Am335 Starterware USB1 "device descriptor read/64, error -110"

Hi Shinu ,

Could you please check whether the device is detecting the RESET being driven by the host?

Hi Vineeth,

Thanks for your reply. The device is detecting SUSPEND being driven by the host then its detecting RESET and SOF.

Thanks

Hi Shinu,

It is possible that you may have missed some configuration . Could you please list out the changes that you have done to make the usb_dev_mouse example work with the instance change.

Hi Vineeth,

In starterware they are using usb instance 0 (USB0) .I changed the USB instance to one (USB1) as we are using USB1 as device.

I ran usb_dev_mouse example in starter kit without any change in code its working fine and receiving SOF continuously.  In our board its detecting SUSPEND ,RESET and stopped receiving SOF after sometime.

Thanks

HI  shinu,

In the main routine in usb_dev_mouse.c , the interrupt controller is being setup for USB0 . I guess you have configured it to work with USB1. 

Could you please put a breakpoint and check whether you are receiving interrupts from the proper instance usbdhandler.c.

Hi Vineeth,

ya i configured it for USB1.  Its receiving interrupts from proper instance only..

Thanks

Am i missing anyother configurations ?

Hi Shinu,

The first thing the usb_dev_mouse ( or any other device example for that matter) does is to disconnect and then connect back to the host. This would cause a connect event on the host which in-turn issues a BUS RESET. Hence you should get a RESET event after the suspend event. Reset event would cause an interrupt and the handler would go ahead and run the reset handler.

As you have confirmed that you are receiving a SUSPEND interrupt from USB1 (USB1DeviceIntHandler() -> USBDeviceIntHandlerInternal ()  ) the interrupt config is alright. 

There can be 2 reasons why you are not receiving RESET

1. Hardware issues cause the host not to see a connect.  - Have you tested this particular USB instance - port combination before using any other software ? 

If not please do post the relevant part of the schematic so  that I can get it reviewed from the concerned.

2. Software :

In USBDCDInit() function USBDevDisconnect is called to disconnect the device from the bus and then USBDevConnect () to connect back on the bus. Could you please confirm whether these functions are being called with the right instance number.

Hi Vineeth,

I think its not the hardware issue because i have checked USB1(device mode) with linux then it was working properly. 

Both  USBDevDisconnect() and USBDevConnect () functions are calling with correct instance from USBDCDInit() function.

Do i need to change the USB instance anywhere else ?

Thanks

Hi Vineeth,

USB device got detected. Actually the issue was with wrong USB instance being passed in some functions of usbdenum.c and usbdhid.c for requests transfer of data to the host on endpoint zero and to handle the change of address request from the host controller.

Thanks

Hi Shinu,

Great !

While going through the code I found a hard coded call in the usbdenum.c.  Is this this call that you have changed? 

  USBDReadAndDispatchRequest(0);  to   USBDReadAndDispatchRequest( index);

Thanks for pointing out this bug.

Hi Vineeth,

I have changed the following functions

USBDReadAndDispatchRequest( index); in usbdenum.c and

USBDCDSendDataEP0(index, (unsigned char *)psDevice->psHIDDescriptor, ulSize);
USBDCDSendDataEP0(index, pucReport, ulSize);
USBDCDSendDataEP0(index, (unsigned char *)&ulTimeout, 1);
USBDCDSendDataEP0(index, (unsigned char *)&ucProtocol, 1);
USBDCDSendDataEP0(index,(unsigned char *)psDevice->ppClassDescriptors[ulDesc], ulSize); in usbdhid.c

Thanks

Hi Shinu,

Thanks for pointing out the issues. I have raised an IR ( Clear Quest number : SDOCM00102713 ) and this would be fixed in the next release of StarterWare.

Apologies for the miss on our part as a silly mistake cost you your time.

Hi Shinu,

I would have to reopen another forum post and then answer the same . Will do that  shortly.