Skip to contentSkip to conversation

drm/i915: Correct max delay for HDMI hotplug live status checking

PATCH Superseded Gary Wang 4 participants 9 messages Dec 14, 2015 – Dec 22, 2015

Subsystems (2)
lore
Share & tools
lore

Gary Wang

author
#0 lore raw

PATCH v2 1/1 drm/i915: Correct max delay for HDMI hotplug live status checking

The total delay of HDMI hotplug detecting with 30ms have already been split into a resolution of 3 retries of 10ms each, for the worst cases. But it still suffered from only waiting 10ms at most in intel_hdmi_detect(). This patch corrects it by reading hotplug status with 4 times at most for 30ms delay.

v2:
- straight up to loop execution for more clear in code readability

Cc Daniel Vetter <[email protected]>
Cc Gavin Hindman <[email protected]>
Cc Sonika Jindal <[email protected]>
Cc Shashank Sharma <[email protected]>
---
 drivers/gpu/drm/i915/intel_hdmi.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
 mode change 100644 => 100755 drivers/gpu/drm/i915/intel_hdmi.c
drivers/gpu/drm/i915/intel_hdmi.c +4 -2
old mode 100644
new mode 100755
index 6825543..97ed16f
@@ -1387,16 +1387,18 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	bool live_status = false;
-	unsigned int retry = 3;
+	unsigned int retry;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, connector->name);
 
 	intel_display_power_get(dev_priv, POWER_DOMAIN_GMBUS);
 
-	while (!live_status && --retry) {
+	for (retry = 0; retry <= 3; retry++) {
 		live_status = intel_digital_port_connected(dev_priv,
 				hdmi_to_dig_port(intel_hdmi));
+		if (live_status || (retry == 3))
+			break;
 		mdelay(10);
 	}
 
Signature · 6 lines
1.9.1

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
maintainer
#1 lore raw
Quoted 5 lines · Gary Wang

The total delay of HDMI hotplug detecting with 30ms have already been split into a resolution of 3 retries of 10ms each, for the worst cases. But it still suffered from only waiting 10ms at most in intel_hdmi_detect(). This patch corrects it by reading hotplug status with 4 times at most for 30ms delay.

It used to try twice, with 10 ms in between, but also with 10 ms after the last attempt, success or not.

Quoted 32 lines

v2:
- straight up to loop execution for more clear in code readability

Cc Daniel Vetter <[email protected]>
Cc Gavin Hindman <[email protected]>
Cc Sonika Jindal <[email protected]>
Cc Shashank Sharma <[email protected]>
---
 drivers/gpu/drm/i915/intel_hdmi.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
 mode change 100644 => 100755 drivers/gpu/drm/i915/intel_hdmi.c
drivers/gpu/drm/i915/intel_hdmi.c +1 -1
old mode 100644
new mode 100755
index 6825543..97ed16f
@@ -1387,16 +1387,18 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	bool live_status = false;
-	unsigned int retry = 3;
+	unsigned int retry;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, connector->name);
 

drivers/gpu/drm/i915/intel_hdmi.c intel_hdmi_detect(struct drm_connector *connector, bool force) · line 1403

Quoted 10 lines
drivers/gpu/drm/i915/intel_hdmi.c +3 -1
 	intel_display_power_get(dev_priv, POWER_DOMAIN_GMBUS);
 
-	while (!live_status && --retry) {
+	for (retry = 0; retry <= 3; retry++) {
 		live_status = intel_digital_port_connected(dev_priv,
 				hdmi_to_dig_port(intel_hdmi));
+		if (live_status || (retry == 3))
+			break;
 		mdelay(10);
 	}

Basically the "retry <= 3" is never the stop condition in your loop, and is thus misleading.

How about this instead, with the paradigm counting for loop:

	for (try = 0; !live_status && try < 4; try++) {
		if (try)
                	mdelay(10);
		live_status = intel_digital_port_connected(...);
	}

Note that I didn't actually check if this is the right thing to do; this is just your loop rewritten as I'd write it... ;)

BR,
Jani.

Signature · 5 lines
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#2 lore raw

Yes, your code is clear. It tried to check hot-plug with 4 times based on delay setup (30ms) in following patch,

commit 237ed86c693d8a8e4db476976aeb30df4deac74b

Author: Sonika Jindal <sonika.jindal@xxxxxxxxx>
Date:   Tue Sep 15 09:44:20 2015 +0530

drm/i915: Check live status before reading edid

I will refine patch for your review again.

