From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20020830
Description of problem:
Upgraded system from 7.3, Logitech Optical USB Scroll mouse which worked fine
under 7.3 no longer functions when USB connected. Workaround - connect as PS2
device. Tried to connect several other USB devices, none are detected at boot
up. USB devices have power after BIOS Post, but are deactivated as soon as the
Kernel selection screen is displayed. Device is autodetected as Generic PS2
Three Button Mouse when connected to PS2 port. USB GRAVIS Gamecontroller no
Version-Release number of selected component (if applicable):
Steps to Reproduce:
Actual Results: No USB Devices are detected.
Expected Results: USB Devices should be detected and activated. Mouse worked
perfected as USB Scroll when running 7.3. Upgrade to 8.0 now only allows device
to work when PS2 connected and then it works only as generic three button, no
scroll functionality. GAME Controller should be autodetected as under 7.3,
however now no longer functions.
System Motherboard is Gigabyte 7VRXP with AMD 1900XP Processor.
Please attach output of "dmesg" and "cat /proc/interrupts".
PLEASE DO NOT DROP THEM IN THE COMMENT BOX.
Created attachment 79806 [details]
Output from DMESG and /proc/interrupts
I've attached the information as requested. Currently mouse is connected as
PS/2 device. No USB devices are currently attached to system (since it doesn't
Trace skillfuly taken, thanks. But the objective is to fix the USB,
and for that we need to see how it fails. Can you plug the mouse
back into a USB port and re-take dmesg? No need to reboot.
Created attachment 79824 [details]
Trace taken after mouse and camera plugged into USB ports. Thanks!
This bug is worse than I thought, EHCI is busted. Arjan, I'm taking it.
Worked offline with the customer, removed ehci-hcd from the
configuration (commented alias usb-controller in modules.conf),
so the box now works in USB 1.1 mode.
Investigating if 2.0 is useful. Motherboard manual:
Created attachment 80045 [details]
lspci -v with 4 1.1 ports and a 2.0 port
If I read it correctly, section "K" on page 23 hints that the EHCI
is connected to the internal header only. It should be fully
functional. Unfortunately, Anaconda cannot guess which wires
are connected where.
I don't understand why this was closed as not a bug. Are you saying the
instructions in the manual are incorrect or that Gigabyte is not provisioning
USB 2.0 to specifications? Can you please explain? Thanks!