On Tue, Dec 04, 2018 at 03:38:25PM -0500, Kyle Williams wrote:
> Description: Some USB device / host controller combinations seem to have
> problems with Link Power management. In particular it is described that
> the combination of certain Logitech devices and other powered media
> devices such as the Atrus device causes 'not enough bandwidth for
> new device state'error.
>
> This patch creates quirk entries for the tested Logitech device
> indicating LPM should remain disabled for the device.
>
> Signed-off-by: Kyle Williams <[email protected]>
> ---
> drivers/usb/core/quirks.c | 16 ++++++++++++++++
> 1 file changed, 16 insertions(+)
Along with what Alan said, this patch was sent in html mode, making it
impossible to apply, and it was rejected by the mailing lists.
Please fix up your email client before resending patches. Or just use
'git send-email'.
thanks,
greg k-h
On Wed, Dec 05, 2018 at 10:39:02AM +0100, Greg Kroah-Hartman wrote:
> On Tue, Dec 04, 2018 at 03:38:25PM -0500, Kyle Williams wrote:
> > Description: Some USB device / host controller combinations seem to have
> > problems with Link Power management. In particular it is described that
> > the combination of certain Logitech devices and other powered media
> > devices such as the Atrus device causes 'not enough bandwidth for
> > new device state'error.
> >
> > This patch creates quirk entries for the tested Logitech device
> > indicating LPM should remain disabled for the device.
> >
> > Signed-off-by: Kyle Williams <[email protected]>
> > ---
> > drivers/usb/core/quirks.c | 16 ++++++++++++++++
> > 1 file changed, 16 insertions(+)
>
> Along with what Alan said, this patch was sent in html mode, making it
> impossible to apply, and it was rejected by the mailing lists.
>
> Please fix up your email client before resending patches. Or just use
> 'git send-email'.
>
> thanks,
>
> greg k-h
Thanks for explaining why the patches were being rejected by the mailing
lists. I will move to using git send-email for sending patches and mutt
for sending replies in the future.
Kyle Williams
From: Kyle Williams <[email protected]>
Description: Some USB device / host controller combinations seem to have
problems with Link Power management. In particular it is described that
the combination of certain Logitech uvc devices and other powered media
devices such causes 'not enough bandwidth for new device state' error.
This patch enables the USB_QUIRK_NO_LPM quirk entries for all connected
Logitech UVC devices indicating LPM should remain disabled for the device.
Signed-off-by: Kyle Williams <[email protected]>
---
Changes in v2:
- changed commit message
- disble lpm for all logitech uvc devices instead of listing manually
- changes to allow the lpm to correctly be disabled after enumeration
drivers/usb/core/hub.c | 87 +++++++++++++++++++++------------------
drivers/usb/core/quirks.c | 2 +-
2 files changed, 49 insertions(+), 40 deletions(-)
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 0f9381b69a3b..8f366ec4d21b 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -365,6 +365,51 @@ static void usb_set_lpm_parameters(struct usb_device *udev)
usb_set_lpm_sel(udev, &udev->u2_params);
}
+/*
+ * There are reports of USB 3.0 devices that say they support USB 2.0 Link PM
+ * when they're plugged into a USB 2.0 port, but they don't work when LPM is
+ * enabled.
+ *
+ * Only enable USB 2.0 Link PM if the port is internal (hardwired), or the
+ * device says it supports the new USB 2.0 Link PM errata by setting the BESL
+ * support bit in the BOS descriptor.