Thanks!

Gary

-----Original Message-----
From: Jani Nikula [mailto:[email protected]]
Sent: Monday, December 14, 2015 7:29 PM
To: Wang, Gary C <[email protected]>; [email protected]
Subject: Re: [Intel-gfx] [PATCH v2] drm/i915: Correct max delay for HDMI hotplug live status checking

Quoted 5 lines · Gary Wang

The total delay of HDMI hotplug detecting with 30ms have already been split into a resolution of 3 retries of 10ms each, for the worst cases. But it still suffered from only waiting 10ms at most in intel_hdmi_detect(). This patch corrects it by reading hotplug status with 4 times at most for 30ms delay.

It used to try twice, with 10 ms in between, but also with 10 ms after the last attempt, success or not.

Quoted 33 lines

v2:
- straight up to loop execution for more clear in code readability

Cc Daniel Vetter <[email protected]>
Cc Gavin Hindman <[email protected]>
Cc Sonika Jindal <[email protected]>
Cc Shashank Sharma <[email protected]>
---
 drivers/gpu/drm/i915/intel_hdmi.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)  mode change 100644 

=> 100755 drivers/gpu/drm/i915/intel_hdmi.c

drivers/gpu/drm/i915/intel_hdmi.c +1 -1
old mode 100644
new mode 100755
index 6825543..97ed16f
@@ -1387,16 +1387,18 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	bool live_status = false;
-	unsigned int retry = 3;
+	unsigned int retry;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, connector->name);
 

drivers/gpu/drm/i915/intel_hdmi.c intel_hdmi_detect(struct drm_connector *connector, bool force) · line 1403

Quoted 10 lines
drivers/gpu/drm/i915/intel_hdmi.c +3 -1
 	intel_display_power_get(dev_priv, POWER_DOMAIN_GMBUS);
 
-	while (!live_status && --retry) {
+	for (retry = 0; retry <= 3; retry++) {
 		live_status = intel_digital_port_connected(dev_priv,
 				hdmi_to_dig_port(intel_hdmi));
+		if (live_status || (retry == 3))
+			break;
 		mdelay(10);
 	}

Basically the "retry <= 3" is never the stop condition in your loop, and is thus misleading.

How about this instead, with the paradigm counting for loop:

	for (try = 0; !live_status && try < 4; try++) {
		if (try)
                	mdelay(10);
		live_status = intel_digital_port_connected(...);
	}

Note that I didn't actually check if this is the right thing to do; this is just your loop rewritten as I'd write it... ;)

BR,
Jani.

--
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

author
#3 lore raw

PATCH v2 1/1 drm/i915: Correct max delay for HDMI hotplug live status checking

In reply to Jani Nikula

The total delay of HDMI hotplug detecting with 30ms have already been split into a resolution of 3 retries of 10ms each, for the worst cases. But it still suffered from only waiting 10ms at most in intel_hdmi_detect(). This patch corrects it by reading hotplug status with 4 times at most for 30ms delay.

v2:
- straight up to loop execution for more clear in code readability
- mdelay will replace with msleep by Daniel's new patch

drm/i915: mdelay(10) considered harmful

- suggest to re-evaluate try times for being compatible to old HDMI monitor

Cc Jani Nikula <[email protected]>
Cc Daniel Vetter <[email protected]>
Cc Gavin Hindman <[email protected]>
Cc Sonika Jindal <[email protected]>
Cc Shashank Sharma <[email protected]>
---
 drivers/gpu/drm/i915/intel_hdmi.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
 mode change 100644 => 100755 drivers/gpu/drm/i915/intel_hdmi.c
drivers/gpu/drm/i915/intel_hdmi.c +4 -3
old mode 100644
new mode 100755
index 6825543..912d6c3
@@ -1387,17 +1387,18 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	bool live_status = false;
-	unsigned int retry = 3;
+	unsigned int try;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, connector->name);
 
 	intel_display_power_get(dev_priv, POWER_DOMAIN_GMBUS);
 
-	while (!live_status && --retry) {
+	for (try = 0; !live_status && try < 4; try++) {
+		if (try)
+			mdelay(10);
 		live_status = intel_digital_port_connected(dev_priv,
 				hdmi_to_dig_port(intel_hdmi));
-		mdelay(10);
 	}
 
 	if (!live_status)
Signature · 6 lines
1.9.1

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
author Reviewed-by Cooper Chiou Tested-by on v2 1/1
#4 lore raw

For BSW platform,
it seems 10ms delay of hot-plug detecting is not enough and to cause high fail-rate on old HDMI monitor, and I would like correct it.

