Bug 6214

Can you test 2.6.16-rc5 please?  This should be fixed there, but if not,
we really need to know.

Here's an explanation of those error messages:

-75 is the numeric code for -EOVERFLOW, which means the device sent more data
than the computer asked for.

"unable to read config index 0 descriptor/start" means that the -EOVERFLOW error
occurred while the computer was trying to read the device's first configuration
descriptor.  This happened twice because the computer retries after a failure.

Then the computer switched to a different strategy.  "device descriptor
read/all" means that the -EOVERFLOW error occurred while the computer was trying
the read the device's device descriptor.  This also happened twice and then the
computer gave up.

I can't think of any reason why installing KDE should cause these errors to
suddenly start happening.

Two things:

First: how can I get 2.6.16-r5? (I downloaded latest and aplied patch and
results in an compilation error)

Second: (recomendation): COuld the error messages be more explicit? (explain
better the problem instead of just -75 for example?)

Thanks..

P.D. If I could get that version of kernel I could try the device and report..
(isn't a full version download prepaired?)

First:  http://www.kernel.org/pub/linux/kernel/v2.6/testing/linux-2.6.16-rc5.tar.bz2
Note also that 2.6.16-rc6 is out now.

Second: Error messages could be more explicit.  But there are thousands of error
messages in the kernel; making them all more explicit would add a tremendous
amount of bloat, with no benefit at all to the vast majority of people who never
get those errors.

I have tested it with 2.6.16 and I get exactly same error.

uname -a:
Linux StormByte 2.6.16-gentoo #1 SMP Tue Mar 28 01:54:43 CEST 2006 x86_64 AMD
Athlon(tm) 64 X2 Dual Core Processor 3800+ AuthenticAMD GNU/Linux

If you need more info, just reply and tell me how to get that info (I am not
much advance user) and I will post it.

Thank you.

P.D. I wonder why it worked once and why not now (and wich features are involved
in this issue).

Thaks again.

Maybe the device will work okay at full speed instead of high speed if you rmmod ehci-hcd.

It might also be interesting to try booting in single-user mode, so that none of the KDE programs are 
running when you plug in the device.  Does that make any difference?

P.S.: I also wonder why it stopped working.  Could it be that your device suddenly broke?  Have you tried 
plugging it into other computers?

Created attachment 7695 [details]
dmesg output (doing modprobe ohci_hcd, etc..)

This is whole output of my probes (first don't detect correctly device and
after, finally all is working good) (but isn't the output a bit extrange?)

I fixed it with linux--gentoo-r7.

I had usb-storage, ohci_hcd, uhci_hcd and ehci_hcd builtin in kernel (and it
didn't work). Now I changed this config as a modules (all mentioned before).

If modprobing ehci_hcd and usb-storage it works well with the dmesg output (very
large) wich is posted above as attachment.

Could this be a bug? (not working when builtin)?

If need more info, reply this.

Thanks again!

In attached dmesg output ehci_hcd is first loaded and mass storage device
connected which results in:
usb 1-6: new high speed USB device using ehci_hcd and address 2
usb 1-6: unable to read config index 0 descriptor/all
usb 1-6: can't read configurations, error -75

Then when ehci_hcd is unloaded and ohci_hcd loaded instead:
usb 1-6: new full speed USB device using ohci_hcd and address 2
usb-storage: USB Mass Storage device detected

And everything is working but it's ohci_hcd which is naturally very slow as USB
2.0 is not used. This is a problem with ehci_hcd and mass storage devices, I
have the same problem with my ipod, using 2.6.15.

It seems that this is very related to Bug 6489 if not even a duplicate.

*** This bug has been marked as a duplicate of 6489 ***