Simon Kirby wrote:
>
> On Mon, Feb 14, 2000 at 09:40:04PM +0000, Tim Waugh wrote:
>
> > I just saw this happen. I was using a DC240 digital camera, and suddenly
> > /dev/kodak (c 180 80) became unopenable.
> >
> > Briefly, camera #0 was detected, but then suddenly #0 went away and was
> > replaced by #1. Unplugging and replugging the camera made things better
> > again.
"should not happen" -- nothing should "suddenly" go away like that.
It's not a symptom I've been able to reproduce. But there's a patch
I've been meaning to submit which might relate to that problem.
> Whoa...you mean it actually worked for you in 2.3 at some point? :)
>
> I haven't been able to get my DC290 to work at all, and I've _always_
> been having a problem similar to what you just posted.
Yeah, but you also had lots of other problems too -- at the level
of lots of "that can't happen" errors. Did you ever get to the
root of those problems? In your clase it seemed like your SMP box
was repeatedly calling the device probe routine -- vs just once.
It was unclear to me that you ever really ruled out it being an SMP
issue with the USB code, by running it as a real uniprocessor, as
just one of the strangnesses you turned up.
> How long has it been working for you? What kernels have worked? Do you
> have anything else on your USB bus(es) other than the camera?
I've been thinking that such problems are more likely than not proof
of some of the problems in HCI code that folk have been recently more
aggessive at tracking down.
I'll be quite pleased if for example the latest kernel handles the OHCI
chip (NEC) in my primary development machine again -- then I can work
on USB without taking down a production machine, even run it routinely.
- Dave
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Feb 23 2000 - 21:00:21 EST