+ */
+static void hub_set_initial_usb2_lpm_policy(struct usb_device *udev)
+{
+ struct usb_hub *hub = usb_hub_to_struct_hub(udev->parent);
+ int connect_type = USB_PORT_CONNECT_TYPE_UNKNOWN;
+
+ if (!udev->usb2_hw_lpm_capable || !udev->bos)
+ return;
+
+ if (hub)
+ connect_type = hub->ports[udev->portnum - 1]->connect_type;
+
+ if ((udev->bos->ext_cap->bmAttributes & cpu_to_le32(USB_BESL_SUPPORT))
+ || connect_type == USB_PORT_CONNECT_TYPE_HARD_WIRED) {
+ udev->usb2_hw_lpm_allowed = 1;
+ usb_set_usb2_hardware_lpm(udev, 1);
+ }
+}
+
+void usb_update_device_lpm(struct usb_hcd *hcd, struct usb_device *udev)
+{
+ int retval;
+
+ if (udev->wusb == 0 && le16_to_cpu(udev->descriptor.bcdUSB) >= 0x0201) {
+ retval = usb_get_bos_descriptor(udev);
+ if (!retval) {
+ udev->lpm_capable = usb_device_supports_lpm(udev);
+ usb_set_lpm_parameters(udev);
+ }
+ }
+
+ /* notify HCD that we have a device connected and addressed */
+ if (hcd->driver->update_device)
+ hcd->driver->update_device(hcd, udev);
+ hub_set_initial_usb2_lpm_policy(udev);
+}
+
/* USB 2.0 spec Section 11.24.4.5 */
static int get_hub_descriptor(struct usb_device *hdev,
struct usb_hub_descriptor *desc)
@@ -2295,7 +2340,6 @@ static int usb_enumerate_device_otg(struct usb_device *udev)
return err;
}
-
/**
* usb_enumerate_device - Read device configs/intfs/otg (usbcore-internal)
* @udev: newly addressed device (in ADDRESS state)
@@ -2351,6 +2395,8 @@ static int usb_enumerate_device(struct usb_device *udev)
usb_detect_interface_quirks(udev);
+ usb_update_device_lpm(hcd, udev);
+
return 0;
}
@@ -4402,33 +4448,6 @@ static int hub_set_address(struct usb_device *udev, int devnum)
return retval;
}
-/*
- * There are reports of USB 3.0 devices that say they support USB 2.0 Link PM
- * when they're plugged into a USB 2.0 port, but they don't work when LPM is
- * enabled.
- *
- * Only enable USB 2.0 Link PM if the port is internal (hardwired), or the
- * device says it supports the new USB 2.0 Link PM errata by setting the BESL
- * support bit in the BOS descriptor.
- */
-static void hub_set_initial_usb2_lpm_policy(struct usb_device *udev)
-{
- struct usb_hub *hub = usb_hub_to_struct_hub(udev->parent);
- int connect_type = USB_PORT_CONNECT_TYPE_UNKNOWN;
-
- if (!udev->usb2_hw_lpm_capable || !udev->bos)
- return;
-
- if (hub)
- connect_type = hub->ports[udev->portnum - 1]->connect_type;
-
- if ((udev->bos->ext_cap->bmAttributes & cpu_to_le32(USB_BESL_SUPPORT)) ||
- connect_type == USB_PORT_CONNECT_TYPE_HARD_WIRED) {
- udev->usb2_hw_lpm_allowed = 1;
- usb_set_usb2_hardware_lpm(udev, 1);
- }
-}
-
static int hub_enable_device(struct usb_device *udev)
{
struct usb_hcd *hcd = bus_to_hcd(udev->bus);
@@ -4779,19 +4798,9 @@ hub_port_init(struct usb_hub *hub, struct usb_device *udev, int port1,
usb_detect_quirks(udev);
- if (udev->wusb == 0 && le16_to_cpu(udev->descriptor.bcdUSB) >= 0x0201) {
- retval = usb_get_bos_descriptor(udev);
- if (!retval) {
- udev->lpm_capable = usb_device_supports_lpm(udev);
- usb_set_lpm_parameters(udev);
- }
- }
+ usb_update_device_lpm(hcd, udev);
retval = 0;
- /* notify HCD that we have a device connected and addressed */
- if (hcd->driver->update_device)
- hcd->driver->update_device(hcd, udev);
- hub_set_initial_usb2_lpm_policy(udev);
fail:
if (retval) {
hub_port_disable(hub, port1, 0);
diff --git a/drivers/usb/core/quirks.c b/drivers/usb/core/quirks.c
index 0690fcff0ea2..bbbb7e6df958 100644
--- a/drivers/usb/core/quirks.c
+++ b/drivers/usb/core/quirks.c
@@ -440,7 +440,7 @@ static const struct usb_device_id usb_quirk_list[] = {
