Steven P. Ulrick wrote:
> All of these items are things that I tried to run relating to USB2 devices.
> As most of you know, "D" means the following (quoted from "man ps"):
> "D Uninterruptible sleep (usually IO)"
>
> I suspect that it is no coincedence that these USB2 related items are all in
> "Uninterruptible sleep" This could at least somewhat explain why I need to
> reboot to get the use of our devices back.
> On a side note, does anyone know of a way to kill the above processes, short
> of rebooting? "kill -9 PID" does not touch them.
To the best of my knowledge, no, there isn't.
"D" state means that the process tried to call the kernel, but the
kernel never returned. Kill signals are not processed while a process is
in kernel mode (they are saved up for when it exits). In this case,
since the hardware has effectively "gone away", the kernel isn't
returning.
This is standard Unix behaviour.
Incidentally, you say that this motherboard is a KD7, and only has the
problem on the rear ports. And this is reproducable with several USB2
devices. Taking a quick look at the documentation suggests that both
front and rear ports use the same VIA USB2 implementation. That means
that as far as software is concerned, they should be identical: any
software problems should show up on any port.
This *thoroughly* suggests that there is something wrong with your rear
USB ports. It could be that the rear USB ports aren't quite wired
properly, or a trace problem on the motherboard, or a defect in the
silicon that handles those USB devices.
Sorry.
James.
--
James Wilkinson | Og just boggle how stupid spammer is. How stupid
Exeter Devon UK | spamhaus is. How stupid spamhaven is. Og thought
E-mail address: james | there was such thing as "evolution". How all these
@westexe.demon.co.uk | stupid people still alive? Og boggle. Boggle Og.
| -- Caveman Og
