ERR(data->gpio_usb_mux))
           dev_info(dev, "can't request USB MUX GPIO\n");
   +    /* process id-pin first so that we start with the right status */
+    queue_delayed_work(system_wq, &data->work, 0);
+    flush_delayed_work(&data->work);
+
       /* register extcon device */
       data->edev = devm_extcon_dev_allocate(dev, int3496_cable);
       if (IS_ERR(data->edev))

drivers/extcon/extcon-intel-int3496.c static int int3496_probe(struct platform_device *pdev) · line 162

Quoted 8 lines
drivers/extcon/extcon-intel-int3496.c +0 -2
@@ -153,9 +157,6 @@ static int int3496_probe(struct platform_device *pdev)
           return ret;
       }
   -    /* queue initial processing of id-pin */
-    queue_delayed_work(system_wq, &data->work, 0);
-
       platform_set_drvdata(pdev, data);
         return 0;
 

Looks good to me.
How about adding the 'Fixes' commit id on patch description?

Quoted 6 lines
Quoted 4 lines

Right, this could use a:

Fixes 2f556bdb9f2e ("extcon: int3496: Add Intel INT3496 ACPI device extcon driver")

Please resend this patch Because you need to send the Linux stable mailing list.

Ok, I've just send a v2 with the Fixes and Cc: stable tags added.

Regards,

Hans