static const struct usb_device_id usb_interface_quirk_list[] = {
/* Logitech UVC Cameras */
{ USB_VENDOR_AND_INTERFACE_INFO(0x046d, USB_CLASS_VIDEO, 1, 0),
- .driver_info = USB_QUIRK_RESET_RESUME },
+ .driver_info = USB_QUIRK_RESET_RESUME | USB_QUIRK_NO_LPM },
{ } /* terminating entry must be last */
};
--
2.20.0.rc2.403.gdbc3b29805-goog
Changes in v3:
- seperated work to allow matched interfaces to use the
USB_QUIRK_NO_LPM quirk to affect a broader range of devices instead of
setting each device individually
Kyle Williams (2):
USB: quirks: Check device interface LPM capability
USB: quirks: Disable LPM for Logitech UVC devices
drivers/usb/core/hub.c | 87 +++++++++++++++++++++------------------
drivers/usb/core/quirks.c | 2 +-
2 files changed, 49 insertions(+), 40 deletions(-)
--
2.20.0.rc2.403.gdbc3b29805-goog
From: Kyle Williams <[email protected]>
Description: enable the ability to disable LPM for all devices matched
by interface information
Signed-off-by: Kyle Williams <[email protected]>
---
drivers/usb/core/hub.c | 87 +++++++++++++++++++++++-------------------
1 file changed, 48 insertions(+), 39 deletions(-)
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 0f9381b69a3b..8f366ec4d21b 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -365,6 +365,51 @@ static void usb_set_lpm_parameters(struct usb_device *udev)
usb_set_lpm_sel(udev, &udev->u2_params);
}
+/*
+ * There are reports of USB 3.0 devices that say they support USB 2.0 Link PM
+ * when they're plugged into a USB 2.0 port, but they don't work when LPM is
+ * enabled.
+ *
+ * Only enable USB 2.0 Link PM if the port is internal (hardwired), or the
+ * device says it supports the new USB 2.0 Link PM errata by setting the BESL
+ * support bit in the BOS descriptor.
+ */
+static void hub_set_initial_usb2_lpm_policy(struct usb_device *udev)
+{
+ struct usb_hub *hub = usb_hub_to_struct_hub(udev->parent);
+ int connect_type = USB_PORT_CONNECT_TYPE_UNKNOWN;
+
+ if (!udev->usb2_hw_lpm_capable || !udev->bos)
+ return;
+
+ if (hub)
+ connect_type = hub->ports[udev->portnum - 1]->connect_type;
+
+ if ((udev->bos->ext_cap->bmAttributes & cpu_to_le32(USB_BESL_SUPPORT))
+ || connect_type == USB_PORT_CONNECT_TYPE_HARD_WIRED) {
+ udev->usb2_hw_lpm_allowed = 1;
+ usb_set_usb2_hardware_lpm(udev, 1);
+ }
+}
+
+void usb_update_device_lpm(struct usb_hcd *hcd, struct usb_device *udev)
+{
+ int retval;
+
+ if (udev->wusb == 0 && le16_to_cpu(udev->descriptor.bcdUSB) >= 0x0201) {
+ retval = usb_get_bos_descriptor(udev);
+ if (!retval) {
+ udev->lpm_capable = usb_device_supports_lpm(udev);
+ usb_set_lpm_parameters(udev);
+ }
+ }
+
+ /* notify HCD that we have a device connected and addressed */
+ if (hcd->driver->update_device)
+ hcd->driver->update_device(hcd, udev);
+ hub_set_initial_usb2_lpm_policy(udev);
+}
+
/* USB 2.0 spec Section 11.24.4.5 */
static int get_hub_descriptor(struct usb_device *hdev,
struct usb_hub_descriptor *desc)
@@ -2295,7 +2340,6 @@ static int usb_enumerate_device_otg(struct usb_device *udev)
return err;
}
-
/**
* usb_enumerate_device - Read device configs/intfs/otg (usbcore-internal)
* @udev: newly addressed device (in ADDRESS state)
@@ -2351,6 +2395,8 @@ static int usb_enumerate_device(struct usb_device *udev)
usb_detect_interface_quirks(udev);
+ usb_update_device_lpm(hcd, udev);
+
return 0;
}
@@ -4402,33 +4448,6 @@ static int hub_set_address(struct usb_device *udev, int devnum)
return retval;
}
-/*
- * There are reports of USB 3.0 devices that say they support USB 2.0 Link PM
- * when they're plugged into a USB 2.0 port, but they don't work when LPM is
- * enabled.
