[linux-usb-devel] Re: [PATCH/RFC 2.6.16.5 1/1] usb/media/quickcam_messenger driver

2006-05-14 Thread Oliver Neukum
Am Sonntag, 14. Mai 2006 16:11 schrieb [EMAIL PROTECTED]: +   struct regval { +   u16 reg; +   u8 val; +   }; +   /* this table is derived from the +   qc-usb-messenger code */ +   struct regval regval_table[] = { +   {

[linux-usb-devel] Re: [PATCH/RFC 2.6.16.5 1/1] usb/media/quickcam_messenger driver

2006-05-14 Thread Oliver Neukum
Am Sonntag, 14. Mai 2006 16:11 schrieb [EMAIL PROTECTED]: +   qcm_setup_input_int(cam, uvd); +   uvd-streaming = 1; +   uvd-curframe = -1; +   qcm_camera_on(uvd); This is a race condition with the URB for the interrupt endpoint. You must set the streaming flag before you

[linux-usb-devel] 64 Bit Linux - USB KB

2006-05-14 Thread Jayaprakash Shanmugam
Hello All, I have 64 bit Linux connected to a USB KB emulator ( Cypress based USB device - Works as a KB in Windows 32/64 bits). Though the firmware gets enumerated as a HID device, the keystrokes are not received by the operating system. Is there any suggestion for me ? Thanks in advance.

[linux-usb-devel] Re: USB keyboards and HID_QUIRK_NOGET

2006-05-14 Thread Vojtech Pavlik
On Fri, May 12, 2006 at 02:16:04PM -0400, Alan Stern wrote: On Wed, 10 May 2006, Vojtech Pavlik wrote: On Wed, May 10, 2006 at 10:50:47AM -0400, Alan Stern wrote: Dmitry and Vojtech: So many USB keyboards require the HID_QUIRK_NOGET blacklist flag, does it make sense to set the

[linux-usb-devel] [RFC][PATCH][RESEND] usbhid: Implement generic framework for descriptor patching

2006-05-14 Thread Matheus Izvekov
The motivation for this is that the quirks framework doesnt scale well. Currently there is only one device needing the descriptor table to be patched, but as soon as something like this gets accepted, im going to implement a fixup for my own crappy keyboard, and over time others will follow.

[linux-usb-devel] Re: [PATCH/RFC 2.6.16.5 1/1] usb/media/quickcam_messenger driver

2006-05-14 Thread Jaya Kumar
On 5/15/06, Oliver Neukum [EMAIL PROTECTED] wrote: Am Sonntag, 14. Mai 2006 16:11 schrieb [EMAIL PROTECTED]: +static int qcm_setup_on_open(struct uvd *uvd) +{ +qcm_sensor_set_gains(uvd, uvd-vpic.hue, +uvd-vpic.colour, uvd-vpic.contrast); +qcm_sensor_set_exposure(uvd, uvd-vpic.brightness);

[linux-usb-devel] [PATCH] usbhid: fix bug when DEBUG_DATA is defined

2006-05-14 Thread Matheus Izvekov
When DEBUG_DATA is defined, the second parameter to hid_parse_report gets overwritten, gettings assigned the value of rsize instead of the return value of hid_get_class_descriptor This patch fixes it by creating a new variable to hold this value, instead of reusing the variable n This patch is

[linux-usb-devel] OHCI_hcd 1.10 overrides EHCI_hcd 2.0

2006-05-14 Thread Ian
The following is part of the readout from Usbview. As it is quite a long readout I've only posted part of it in the hope someone can help. The main controller EHCI_hcd is 2.00 then there seem to be 3 secondary controllers OHCI_hcd 1.10. Problem, the ohci_hcd driver is on all ports and I want to