[Nut-upsdev] nut + usb + udev with kernel 2.6.22

[Arjen de Korte]

Gilad Arnold wrote:

>>> SUBSYSTEM=="usb_device", GOTO="nut-usbups_rules_real"
>>> SUBSYSTEM=="usb", GOTO="nut-usbups_rules_real"
>>> GOTO="nut-usbups_rules_end"
>> Well, I suppose Arnaud had a reason to include the BUS here, so just
>> stripping this doesn't seem to be the right thing to do. Since it
>> doesn't hurt to have it, I still don't grasp the reason why it should
>> be removed.
> I thought we concluded that there is no known reason to compare with the
> BUS variable;

I concluded that there are no objections against comparing to it if it
doesn't exist, not that it should be removed.

> there is no evidence where it is needed, nor is it
> addressed by current udev rules (which is not reported as a problem).

Since Arnaud came up with this, I suppose he had a reason to do so. He
has not replied, so until he states that it was a mistake, I'll take his
word for it that there *is* a valid reason to include it. Maybe not for
Linux, but there could be other architectures where this is valid.

> Additionally, I thought you suggested not to mix 'equals' with
> 'not-equals' guards (which I agree with).

I still do. I would prefer to see positive logic here (BUS=="usb) and
the fall through default to a GOTO line as suggested.

> Hence, I concluded that pruning the 'BUS equals' clause would solve
both.  Did I get it wrong?

You got it almost right.

Best regards, Arjen