Then user needs to wait screen blank for 2 mins or re-plugin cable (impact to user experience).

Thanks for your review.

Gary

-----Original Message-----
From: Wang, Gary C
Sent: Tuesday, December 15, 2015 12:41 PM
To: [email protected]
Cc: Wang, Gary C <[email protected]>; Jani Nikula <[email protected]>; Daniel Vetter <[email protected]>; Hindman, Gavin <[email protected]>; Jindal, Sonika <[email protected]>; Sharma, Shashank <[email protected]>
Subject: [PATCH v2] drm/i915: Correct max delay for HDMI hotplug live status checking

The total delay of HDMI hotplug detecting with 30ms have already been split into a resolution of 3 retries of 10ms each, for the worst cases. But it still suffered from only waiting 10ms at most in intel_hdmi_detect(). This patch corrects it by reading hotplug status with 4 times at most for 30ms delay.

v2:
- straight up to loop execution for more clear in code readability
- mdelay will replace with msleep by Daniel's new patch

drm/i915: mdelay(10) considered harmful

- suggest to re-evaluate try times for being compatible to old HDMI monitor

Cc Jani Nikula <[email protected]>
Cc Daniel Vetter <[email protected]>
Cc Gavin Hindman <[email protected]>
Cc Sonika Jindal <[email protected]>
Cc Shashank Sharma <[email protected]>
---
 drivers/gpu/drm/i915/intel_hdmi.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)  mode change 100644 => 100755 drivers/gpu/drm/i915/intel_hdmi.c
drivers/gpu/drm/i915/intel_hdmi.c +4 -3
old mode 100644
new mode 100755
index 6825543..912d6c3
@@ -1387,17 +1387,18 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	bool live_status = false;
-	unsigned int retry = 3;
+	unsigned int try;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, connector->name);
 
 	intel_display_power_get(dev_priv, POWER_DOMAIN_GMBUS);
 
-	while (!live_status && --retry) {
+	for (try = 0; !live_status && try < 4; try++) {
+		if (try)
+			mdelay(10);
 		live_status = intel_digital_port_connected(dev_priv,
 				hdmi_to_dig_port(intel_hdmi));
-		mdelay(10);
 	}
 
 	if (!live_status)

--
1.9.1

_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

#5 lore raw
In reply to Gary Wang
Quoted 14 lines · Gary Wang

The total delay of HDMI hotplug detecting with 30ms have already been split into a resolution of 3 retries of 10ms each, for the worst cases. But it still suffered from only waiting 10ms at most in intel_hdmi_detect(). This patch corrects it by reading hotplug status with 4 times at most for 30ms delay.

v2:
- straight up to loop execution for more clear in code readability
- mdelay will replace with msleep by Daniel's new patch

drm/i915: mdelay(10) considered harmful

- suggest to re-evaluate try times for being compatible to old HDMI monitor

Quoted 8 lines
Cc Jani Nikula <[email protected]>
Cc Daniel Vetter <[email protected]>
Cc Gavin Hindman <[email protected]>
Cc Sonika Jindal <[email protected]>
Cc Shashank Sharma <[email protected]>

Queued for -next (including cc: fixes), thanks for the patch. -Daniel

Quoted 36 lines
---
 drivers/gpu/drm/i915/intel_hdmi.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
 mode change 100644 => 100755 drivers/gpu/drm/i915/intel_hdmi.c
drivers/gpu/drm/i915/intel_hdmi.c +4 -3
old mode 100644
new mode 100755
index 6825543..912d6c3
@@ -1387,17 +1387,18 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	bool live_status = false;
-	unsigned int retry = 3;
+	unsigned int try;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, connector->name);
 
 	intel_display_power_get(dev_priv, POWER_DOMAIN_GMBUS);
 
