[BISECTED] (?) since 3.16: USB video cam of an HP Pavilion dv7 laptop not working

[BISECTED] (?) since 3.16: USB video cam of an HP Pavilion dv7 laptop not working

[Jörg-Volker Peetz]

Since 3.16 the builtin USB video cam of an HP Pavilion dv7 laptop shows only a
black screen with the application guvcview although the video LED is turned on.
There is no hint in dmesg. Also the output of guvcview shows no conspicuousness.

With kernel version 3.15.9 the camera works.

I tried to bisect between good v3.15 and bad v3.16, which identified

commit 9262c19d14c433a6a1ba25c3ff897cb89e412309
Author: Dan Williams <dan.j.williams@intel.com>
Date:   Tue May 20 18:08:12 2014 -0700

    usb: disable port power control if not supported in wHubCharacteristics

    A hub indicates whether it supports per-port power control via the
    wHubCharacteristics field in its descriptor.  If it is not supported
    a hub will still emulate ClearPortPower(PORT_POWER) requests by
    stopping the link state machine.  However, since this does not save
    power do not bother suspending.

    This also consolidates support checks into a
    hub_is_port_power_switchable() helper.

    Acked-by: Alan Stern <stern@rowland.harvard.edu>
    Signed-off-by: Dan Williams <dan.j.williams@intel.com>
    Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

In the bad bisection steps the video device even was not found. Therefore, I'm
not sure if I identified the right culprit.

Thanks for any help,
Jörg-Volker.

Re: [BISECTED] (?) since 3.16: USB video cam of an HP Pavilion dv7 laptop not working

[Jörg-Volker Peetz]

[-- Attachment #1: Type: text/plain, Size: 1690 bytes --]

Jörg-Volker Peetz wrote on 09/08/2014 17:17:
> Since 3.16 the builtin USB video cam of an HP Pavilion dv7 laptop shows only a
> black screen with the application guvcview although the video LED is turned on.
> There is no hint in dmesg. Also the output of guvcview shows no conspicuousness.
> 
> With kernel version 3.15.9 the camera works.
> 
> I tried to bisect between good v3.15 and bad v3.16, which identified
> 
> commit 9262c19d14c433a6a1ba25c3ff897cb89e412309
> Author: Dan Williams <dan.j.williams@intel.com>
> Date:   Tue May 20 18:08:12 2014 -0700
> 
>     usb: disable port power control if not supported in wHubCharacteristics
> 
>     A hub indicates whether it supports per-port power control via the
>     wHubCharacteristics field in its descriptor.  If it is not supported
>     a hub will still emulate ClearPortPower(PORT_POWER) requests by
>     stopping the link state machine.  However, since this does not save
>     power do not bother suspending.
> 
>     This also consolidates support checks into a
>     hub_is_port_power_switchable() helper.
> 
>     Acked-by: Alan Stern <stern@rowland.harvard.edu>
>     Signed-off-by: Dan Williams <dan.j.williams@intel.com>
>     Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> 
> In the bad bisection steps the video device even was not found. Therefore, I'm
> not sure if I identified the right culprit.
> 

As you suggested Dan, I reverted this commit on top of 3.16.0 (the commit could
not be reverted cleanly, I had to do it manually, patch appended). But still I
see a black picture from the camera. So this commit is not the culprit.

How to proceed with the bisection?
-- 
Regards,
Jörg-Volker.


[-- Attachment #2: my-patch --]
[-- Type: text/plain, Size: 2926 bytes --]

--- hub.c.orig	2014-08-04 00:25:02.000000000 +0200
+++ hub.c	2014-09-09 11:56:24.721312469 +0200
@@ -820,6 +820,8 @@ EXPORT_SYMBOL_GPL(usb_hub_clear_tt_buffe
 static void hub_power_on(struct usb_hub *hub, bool do_delay)
 {
 	int port1;
+	u16 wHubCharacteristics =
+			le16_to_cpu(hub->descriptor->wHubCharacteristics);
 
