On Tue, Apr 11, 2000 at 11:38:02PM +0000, Brad Hards wrote:
> David Brownell wrote:
> > Seeing an undending stream of interrupts reported is odd,
> > and given that I saw it with OHCI when the unit wasn't
> > even plugged in any more ... it feels like there's a bug
> > somewhere outside the {U,O}HCI code that's somehow been
> > tweaked. Or else the *HCI code is starting to get more
> > compatible (bug-for-bug :-) than we anticipated!
>
> I first noticed this last night, with a minimal USB tree, using UHCI
> controller. A little more checking of the kernel logs showed this:
<...>
This is a problem I also have with my TI-hub. It is extremly sensitive to
EMI. Even touching the hub shielding with a "free floating" USB-cable
makes the hub simetimes disabling all ports. I changed the hub code to
detect this and re-enable the ports (under Windows you are now lost...).
There are maybe some devices that don't like this recovery.
> Another potentially important section is this:
>
> Apr 11 23:05:46 zhora kernel: hub.c: nonzero status in irq -84
> Apr 11 23:05:47 zhora kernel: usb-uhci.c: interrupt, status 3, frame#
> 960
> Apr 11 23:05:47 zhora kernel: hub.c: nonzero status in irq -84
> Apr 11 23:05:47 zhora kernel: usb-uhci.c: interrupt, status 3, frame#
> 1344
> Apr 11 23:05:47 zhora kernel: hub.c: nonzero status in irq -84
> Apr 11 23:05:47 zhora kernel: usb-uhci.c: interrupt, status 3, frame#
> 1728
> Apr 11 23:05:47 zhora kernel: hub.c: nonzero status in irq -84
> Apr 11 23:05:48 zhora kernel: usb-uhci.c: interrupt, status 3, frame# 64
> Apr 11 23:05:48 zhora kernel: hub.c: nonzero status in irq -84
> Apr 11 23:05:48 zhora kernel: usb-uhci.c: interrupt, status 3, frame#
> 448
> Apr 11 23:05:48 zhora kernel: hub.c: nonzero status in irq -84
> Apr 11 23:05:48 zhora kernel: usb-uhci.c: interrupt, status 3, frame#
> 832
> Apr 11 23:05:48 zhora kernel: hub.c: nonzero status in irq -84
> Apr 11 23:05:49 zhora kernel: usb-uhci.c: interrupt, status 3, frame#
> 1216
> Apr 11 23:05:49 zhora kernel: hub.c: nonzero status in irq -84
> Apr 11 23:05:49 zhora kernel: usb-uhci.c: interrupt, status 3, frame#
> 1600
>
> This is only once, and this is the time I noticed it. Lots and lots of
> messages, only stopping on reboot.
-84=-EILSEQ=Timeout during IN, so the hub doesn't respond anymore to the
interrupt requests. Does the root hub still recognize connects/reconnects?
If it does but the messages appear afte the TI hub gets disconnected, then
the interrupt-transfer isn't killed in hub.c.
> It looks like the hub in my bus powered keyboard (which is a TI one, I
> think) is switching off the downstream connection to the self-powered
> hub (also a TI one), and then restarting.
BTW:
Does this happen with both usb-uhci and uhci?
--
Bye
Georg Acher, [EMAIL PROTECTED]
http://www.in.tum.de/~acher/
"Oh no, not again !" The bowl of petunias
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