- *
- * Only enable USB 2.0 Link PM if the port is internal (hardwired), or the
- * device says it supports the new USB 2.0 Link PM errata by setting the BESL
- * support bit in the BOS descriptor.
- */
-static void hub_set_initial_usb2_lpm_policy(struct usb_device *udev)
-{
- struct usb_hub *hub = usb_hub_to_struct_hub(udev->parent);
- int connect_type = USB_PORT_CONNECT_TYPE_UNKNOWN;
-
- if (!udev->usb2_hw_lpm_capable || !udev->bos)
- return;
-
- if (hub)
- connect_type = hub->ports[udev->portnum - 1]->connect_type;
-
- if ((udev->bos->ext_cap->bmAttributes & cpu_to_le32(USB_BESL_SUPPORT)) ||
- connect_type == USB_PORT_CONNECT_TYPE_HARD_WIRED) {
- udev->usb2_hw_lpm_allowed = 1;
- usb_set_usb2_hardware_lpm(udev, 1);
- }
-}
-
static int hub_enable_device(struct usb_device *udev)
{
struct usb_hcd *hcd = bus_to_hcd(udev->bus);
@@ -4779,19 +4798,9 @@ hub_port_init(struct usb_hub *hub, struct usb_device *udev, int port1,
usb_detect_quirks(udev);
- if (udev->wusb == 0 && le16_to_cpu(udev->descriptor.bcdUSB) >= 0x0201) {
- retval = usb_get_bos_descriptor(udev);
- if (!retval) {
- udev->lpm_capable = usb_device_supports_lpm(udev);
- usb_set_lpm_parameters(udev);
- }
- }
+ usb_update_device_lpm(hcd, udev);
retval = 0;
- /* notify HCD that we have a device connected and addressed */
- if (hcd->driver->update_device)
- hcd->driver->update_device(hcd, udev);
- hub_set_initial_usb2_lpm_policy(udev);
fail:
if (retval) {
hub_port_disable(hub, port1, 0);
--
2.20.0.rc2.403.gdbc3b29805-goog
From: Kyle Williams <[email protected]>
Description: Some USB device / host controller combinations seem to have
problems with Link Power management. In particular it is described that
the combination of certain Logitech uvc devices and other powered media
devices such causes 'not enough bandwidth for new device state' error.
This patch enables the USB_QUIRK_NO_LPM quirk entries for all connected
Logitech UVC devices indicating LPM should remain disabled for the device.
