Skip to contentSkip to conversation

USB modem no longer detected in -test4

DISCUSSION Pavel Machek 5 participants 13 messages Sep 3, 2003 – Sep 8, 2003

lore
Share & tools
lore
#0 lore raw

Hi!

In 2.6.0-test4, USB ELSA modem no longer works. This is UHCI (on toshiba 4030cdt).

Relevant messages seem to be:

PM: Adding info for usb:1-1.2 drivers/usb/class/cdc-acm.c: need inactive config#2
PM: Adding info for usb:1-1.2:0 drivers/usb/class/cdc-acm.c: need inactive config#2

Pavel

Signature · 2 lines
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]
#1 lore raw
Quoted 11 lines · Pavel Machek

Hi!

In 2.6.0-test4, USB ELSA modem no longer works. This is UHCI (on toshiba 4030cdt).

Relevant messages seem to be:

PM: Adding info for usb:1-1.2 drivers/usb/class/cdc-acm.c: need inactive config#2
PM: Adding info for usb:1-1.2:0 drivers/usb/class/cdc-acm.c: need inactive config#2

And this worked just fine in 2.6.0-test3?

thanks,

greg k-h

#2 lore raw

Hi!

Quoted 11 lines
Quoted 9 lines

In 2.6.0-test4, USB ELSA modem no longer works. This is UHCI (on toshiba 4030cdt).

Relevant messages seem to be:

PM: Adding info for usb:1-1.2 drivers/usb/class/cdc-acm.c: need inactive config#2
PM: Adding info for usb:1-1.2:0 drivers/usb/class/cdc-acm.c: need inactive config#2

And this worked just fine in 2.6.0-test3?

It worked okay in 2.5.62, and probably in many later versions, but I'm not sure if I specificaly tested it with -test3.
Pavel

Signature · 2 lines
Horseback riding is like software...
...vgf orggre jura vgf serr.
#3 lore raw
Quoted 12 lines · Pavel Machek

Hi!

Quoted 10 lines
Quoted 9 lines

In 2.6.0-test4, USB ELSA modem no longer works. This is UHCI (on toshiba 4030cdt).

Relevant messages seem to be:

PM: Adding info for usb:1-1.2 drivers/usb/class/cdc-acm.c: need inactive config#2
PM: Adding info for usb:1-1.2:0 drivers/usb/class/cdc-acm.c: need inactive config#2

Quoted 4 lines
Quoted 1 line

And this worked just fine in 2.6.0-test3?

It worked okay in 2.5.62, and probably in many later versions, but I'm not sure if I specificaly tested it with -test3.

2.5.62??? You are going to have to help narrow it down a bit more :)

thanks,

greg k-h

#4 lore raw
Quoted 1 line · Greg KH

2.5.62??? You are going to have to help narrow it down a bit more :)

It worked fine in 2.5.75 too. I have the same problem as Pavel, with a different USB modem in -test4.

-- Jamie

#5 lore raw
Quoted 5 lines · Jamie Lokier
Quoted 1 line · Greg KH

2.5.62??? You are going to have to help narrow it down a bit more :)

It worked fine in 2.5.75 too. I have the same problem as Pavel, with a different USB modem in -test4.

Again, that was a while ago. Care to help narrow it down some?

thanks,

greg k-h

#6 lore raw
In reply to Jamie Lokier
Quoted 7 lines · Jamie Lokier
Quoted 1 line · Greg KH

2.5.62??? You are going to have to help narrow it down a bit more :)

It worked fine in 2.5.75 too. I have the same problem as Pavel, with a different USB modem in -test4.

And ... does my suggestion to Pavel then improve things?

http://marc.theaimsgroup.com/?l=linux-kernel&m=106263361810553&w=2

referring to this patch

http://marc.theaimsgroup.com/?l=linux-usb-devel&m=106260875713372&w=2

Same symptoms should in this case mean same fix ...

- Dave

#7 lore raw
Quoted 6 lines · David Brownell
Quoted 2 lines

It worked fine in 2.5.75 too. I have the same problem as Pavel, with a different USB modem in -test4.

And ... does my suggestion to Pavel then improve things?
http://marc.theaimsgroup.com/?l=linux-kernel&m=106263361810553&w=2
http://marc.theaimsgroup.com/?l=linux-usb-devel&m=106260875713372&w=2

Yes it does.

When loading the controller, these are some of the messages:

PM: Adding info for usb:1-1:0
PM: Adding info for usb:1-1:1 drivers/usb/class/cdc-acm.c: need inactive config #2 drivers/usb/class/cdc-acm.c: need inactive config #2

On loading cdc_acm:

usb 1-1: control timeout on ep0in

When I do the echo 2 >/sys/bus/usb/devices/1-1/bConfigurationValue:

acm: probe of 1-1:1 failed with error -5

I am able to open the modem device and use it after the echo, and not before. (I'm using it now).

