BLFS Trac wrote:
> #1680: /proc/bus/usb is obsolete
> Comment (by [EMAIL PROTECTED]):
>
> There is a conflict between the LFS and BLFS rules. LFS already names USB
> devices (basically the first rule but with NAME=... instead of SYMLINK),
> thus, the BLFS rule should be:
>
> SUBSYSTEM=="usb_device", GROUP="usb"
>
> But, that's insecure. The "usb" group must die (it was a hack invented by
> me in order to avoid the dependency of SANE on Hotplug). Rules of the
> second kind (matching on vendor and product IDs are already a part of SANE
> package, but are not installed in BLFS (that's a bug). For libgphoto2,
> there is a Debian-specific patch that adds a generator for such rules.
Alexander,
I'm getting frustrated here. Which is it? Should libusb have a udev
rule or not? Is the rule above the complete rule?
The rule now is
SUBSYSTEM=="usb_device", \
PROGRAM="/bin/sh -c 'X=%k X=$${X#usbdev} B=$${X%%%%.*} D=$${X#*.}; \
echo bus/usb/$$B/$$D'", SYMLINK+="%c", GROUP="usb"
which *you* originally put into the ticket.
I put the rule at /etc/udev/rules.d/23-usb-rules, which comes before the
25-lfs-rules. My understanding, admittedly incomplete, is that the
first rule encountered will be executed and others skipped. Is this not
true?
Secondly, this is a ticket about libusb. Don't start introducing issues
with libgphoto2 and sane here. If there are multiple issues to be
discussed, lets do it on blfs-dev.
> As for the /proc/bus/usb mount point, it would be nice to list an example
> of an application that does require it: VMware.
Done but not yet committed.
-- Bruce
--
http://linuxfromscratch.org/mailman/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
