Hello, I am using the USB Framework from Atmel to develop our product. I
imported all source code in usb-device-core-project-at91sam7se-ek-gnu.zip to
CrossWorks IDE and tried to debug the code. But I found the deviceState never go
up to "USBD_STATE_ADDRESS". After the deviceState becomes "USBD_STATE_DEFAULT"
and USBDCallbacks_Reset() is called, the deviceState goes back to 0, which is
suspend, and Windows poped up a warning message at the right corner. Can you
please give me some idea what is wrong with this?
THANKS!
problem in USB Core Stack for AT91sam7SE-EK
Started by ●February 28, 2011
Reply by ●March 3, 20112011-03-03
Problems were solved by disabling all trace functions in source code.
--- In A..., "wuqingwh" wrote:
>
> Hello, I am using the USB Framework from Atmel to develop our product. I imported all source code in usb-device-core-project-at91sam7se-ek-gnu.zip to CrossWorks IDE and tried to debug the code. But I found the deviceState never go up to "USBD_STATE_ADDRESS". After the deviceState becomes "USBD_STATE_DEFAULT" and USBDCallbacks_Reset() is called, the deviceState goes back to 0, which is suspend, and Windows poped up a warning message at the right corner. Can you please give me some idea what is wrong with this?
>
> THANKS!
>
--- In A..., "wuqingwh" wrote:
>
> Hello, I am using the USB Framework from Atmel to develop our product. I imported all source code in usb-device-core-project-at91sam7se-ek-gnu.zip to CrossWorks IDE and tried to debug the code. But I found the deviceState never go up to "USBD_STATE_ADDRESS". After the deviceState becomes "USBD_STATE_DEFAULT" and USBDCallbacks_Reset() is called, the deviceState goes back to 0, which is suspend, and Windows poped up a warning message at the right corner. Can you please give me some idea what is wrong with this?
>
> THANKS!
>
Reply by ●March 3, 20112011-03-03
Yes, I found this too - USB timing is very tight and the trace functions
inject too much delay. But a larger question: has anyone actually
shipped a product based on the USB Framework? Has anyone gotten it to
work with usbser.sys? After months of attention I'm still having issues
with random hangs and disconnects from a Windows host. I've found posts
suggesting usbser.sys does not retry timed out bulk transfers and that a
better path is to write a custom USB driver on the windows side.
On 3/2/2011 3:18 PM, wuqingwh wrote:
>
> Problems were solved by disabling all trace functions in source code.
>
> --- In A... ,
> "wuqingwh" wrote:
> >
> > Hello, I am using the USB Framework from Atmel to develop our
> product. I imported all source code in
> usb-device-core-project-at91sam7se-ek-gnu.zip to CrossWorks IDE and
> tried to debug the code. But I found the deviceState never go up to
> "USBD_STATE_ADDRESS". After the deviceState becomes
> "USBD_STATE_DEFAULT" and USBDCallbacks_Reset() is called, the
> deviceState goes back to 0, which is suspend, and Windows poped up a
> warning message at the right corner. Can you please give me some idea
> what is wrong with this?
> >
> > THANKS!
> >
inject too much delay. But a larger question: has anyone actually
shipped a product based on the USB Framework? Has anyone gotten it to
work with usbser.sys? After months of attention I'm still having issues
with random hangs and disconnects from a Windows host. I've found posts
suggesting usbser.sys does not retry timed out bulk transfers and that a
better path is to write a custom USB driver on the windows side.
On 3/2/2011 3:18 PM, wuqingwh wrote:
>
> Problems were solved by disabling all trace functions in source code.
>
> --- In A... ,
> "wuqingwh" wrote:
> >
> > Hello, I am using the USB Framework from Atmel to develop our
> product. I imported all source code in
> usb-device-core-project-at91sam7se-ek-gnu.zip to CrossWorks IDE and
> tried to debug the code. But I found the deviceState never go up to
> "USBD_STATE_ADDRESS". After the deviceState becomes
> "USBD_STATE_DEFAULT" and USBDCallbacks_Reset() is called, the
> deviceState goes back to 0, which is suspend, and Windows poped up a
> warning message at the right corner. Can you please give me some idea
> what is wrong with this?
> >
> > THANKS!
> >
