Bug 225845

Summary:	kernel 2.6.25-gentoo-r4: ehci_hcd HC died
Product:	Gentoo Linux	Reporter:	Robin Hartley <robin.hartley>
Component:	[OLD] Core system	Assignee:	Gentoo Linux bug wranglers <bug-wranglers>
Status:	RESOLVED CANTFIX
Severity:	normal
Priority:	High
Version:	unspecified
Hardware:	AMD64
OS:	Linux
Whiteboard:
Package list:		Runtime testing required:	---

Description Robin Hartley 2008-06-11 08:28:16 UTC

While playing with a flaky webcam the USB2 controller dies.
var/log/messages has one message "ehci_hcd 0000:00:0b.1: HC died; cleaning up".

After this the USB2 hub disappears from lsusb. the USB1.0 controller remains.

Problem not seen in 2.6.24-gentoo-r4.



Reproducible: Sometimes

Steps to Reproduce:
1.unplug webcam
2.plug in webcam
3.repeat 4 or 5 times & fails. I reboot to restore (I don't know how to restart USB any other way)




Fresh build from gentoo-source. I note it's marked as ~64.
I down loaded it based on a recomendation from the UVC people as there is supposed to be a fix for a possible USB bug in this version.
I was hoping this would fix my webcam problems.

It's not a problem for me but since there has been a USB fix done this could be related & this may be of interest to the upstream kernel people.

Feel free to change the priority if my selection is not appropriate.

Comment 1 Jeroen Roovers (RETIRED) gentoo-dev

2008-06-11 12:57:59 UTC

I would say this isn't a bug (and if it were, it wouldn't be a bug in Gentoo) but a feature. Software must bail out when hardware fails. The best you can do is a) buy new hardware and b) write a patch against the kernel that distinguishes between flaky USB end devices and non-flaky USB controllers.

Comment 2 Robin Hartley 2008-06-12 08:01:49 UTC

(In reply to comment #1)
> I would say this isn't a bug (and if it were, it wouldn't be a bug in Gentoo)
> but a feature. Software must bail out when hardware fails. The best you can do
> is a) buy new hardware and b) write a patch against the kernel that
> distinguishes between flaky USB end devices and non-flaky USB controllers.
> 

I'm not sure of what the exact problem is with the end device but I don't see why it should cause the whole USB2 process(?) to crash. My main reason for reporting it is that I understand there are some changes to the USB in this version & it's a change in behaviour. I don't know how to report the issue to the right people so I hought I'd start with the knowledgable people here.
I've never been able to crash the USB system on the previous version.

I'm happy to close the report if this isn't the correct way to report it.