Yes, 9 is the value that I send in with
usb_fill_int_urb (), which assigns 256 to interval.
The device is powered externally, and won't draw power
from  the bus.

/proc/bus/usb/devices:
T:  Bus=01 Lev=03 Prnt=06 Port=01 Cnt=02 Dev#= 23
Spd=480 MxCh= 0
D:  Ver= 2.00 Cls=ff(vend.) Sub=00 Prot=ff MxPS=64
#Cfgs=  1
C:* #Ifs= 1 Cfg#= 1 Atr=80 MxPwr=500mA
I:  If#= 0 Alt= 0 #EPs= 6 Cls=ff(vend.) Sub=00 Prot=00
Driver=prod_drvr
E:  Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=84(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=86(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=88(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=90(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=92(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms

I just call usb_submit_urb (urb, GFP_KERNEL); when I
need to submit an interrupt urb.

As the behavior is totally random, I thought that I
should check that route too.

Thanks.


--- Alan Stern <[EMAIL PROTECTED]> wrote:

> On Wed, 7 Jun 2006, usb usb wrote:
> 
> > Let me clarify one more thing - it's an interrupt
> URB.
> > Just before I submit the URB, I do some more
> checking.
> > I set some values, and also check them.
> > status = 0, interval = 256 (I send in 9=>2^8),
> 
> Um, what do you mean by "send in"?  Is 9 the value
> you pass to 
> usb_fill_int_urb?  If so then it's correct.  Or is 9
> the value you 
> assign to urb->interval?  If so then it's wrong:
> urb->interval should 
> be set to 256.
> 
> >  buffer
> > len = 4 (my requirement), and the callback
> function.
> 
> If this doesn't already answer your question,
> perhaps you should post the 
> relevant portion of your source code.  Include also
> the 
> /proc/bus/usb/devices entry for your device, so we
> can check that the 
> values you use are acceptable to the device.
> 
> > Everything seems to be alright. I will try to
> check
> > for halt (_get_status, _clear_halt) , and follow
> USB
> > specs (5.7.5), and see what happens. Is it a good
> > idea/practise to do those sycnhronous calls
> directly
> > from my code?
> 
> It's not a good idea to check things like that
> unless you have some good 
> reason for it.  In this case an endpoint halt would
> give you -EPIPE, not 
> -EINVAL.
> 
> > Is there a significant change in 2.6.16?
> 
> I don't think so.
> 
> Alan Stern
> 
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 


_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-users

Reply via email to