Ubuntu Should Automatically Mount usbfs At Boot Time

Bug #70968 reported by Forest Bond
8
Affects Status Importance Assigned to Milestone
Ubuntu
Invalid
Wishlist
Unassigned

Bug Description

In particular, midisport-firmware (and some other packages as well, I assume), depend on usbfs having been mounted. I am not aware of any disadvantage to doing this.

-Forest

Revision history for this message
Laurent Bigonville (bigon) wrote :

I think that usbfs in /proc is deprecated. Now ubuntu uses udev and the new location for usbfs is in /dev. See bug #30702

Revision history for this message
Andrew Ash (ash211) wrote :

What release are you using, Forest?

Revision history for this message
Forest Bond (forest-bond) wrote : Re: [Bug 70968] Re: Ubuntu Should Automatically Mount usbfs At Boot Time

On Thu, Nov 09, 2006 at 01:29:08AM -0000, Andrew Ash wrote:
> What release are you using, Forest?

edgy

I believe this worked fine in dapper.

I am the maintainer of midisport-firmware, and am happy to make changes to it if
it needs it; upstream is not extraordinarilly friendly, but I don't mind
managing a delta of some kind. Otherwise, I wasn't sure what package this bug
should be filed against.

-Forest

Revision history for this message
Daniel T Chen (crimsun) wrote :

Will need to consult Scott regarding whether this feature request is feasible for Feisty. In the meantime, I see no hitches with adding a section to README.Debian informing the user that

none /dev/bus/usb usbfs defaults 0 0

needs to be added to fstab(5).

Revision history for this message
Forest Bond (forest-bond) wrote :

On Thu, Nov 09, 2006 at 08:42:56AM -0000, Daniel T Chen wrote:
> Will need to consult Scott regarding whether this feature request is
> feasible for Feisty. In the meantime, I see no hitches with adding a
> section to README.Debian informing the user that
>
> none /dev/bus/usb usbfs defaults 0 0
>
> needs to be added to fstab(5).

Sounds good, I will attach a debdiff shortly.

-Forest

Revision history for this message
Forest Bond (forest-bond) wrote :

On Thu, Nov 09, 2006 at 08:42:56AM -0000, Daniel T Chen wrote:
> Will need to consult Scott regarding whether this feature request is
> feasible for Feisty.

Should I subscribe Scott to this bug, then?

-Forest

Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote :

I am confused by this bug ...

/proc/bus/usb is mounted at boot time, and contains the tree and devices nodes you would expect

Revision history for this message
Forest Bond (forest-bond) wrote :

Yes, apologies. This was misunderstood and consequently misreported. Just trying to get some action one way or another. :)

Closing.

thanks,
Forest

Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote :

Having looked at the package, your problem is actually that your udev rules are being run in the wrong place.

In order to be able to access the /proc/bus/usb/XXX/YYY device node, you need to run your program as a result of the usb_device subsystem, not the usb subsystem:

For edgy this would look like:

ACTION=="add", SUBSYSTEM=="usb_device", SYSFS{idVendor}=="0763", SYSFS{idProduct}=="1001", RUN+="@fxload@ -s @firmwaredir@/MidiSportLoader.ihx -I @firmwaredir@/MidiSport2x2.ihx"

And for feisty:

ACTION=="add", SUBSYSTEM=="usb_device", ATTR{idVendor}=="0763", ATTR{idProduct}=="1001", RUN+="@fxload@ -s @firmwaredir@/MidiSportLoader.ihx -I @firmwaredir@/MidiSport2x2.ihx"

Revision history for this message
Scott James Remnant (Canonical) (canonical-scott) wrote :

sorry, on feisty those should be ATTRS{...}==

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.