-	while (!live_status && --retry) {
+	for (try = 0; !live_status && try < 4; try++) {
+		if (try)
+			mdelay(10);
 		live_status = intel_digital_port_connected(dev_priv,
 				hdmi_to_dig_port(intel_hdmi));
-		mdelay(10);
 	}
 
 	if (!live_status)
Signature · 1 line
1.9.1
Signature · 7 lines
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#6 lore raw
Quoted 24 lines · Daniel Vetter
Quoted 14 lines · Gary Wang

The total delay of HDMI hotplug detecting with 30ms have already been split into a resolution of 3 retries of 10ms each, for the worst cases. But it still suffered from only waiting 10ms at most in intel_hdmi_detect(). This patch corrects it by reading hotplug status with 4 times at most for 30ms delay.

v2:
- straight up to loop execution for more clear in code readability
- mdelay will replace with msleep by Daniel's new patch

drm/i915: mdelay(10) considered harmful

- suggest to re-evaluate try times for being compatible to old HDMI monitor

Quoted 8 lines
Cc Jani Nikula <[email protected]>
Cc Daniel Vetter <[email protected]>
Cc Gavin Hindman <[email protected]>
Cc Sonika Jindal <[email protected]>
Cc Shashank Sharma <[email protected]>
Quoted 1 line

Queued for -next (including cc: fixes), thanks for the patch.

Using mdelay() or fixed to use msleep()?
-Chris

Signature · 5 lines
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#7 lore raw
Quoted 25 lines · Chris Wilson
Quoted 24 lines · Daniel Vetter
Quoted 22 lines · Gary Wang

The total delay of HDMI hotplug detecting with 30ms have already been split into a resolution of 3 retries of 10ms each, for the worst cases. But it still suffered from only waiting 10ms at most in intel_hdmi_detect(). This patch corrects it by reading hotplug status with 4 times at most for 30ms delay.

v2:
- straight up to loop execution for more clear in code readability
- mdelay will replace with msleep by Daniel's new patch

drm/i915: mdelay(10) considered harmful

- suggest to re-evaluate try times for being compatible to old HDMI monitor

Cc Jani Nikula <[email protected]>
Cc Daniel Vetter <[email protected]>
Cc Gavin Hindman <[email protected]>
Cc Sonika Jindal <[email protected]>
Cc Shashank Sharma <[email protected]>
Quoted 3 lines
Quoted 1 line

Queued for -next (including cc: fixes), thanks for the patch.

Using mdelay() or fixed to use msleep()?

msleep patch was merged already, I just had to resolve the conflict. -Daniel

Signature · 7 lines
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#8 lore raw

Sorry for my incorrect grammar on my patch. It should be following description (mdelay-->msleep),
...
v2:
- straight up to loop execution for more clear in code readability
- mdelay will be replaced with msleep by Daniel's new patch
drm/i915: mdelay(10) considered harmful
...

Gary

-----Original Message-----
From: Daniel Vetter [mailto:[email protected]] On Behalf Of Daniel Vetter
Sent: Monday, December 21, 2015 7:52 PM
To: Chris Wilson <[email protected]>; Daniel Vetter <[email protected]>; Wang, Gary C <[email protected]>; [email protected]
Subject: Re: [Intel-gfx] [PATCH v2] drm/i915: Correct max delay for HDMI hotplug live status checking

Quoted 26 lines · Chris Wilson
Quoted 25 lines · Daniel Vetter
Quoted 23 lines · Gary Wang

The total delay of HDMI hotplug detecting with 30ms have already been split into a resolution of 3 retries of 10ms each, for the worst cases. But it still suffered from only waiting 10ms at most in intel_hdmi_detect(). This patch corrects it by reading hotplug status with 4 times at most for 30ms delay.

v2:
- straight up to loop execution for more clear in code readability
- mdelay will replace with msleep by Daniel's new patch

drm/i915: mdelay(10) considered harmful

- suggest to re-evaluate try times for being compatible to old HDMI monitor

Cc Jani Nikula <[email protected]>
Cc Daniel Vetter <[email protected]>
Cc Gavin Hindman <[email protected]>
Cc Sonika Jindal <[email protected]>
Cc Shashank Sharma <[email protected]>
Quoted 3 lines
Quoted 1 line

Queued for -next (including cc: fixes), thanks for the patch.

Using mdelay() or fixed to use msleep()?

msleep patch was merged already, I just had to resolve the conflict. -Daniel
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx