Thanks Roger,

Quoting Roger Larsson <[EMAIL PROTECTED]>:

> On Monday 18 August 2003 18.18, Dmitri Katchalov wrote:
> > Hi,
> >
> > "USB device not accepting new address"
> >
> > I've got this Atmel AT76C503A based wireless USB adapter.
> > There is a driver for this wireless card  (2 of them in fact), they appear
> > to work with 2.4.18 kernel.
> 
> "appear to work" - do they work or don't they? 

I couldn't get it connected to my network due to other unrelated issues 
but at least I was able to bring up wvlan0 interface and view/change 
settings using iwconfig. This and the log files indicate that the startup 
sequence has completed successfully. 

> Are you using the same host hardware? 
Yes. The hardware is all new so there may be a problem.
It's Intel Pentium 4 system with ABIT IS7 M/B, i865/ICH5 chipset, 
UHCI/EHCI host controller.

> Summary: If it does work with 2.4.18 kernel but not with a 2.6 (or 2.4.20?)
> on
> the same HW then it would be very interesting to find out why. But if it is 
> not the same HW (including the host side) - then there are a lot of possible
> 
> causes (BIOS, actual hardware - like bugs in OHCI HW implementation, 
> drivers, ...)
> > Unfortunately running 2.4.18 kernel is not 
> > an option for me as I have a few other problems with it and I need
> > some of the new stuff in 2.6.x kernel anyway. So I'm in the process
> > of upgrading the driver to 2.6.x format. The adapter is a bit tricky in
> > that it requires firmware to be downloaded first and then a reset.
> > I've got the firmware downloading stage working and now I have
> > this problem with reset. I have now disabled everything except the
> > reset call and still I'm consistently getting the same error. 
> 
> > Needless to say it works flawlessly with the 'other' OS.
> Using the same HW and generic drivers for both the device and host side?
> (A lot of hardware bugs are covered up in drivers)

Yes, same hardware. AFAIK Atmel linux driver is derived from their 
windows NDIS driver. I have made a trace of USB traffic on Windows 
and it matches linux driver source. This rules out any potential 
cover-up in the Atmel driver.

BTW while I am at it I have one particular issue with usb_reset_device():
- If you want to reset the device you own you have to release the 
interface before calling reset and then reclaim it back afterwards as
usb_destroy_configuration wipes out usb_interface without releasing any 
attached drivers.
- When you release the interface it tries to set default altsetting for the 
interface by sending it a corresponding control URB. At this point in time 
the device is waiting for a reset and not responding to any commands. This 
causes set_interface to time out. 
Bottomline: there must be a way of resetting USB port without assuming 
anything about the state of the device attached to it.

Regards,
Dmitri




-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to