Bug 491 - On Abit KD7 Usb doesn't "exist" while with linux 2.4.20 it runs perfectly
Summary: On Abit KD7 Usb doesn't "exist" while with linux 2.4.20 it runs perfectly
Status: REJECTED DOCUMENTED
Alias: None
Product: Drivers
Classification: Unclassified
Component: USB (show other bugs)
Hardware: i386 Linux
: P2 high
Assignee: Greg Kroah-Hartman
URL:
Keywords:
Depends on: 10
Blocks:
  Show dependency tree
 
Reported: 2003-03-23 11:08 UTC by Samuele Kaplun
Modified: 2003-04-16 20:25 UTC (History)
1 user (show)

See Also:
Kernel Version: 2.5.67
Subsystem:
Regression: ---
Bisected commit-id:


Attachments
This is the dmesg output, with usb debug on (14.93 KB, text/plain)
2003-03-23 11:10 UTC, Samuele Kaplun
Details
This are all the modules I loaded in my test... (927 bytes, text/plain)
2003-03-23 11:11 UTC, Samuele Kaplun
Details
This is the configuration of my kernel (24.40 KB, text/plain)
2003-03-23 11:12 UTC, Samuele Kaplun
Details
This is the dmesg output of 2.4.20 on the same platform... I hope it's useful (10.33 KB, text/plain)
2003-03-25 13:45 UTC, Samuele Kaplun
Details
This is the content of my /proc/bus/usb/device file (1.53 KB, text/plain)
2003-03-26 11:46 UTC, Samuele Kaplun
Details
This is the what happens after plugging all my usb device... (227 bytes, text/plain)
2003-04-01 10:05 UTC, Samuele Kaplun
Details
What happens with modprobe uhci-hcd and no devices plugged in (4.15 KB, text/plain)
2003-04-01 10:08 UTC, Samuele Kaplun
Details

Description Samuele Kaplun 2003-03-23 11:08:41 UTC
Distribution: Slackware 8.1 (well with quite a lot of upgrades) 
Hardware Environment: Amd AthlonXP 2100+, Abit KD7 
Software Environment: kernel 2.5.65, gcc 3.2.2, glibc-2.2.5 
Problem Description: with this motherboard usb modules fire up, but no devices 
are found... with kernel 2.4.20, the problem doesn't exist... 
 
Steps to reproduce: running it on this VIA KT400, motherboard...
Comment 1 Samuele Kaplun 2003-03-23 11:10:44 UTC
Created attachment 228 [details]
This is the dmesg output, with usb debug on

I can't understand what the usb messages means... I can't find any wrong
message, but the devices (a mouse, a webcam, a printer, a scanner) aren't found
Comment 2 Samuele Kaplun 2003-03-23 11:11:25 UTC
Created attachment 229 [details]
This are all the modules I loaded in my test...
Comment 3 Samuele Kaplun 2003-03-23 11:12:40 UTC
Created attachment 230 [details]
This is the configuration of my kernel

I also modified my Makefile to include these gcc options:
-march=athlon-xp -mcpu=athlon-xp -mmmx -msse -m3dnow -mfpmath=sse
Comment 4 Greg Kroah-Hartman 2003-03-24 19:42:29 UTC
Um, it looks like your usb device is found just fine.

What is the symptom of the problem?
Comment 5 Samuele Kaplun 2003-03-25 13:24:42 UTC
It looks like the modules load very well, while the devices don't run at all: I mean, I have an 
optical mouse, which doesn't fire up it's light, a little webcam which doesn't receive it's power 
from the computer and looks switched off... (with kernel 2.5.65 and earlier too) 
(I had a Gygabyte ViaKT333 since last two week, and all these devices worked perfectly with 
the same kernel...) 
Sorry for my disgusting English :-) 
Comment 6 Samuele Kaplun 2003-03-25 13:25:43 UTC
[Aargh!!! read every "it's" like an "its"... :-D Sorry!] 
Comment 7 Samuele Kaplun 2003-03-25 13:45:55 UTC
Created attachment 233 [details]
This is the dmesg output of 2.4.20 on the same platform... I hope it's useful
Comment 8 Greg Kroah-Hartman 2003-03-26 11:09:49 UTC
What does /proc/bus/usb/devices show on 2.5 with your devices plugged in?
Comment 9 Samuele Kaplun 2003-03-26 11:46:37 UTC
Created attachment 235 [details]
This is the content of my /proc/bus/usb/device file

I noticed one thing... for just a second, my mouse seems to react when its
module is loading, but then it suddenly shutdown...
Comment 10 Greg Kroah-Hartman 2003-03-31 23:44:13 UTC
ok, so the host controller drivers are loaded.

Now try the following as root:

dmesg -c    # clears out the dmesg buffer

plug the device into the usb bus

dmesg

Please attach the result of the final dmesg output above.
Comment 11 Samuele Kaplun 2003-04-01 10:05:26 UTC
Created attachment 256 [details]
This is the what happens after plugging all my usb device...

I've done this after loading uhci-hcd.ko, clearing my dmesg buffer...
I think it's also important the messages after modprobe uhci-hcd which I attach
next...
Comment 12 Samuele Kaplun 2003-04-01 10:08:42 UTC
Created attachment 257 [details]
What happens with modprobe uhci-hcd and no devices plugged in

I activated all of the possible debug options and tricks in the kernel...
Comment 13 Samuele Kaplun 2003-04-13 05:18:11 UTC
I patched kernel 2.4.20, aginst Con Kolivas patch set ck5... 
The usb system, stop to run properly, when backporting to 2.4 the acpi system of 
2.5, as done with this patch set, I think... 
So I think the problem is concerned with acpi, too (as in bug 10) 
[I need to learn English, I know...] 
Comment 14 Samuele Kaplun 2003-04-13 05:47:21 UTC
Wow! I disabled APIC and IOAPIC, in my 2.5.67 kernel, and the usb system 
returned to Life! But it's only a workaround, isn't it? 
Comment 15 Greg Kroah-Hartman 2003-04-16 20:25:19 UTC
No, that's not a "workaround", it's the way your hardware is wired :)


Note You need to log in before you can comment on or make changes to this bug.