 	/* Enable power on each port.  Some hubs have reserved values
 	 * of LPSM (> 2) in their descriptors, even though they are
@@ -827,7 +829,7 @@ static void hub_power_on(struct usb_hub
 	 * but only emulate it.  In all cases, the ports won't work
 	 * unless we send these messages to the hub.
 	 */
-	if (hub_is_port_power_switchable(hub))
+	if ((wHubCharacteristics & HUB_CHAR_LPSM) < 2)
 		dev_dbg(hub->intfdev, "enabling power on all ports\n");
 	else
 		dev_dbg(hub->intfdev, "trying to enable port power on "
@@ -4548,6 +4550,8 @@ static void hub_port_connect(struct usb_
 	unsigned unit_load;
 	struct usb_device *hdev = hub->hdev;
 	struct usb_hcd *hcd = bus_to_hcd(hdev->bus);
+	unsigned wHubCharacteristics =
+			le16_to_cpu(hub->descriptor->wHubCharacteristics);
 	struct usb_port *port_dev = hub->ports[port1 - 1];
 	struct usb_device *udev = port_dev->child;
 
@@ -4586,7 +4590,7 @@ static void hub_port_connect(struct usb_
 			test_bit(port1, hub->removed_bits)) {
 
 		/* maybe switch power back on (e.g. root hub was reset) */
-		if (hub_is_port_power_switchable(hub)
+		if ((wHubCharacteristics & HUB_CHAR_LPSM) < 2
 				&& !port_is_power_on(hub, portstatus))
 			set_port_feature(hdev, port1, USB_PORT_FEAT_POWER);
 
--- hub.h.orig	2014-08-04 00:25:02.000000000 +0200
+++ hub.h	2014-09-09 12:52:40.027903791 +0200
@@ -119,16 +119,6 @@ extern int hub_port_debounce(struct usb_
 extern int usb_clear_port_feature(struct usb_device *hdev,
 		int port1, int feature);
 
-static inline bool hub_is_port_power_switchable(struct usb_hub *hub)
-{
-	__le16 hcs;
-
-	if (!hub)
-		return false;
-	hcs = hub->descriptor->wHubCharacteristics;
-	return (le16_to_cpu(hcs) & HUB_CHAR_LPSM) < HUB_CHAR_NO_LPSM;
-}
-
 static inline int hub_is_superspeed(struct usb_device *hdev)
 {
 	return hdev->descriptor.bDeviceProtocol == USB_HUB_PR_SS;
--- port.c.orig	2014-08-04 00:25:02.000000000 +0200
+++ port.c	2014-09-09 12:58:46.521229430 +0200
@@ -445,16 +445,11 @@ int usb_hub_create_port_device(struct us
 	 */
 	pm_runtime_set_active(&port_dev->dev);
 	pm_runtime_get_noresume(&port_dev->dev);
-	pm_runtime_enable(&port_dev->dev);
+	if (!dev_pm_qos_expose_flags(&port_dev->dev,
+			PM_QOS_FLAG_NO_POWER_OFF))
+		pm_runtime_enable(&port_dev->dev);
 	device_enable_async_suspend(&port_dev->dev);
 
-	/*
-	 * Keep hidden the ability to enable port-poweroff if the hub
-	 * does not support power switching.
-	 */
-	if (!hub_is_port_power_switchable(hub))
-		return 0;
-
 	/* Attempt to let userspace take over the policy. */
 	retval = dev_pm_qos_expose_flags(&port_dev->dev,
 			PM_QOS_FLAG_NO_POWER_OFF);

Re: [BISECTED] (?) since 3.16: USB video cam of an HP Pavilion dv7 laptop not working

[Dominik Kopp]

Jörg-Volker Peetz wrote:

> Since 3.16 the builtin USB video cam of an HP Pavilion dv7 laptop shows
> only a black screen with the application guvcview although the video LED
> is turned on. There is no hint in dmesg. Also the output of guvcview shows
> no conspicuousness.
> 
> With kernel version 3.15.9 the camera works.
> 
> I tried to bisect between good v3.15 and bad v3.16, which identified
> 
> commit 9262c19d14c433a6a1ba25c3ff897cb89e412309
...
> 
> Thanks for any help,
> Jörg-Volker.

please have a look at
https://bugzilla.kernel.org/show_bug.cgi?id=81611

Due to a change in the driver, guvcview won't work with kernel 3.16.
You have to revert up to two kernel commits or use the latest version of 
guvcview (git version >2.0.0)

regards
Dominik Kopp