Quoted 3 lines

I'll submit a cdc-acm patch later to get rid of the need for the manual workaround. And meanwhile, a hotplug script could automate this for you, in /etc/hotplug/usb/cdc-acm ...

So many other things don't work automatically for me in 2.6 that one little echo for cdc_acm is a little thing. Besides, hotplug doesn't work either - something about the arguments to /sbin/hotplug has changed since 2.4 and I am in no rush to install a new version.

Thanks for the modem fix. I can now test the futex bug without going offline :)

Cheers,
-- Jamie

#8 lore raw
Quoted 5 lines · Jamie Lokier

So many other things don't work automatically for me in 2.6 that one little echo for cdc_acm is a little thing. Besides, hotplug doesn't work either - something about the arguments to /sbin/hotplug has changed since 2.4 and I am in no rush to install a new version.

Sorry, but if you want hotplug to work in 2.6, you will have to install a new version due to some changes to the network arguments, and due to a bug in the older versions of the scripts.

thanks,

greg k-h

#9 lore raw
Quoted 10 lines · Greg KH
Quoted 5 lines · Jamie Lokier

So many other things don't work automatically for me in 2.6 that one little echo for cdc_acm is a little thing. Besides, hotplug doesn't work either - something about the arguments to /sbin/hotplug has changed since 2.4 and I am in no rush to install a new version.

Sorry, but if you want hotplug to work in 2.6, you will have to install a new version due to some changes to the network arguments, and due to a bug in the older versions of the scripts.

What release date should the hotplug scripts be? I still have that hotplug related oops that I told you about a while ago...

#10 lore raw
Quoted 8 lines · Mike Fedyk
Quoted 7 lines · Greg KH
Quoted 5 lines · Jamie Lokier

So many other things don't work automatically for me in 2.6 that one little echo for cdc_acm is a little thing. Besides, hotplug doesn't work either - something about the arguments to /sbin/hotplug has changed since 2.4 and I am in no rush to install a new version.

Quoted 6 lines
Quoted 3 lines

Sorry, but if you want hotplug to work in 2.6, you will have to install a new version due to some changes to the network arguments, and due to a bug in the older versions of the scripts.

What release date should the hotplug scripts be? I still have that hotplug related oops that I told you about a while ago...

Try the latest :)

But that oops should have nothing to do with the scripts, that's a kernel oops that I could never duplicate :(

thanks,

greg k-h

#11 lore raw
Quoted 13 lines · Greg KH
Quoted 12 lines · Mike Fedyk
Quoted 10 lines · Greg KH
Quoted 5 lines · Jamie Lokier

So many other things don't work automatically for me in 2.6 that one little echo for cdc_acm is a little thing. Besides, hotplug doesn't work either - something about the arguments to /sbin/hotplug has changed since 2.4 and I am in no rush to install a new version.

Sorry, but if you want hotplug to work in 2.6, you will have to install a new version due to some changes to the network arguments, and due to a bug in the older versions of the scripts.

Quoted 7 lines
Quoted 2 lines

What release date should the hotplug scripts be? I still have that hotplug related oops that I told you about a while ago...

Try the latest :)

But that oops should have nothing to do with the scripts, that's a kernel oops that I could never duplicate :(

Are the kksymoops reports parsable by the userspace ksymoops? Whenever I run the 2.6 kernel resolved oops through the user space ksymoops, it points to a different part of the kernel, leading me to suspect that i shouldn't be doing that.

Is that right? Also I use modules a lot, will that affect this?

I'd like to get this fixed. Any instructions / patches would be helpful.

#12 lore raw
In reply to Pavel Machek
Quoted 11 lines · Pavel Machek

Hi!

In 2.6.0-test4, USB ELSA modem no longer works. This is UHCI (on toshiba 4030cdt).

Relevant messages seem to be:

PM: Adding info for usb:1-1.2 drivers/usb/class/cdc-acm.c: need inactive config#2
PM: Adding info for usb:1-1.2:0 drivers/usb/class/cdc-acm.c: need inactive config#2

Try the usb_set_configuration() patch I just posted this morning (to linux-usb-devel).

Apply it, and you'll still get these messages ... but then you'll be able to

   # echo 2 > /sys/bus/usb/devices/1-1.2/bConfigurationValue

and then it should work as before. Granted that works as I expect, I'll submit a cdc-acm patch later to get rid of the need for the manual workaround. And meanwhile, a hotplug script could automate this for you, in /etc/hotplug/usb/cdc-acm ...

- Dave