Actually, I've had several people email me regarding this problem asking if I'd gotten around my incarnation of the problem. (The answer is still no.)
> (And "broken" is fully compatible > with "works in Windows"; especially since MSFT has access to > chip errata and workarounds that aren't generally known.)
Yes, that's probable in this case, although I would expect a non-compliant host to cause the driver to fail initialization, not lock up the system...
> So far as I can tell, I never got any of that information.
? How odd.. I remember gathering the information... It should have been sent, but oh well.
I'll re-do all of this with 2.6.1, but in the old kernel it went this way:
>>> The tweak is enabling EHCI_VERBOSE_DEBUG as well CONFIG_USB_DEBUG, >>> then Alt-SysRq-8 to make debug messages go to your console (you're >>> not using X11 here, "telinit 2" and login that way).
All done.
>>> Then when you plug in whatever device, you'll see messages coming >>> out on the console. The last ones before the hang are what I'll >>> be wanting to know. ...
No, actually, I don't. There are two possible approaches to causing the failure. Loading ehci-hcd with a USB device plugged in, or loading ehci-hcd and plugging a device in. If ehci-hcd isn't loaded, yes infos about whatever the device is pop up on the console, but whenever ehci gets loaded, the system locks and there are no additional messages. If ehci-hcd is loaded already and a device is plugged in, there are NO debugging messages regarding the device being plugged at all.
>>> Various things can misbehave, there's no point in wondering what >>> unless you can report the last things going on before lockup.
I can't, because there don't seem to be any debugging messages accociated with whatever the problem is.
Like I said, I'll re-do all of this with 2.6.1 and capture a full text session to mail in.
Hopefully I'll have that mailed in by monday.
cheers --Alex
David Brownell wrote:
Alexander Markley wrote:
Hey all... Some of you may remember me and my EHCI problem, but for those that don't, I'll summerize the details.
Basically, if I load ehci-hcd into my kernel, and plug ANY USB devices in, the computer completely hangs. (No magic sysrq keys, no kernel panic messages, no indication at all that the system is about to go to the great big bitbucket in the sky.)
The last email messages I have between us were inconclusive. I almost suspect your EHCI is broken; nobody else has ever reported such a problem. (And "broken" is fully compatible with "works in Windows"; especially since MSFT has access to chip errata and workarounds that aren't generally known.)
However, as of August 25 I was expecting more info from you:
The tweak is enabling EHCI_VERBOSE_DEBUG as well CONFIG_USB_DEBUG, then Alt-SysRq-8 to make debug messages go to your console (you're not using X11 here, "telinit 2" and login that way).
Then when you plug in whatever device, you'll see messages coming out on the console. The last ones before the hang are what I'll be wanting to know. ...
Various things can misbehave, there's no point in wondering what unless you can report the last things going on before lockup.
So far as I can tell, I never got any of that information.
- Dave
------------------------------------------------------- The SF.Net email is sponsored by EclipseCon 2004 Premiere Conference on Open Tools Development and Integration See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. http://www.eclipsecon.org/osdn _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
