[sane-devel] libusb vs linux kernel scanner driver

[m. allan noah]

ok, this seems highly intermittent. various combinations of:
a. scanner driver already loaded at plugin time,
b. scanner driver loaded by hotplug
c. manually rmmod and insmod scanner driver, 
d. allowing the scanner to go in and out of power saving mode 

are causing many errors in the syslog, and pauses for many seconds either 
when initially attaching to the scanner, or after the adf loads the first 
sheet from the hopper, but before it returns image data.

i am having some trouble being able to repeat the errors on demand, but it 
may have something to do with sanei_usb scanning the /dev/usb/scanner0 
files, and then immediately scanning the same device via libusb, but i 
cannot say for certain.

i will continue to look.

allan

On Fri, 11 Jul 2003, Henning Meier-Geinitz wrote:

> Hi,
> 
> On Fri, Jul 11, 2003 at 02:57:43PM -0400, m. allan noah wrote:
> > no, the kernel driver starts the scan the second you ask, with no errors in the 
> > log, the libusb connection has a couple timeouts, and the scan starts after the 
> > timeouts are over, maybe 30 seconds. i am thinking that the logic in the kernel 
> > scanner driver and the libusb portions of sanei_usb must be sending different 
> > packets to the scanner, even though the backend does not distinguish between the 
> > two.
> 
> The scanner driver limits the maximum size of the packets to 32k I
> think. Maybe that's the difference?
> 
> Can you show me a some lines of the log with SANE_DEBUG_SANEI_USB=255
> and libusb? e.g. the part where the timeout occurs plus 20 lines
> before that happens?
> 
> Bye,
>   Henning
> _______________________________________________
> Sane-devel mailing list
> Sane-devel at www.mostang.com
> http://www.mostang.com/mailman/listinfo/sane-devel
> 

-- 
"so don't tell us it can't be done, putting down what you don't know.
money isn't our god, integrity will free our souls" - Max Cavalera