[Please CC me, I'm not subscribed]
Some days ago I have wrote to this list to expose a problem I have using kernel 2.6.0-textX + stv680 + UHCI_HCD driver... the subject was: "USB-Problems with uhci_hcd + V4L + stv680 driver" http://marc.theaimsgroup.com/?l=linux-usb-devel&m=106493648430605&w=2
An oops in the stv680 code ... hmm, that looks much like what I see if I build that driver on my system:
000002b0 <stv_set_config>:
2b0: 83 ec 20 sub $0x20,%esp
2b3: 89 74 24 18 mov %esi,0x18(%esp,1)
2b7: 89 5c 24 14 mov %ebx,0x14(%esp,1)
2bb: 89 7c 24 1c mov %edi,0x1c(%esp,1)
2bf: 8b 74 24 24 mov 0x24(%esp,1),%esi
2c3: 8b 5c 24 28 mov 0x28(%esp,1),%ebx
2c7: 8b 7c 24 30 mov 0x30(%esp,1),%edi
2cb: 8b 56 04 mov 0x4(%esi),%edx
2ce: 8b 82 90 01 00 00 mov 0x190(%edx),%eax// OOPS HERE
2d4: 0f b6 40 05 movzbl 0x5(%eax),%eax
2d8: 39 c3 cmp %eax,%ebx
2da: 74 4e je 32a <stv_set_config+0x7a>In "configuration != dev->udev->actconfig->desc.bConfigurationValue", where bConfigurationValue is offset 5.
Looks as if dev->actconfig is somehow null. In this case I'll suspect that the problem is the stv680 driver accessing a device after it returned from disconnect() -- given that stack trace, and the fact that stv680_disconnect() doesn't null out the device pointer if the camera is in use.
Short summary of previous message: - using a software like camorama the video stops after some seconds
That could be an stv680 bug.
Do you also see it with an OHCI controller? If it happens with both controllers, it's more likely to be in stv680.
- then, unplugging/replugging the web-cam, I get a message like:
Unable to handle kernel NULL pointer dereference at virtual...
See above.
I think "2.4.x UHCI alternate driver" && "2.6.x UHCI-HCD" have some problems on my hardware (my chipset is VIA KT133 on Asus motherboard).
I seem to recall people saying that VIA has some problems with its ISO support through UHCI. It'd be worth grabbing one of the many inexpensive PCI-based USB adapters that don't have a VIA chip (I've seen them at under US$ 10); they'll have OHCI, and you can at least rule out that factor.
- Dave
------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. SourceForge.net hosts over 70,000 Open Source Projects. See the people who have HELPED US provide better services: Click here: http://sourceforge.net/supporters.php _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
