On Fri, Sep 22, 2006 at 05:55:24PM +0600, Alexander E. Patrakov wrote:
> >I wasn't sure if the disk was actually
> >on the "ancestor" chain of the partition, so I left it as-is instead of
> >converting to ATTRS{removable}.
> >
> >It should probably be changed to the upstream version though.
>
> Test and report.
Will do, if I have any devices that show up with ATTR{removable}=="1".
I can probably tell from udevinfo whether the whole-disk device is a
parent of the partition device, though. (And from Kay's response on
linux-hotplug-devel, it sounds like it is.)
> >Do you know if ATTRS{ieee1394_id}=="*" matches if the ieee1394_id
> >attribute is missing entirely?
>
> It doesn't.
That's good to know, thanks. So do you think it should be OK to leave
it as "*", to match any device that has an _id attribute, even if that
attribute is empty?
> Anyway, Kay already gave his opinion on
> linux-hotplug-devel: "KERNEL=="hd*[!0-9]", ATTRS{removable}=="1" could
> be ATTR I think, cause it will not run for a partition anyway cause of
> the KERNEL match."
Yep, I saw that. I'll change that later today.
> >As far as 95-udev-late goes: That's just for udevmonitor, correct? I
> >think we should probably still have it, just wondering what uses it.
>
> udevmonitor is a debugging tool shipped with udev.
Yes, and I think we should have a rule to enable it. Just wondering if
anything else uses that Unix-domain socket though. (I'm guessing not;
from the name, it seems pretty much udevmonitor-specific.)
pgpZnkvpsBMRa.pgp
Description: PGP signature
-- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