Signed-off-by: Kyle Williams <[email protected]>
---
drivers/usb/core/quirks.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/usb/core/quirks.c b/drivers/usb/core/quirks.c
index 0690fcff0ea2..bbbb7e6df958 100644
--- a/drivers/usb/core/quirks.c
+++ b/drivers/usb/core/quirks.c
@@ -440,7 +440,7 @@ static const struct usb_device_id usb_quirk_list[] = {
static const struct usb_device_id usb_interface_quirk_list[] = {
/* Logitech UVC Cameras */
{ USB_VENDOR_AND_INTERFACE_INFO(0x046d, USB_CLASS_VIDEO, 1, 0),
- .driver_info = USB_QUIRK_RESET_RESUME },
+ .driver_info = USB_QUIRK_RESET_RESUME | USB_QUIRK_NO_LPM },
{ } /* terminating entry must be last */
};
--
2.20.0.rc2.403.gdbc3b29805-goog
On Fri, Dec 07, 2018 at 08:47:21PM -0500, Kyle Williams wrote:
> From: Kyle Williams <[email protected]>
>
> Description: enable the ability to disable LPM for all devices matched
> by interface information
Why have "Description:" in here? Of course this is the description :)
And the subject doesn't make much sense, what does this have to do with
quirks?
Also, you need more information here in the description about why you
are doing all of this, as it is, I have no idea why this patch is needed
at all.
thanks,
greg k-h
On Fri, Dec 07, 2018 at 08:47:22PM -0500, Kyle Williams wrote:
> From: Kyle Williams <[email protected]>
>
> Description: Some USB device / host controller combinations seem to have
> problems with Link Power management. In particular it is described that
> the combination of certain Logitech uvc devices and other powered media
> devices such causes 'not enough bandwidth for new device state' error.
Same thing with "Description:" :(
On Wed, Dec 12, 2018 at 12:37:09PM +0100, Greg Kroah-Hartman wrote:
> On Fri, Dec 07, 2018 at 08:47:21PM -0500, Kyle Williams wrote:
> > From: Kyle Williams <[email protected]>
> >
> > Description: enable the ability to disable LPM for all devices matched
> > by interface information
>
> Why have "Description:" in here? Of course this is the description :)
>
> And the subject doesn't make much sense, what does this have to do with
> quirks?
>
> Also, you need more information here in the description about why you
> are doing all of this, as it is, I have no idea why this patch is needed
> at all.
>
> thanks,
>
> greg k-h
It seems the underlying issue has been solved in
0472bf0 xhci: Prevent U1/U2 link pm states if exit latency is too long
However, the patch to fix the issue when matching the USB_QUIRK_NO_LPM
quirk by iterface could be useful in the future. Currently nothing will
happen that quirk is matched by interface instead of by vid/pid pair.
Is their anything that would need to be changed besides removing the
Description tag?
-Kyle
On Thu, Jan 10, 2019 at 10:49:09AM -0500, Kyle Williams wrote:
> On Wed, Dec 12, 2018 at 12:37:09PM +0100, Greg Kroah-Hartman wrote:
> > On Fri, Dec 07, 2018 at 08:47:21PM -0500, Kyle Williams wrote:
> > > From: Kyle Williams <[email protected]>
> > >
> > > Description: enable the ability to disable LPM for all devices matched
> > > by interface information
> >
> > Why have "Description:" in here? Of course this is the description :)
> >
> > And the subject doesn't make much sense, what does this have to do with
> > quirks?
> >
> > Also, you need more information here in the description about why you
> > are doing all of this, as it is, I have no idea why this patch is needed
> > at all.
> >
> > thanks,
> >
> > greg k-h
>
> It seems the underlying issue has been solved in
> 0472bf0 xhci: Prevent U1/U2 link pm states if exit latency is too long
>
> However, the patch to fix the issue when matching the USB_QUIRK_NO_LPM
> quirk by iterface could be useful in the future. Currently nothing will
> happen that quirk is matched by interface instead of by vid/pid pair.
> Is their anything that would need to be changed besides removing the
> Description tag?
Don't add code to the kernel that is not used now. If in the future
someone needs to add this type of functionality, worry about it then...
thanks,
greg k-h