I just posted this to LKML, but I figured I'd follow up here too (less
noise).
I found out that if SMP is disabled, you can print with a USB printer
while using usb-uhci.o. If SMP is not disabled, then you need to use
uhci.o or your system will hang.
I'm not sure if this is just an SMP thing or an SMP-Via chipset thing.
I've heard from two other people about this. I don't know what chipset
one of them is using, but the other looks to be using a Via chipset, but
one that's different from mine (Apollo Pro 133a).
Can anyone help with actually finding the bug? My kernel programming
and USB stills are weak, so I really can't do much with this, but I have
no problems testing patches or suggested fixes.
pete
On Thu, 05 Jul 2001, Pete Toscano wrote:
> Hello,
>
> I wrote to this list a few days ago asking if anyone else is seeing USB
> printing causing lockups. I heard back from one person, but I'll bet
> there are other out there.
>
> I didn't have time at the time to track this down further, but I have
> looked into this a bit more now. First, the problem. Whenever I would
> print anything to a USB device (/dev/usb/lp0, in my case), the machine
> would seem to lock up. Up until yesterday, I'd just be saying that it
> locks up, but yesterday, after it "hung", I let it sit for a few hours
> while I went on with my life. When I got back, it looks like it did
> print stuff, but mostly junk. Some of this junk was recognizable as
> various files (completely unrelated) from my filesystem. The machine
> was still unresponsive to anything, but my guess now is that it was
> doing something, but just stuck in a very busy loop.
>
> Anyway, I traced this problem back to the transition between 2.4.3 and
> 2.4.4. 2.4.5 and 2.4.6 are both affected by this problem. I also
> discovered that this problem went away if I used uhci.o instead of
> usb-uhci.o. Finally, the other guy I heard from, with similar printing
> problems, also found that it went away when using uhci.o. In addition,
> we're both using VIA chipsets.
>
> My system is a daul P3-600 system on a Tyan Tiger 133 (Apollo Pro 133a
> chipset). I'm using an SMP-enabled kernel, but I'm disabling APIC so
> that I can actually use USB. The other USB devices (mouse, SanDisk
> SDDR-31, and Rio 500) all seem to work fine with the usb-uhci.o module.
>
> I took a look at the patch to usb-uhci.c for 2.4.4, but the details of
> the changes were just way beyond my comprehension. I'm more than happy
> to try experimental patches, give extra system information, and just
> basically work with anyone who has the skills to fix this (Johannes?
> Greg?).
>
> (BTW, besides the wonderful debugging possibilities of having two UHCI
> modules in the kernel tree, what is the purpose of having both UHCI and
> UHCI_ALT?)
>
> Thanks,
> pete
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
http://lists.sourceforge.net/lists/listinfo/linux-usb-users
