Bryan Kadzban wrote:
> Ken Moffat wrote:
>> On Thu, Jun 07, 2012 at 04:21:05PM -0500, Bruce Dubbs wrote:
>>> $ ls -l /lib/udev
>>> total 1528
>>> -rwxr-xr-x 1 root root 196379 May 21 15:41 accelerometer
>>> -rwxr-xr-x 1 root root  91401 May 21 15:41 ata_id
>>> -rwxr-xr-x 1 root root 138252 May 21 15:41 cdrom_id
>>> -rwxr-xr-x 1 root root  85791 May 21 15:41 collect
>>> drwxr-xr-x 3 root root   4096 May 15 02:24 devices
>>> -rwxr-xr-x 1 root root   2613 May 21 15:41 findkeyboards
>>> -rwxr-xr-x 1 root root    625 May 21 15:41 keyboard-force-release.sh
>>> -rwxr-xr-x 1 root root  72705 May 21 15:41 keymap
>>> drwxr-xr-x 3 root root   4096 May 21 15:41 keymaps
>>> -rwxr-xr-x 1 root root  15826 May 21 15:41 mtd_probe
>>> -rw-r--r-- 1 root root   3481 May 15 02:24 rule_generator.functions
>>> drwxr-xr-x 2 root root   4096 May 25 22:05 rules.d
>>> -rwxr-xr-x 1 root root 139696 May 21 15:41 scsi_id
>>> -rwxr-xr-x 1 root root 760509 May 21 15:41 udevd
>>> -rwxr-xr-x 1 root root  12484 May 21 15:41 v4l_id
>>> -rwxr-xr-x 1 root root   3595 May 15 02:24 write_cd_rules
>>> -rwxr-xr-x 1 root root   4384 May 15 02:24 write_net_rules
>>>
>>  I think that's a fuller install than what we've been doing up to
>> now in the LFS book.  On 7.1 I don't have findkeyboards, nor any of
>> the key* files.
> 
> --disable-keymap, as we do in -182, turns them all off.  :-)
> 
> Sadly it doesn't look like upstream likes the patch as-is. Not sure if
> they'll take the version that the William Hubbs (the gentoo guy from the
> linux-hotplug thread) and I are going to put together, which has various
> fragments split out from Makefile.am, and then enabled / disabled
> individually.
> 
> We *could* apply the patch, then autoreconf, then regenerate a new
> patch, for LFS (to avoid pkg-config or other weirdness causing issues in
> chapter 6's autoreconf run). Or we could give it a while (leaving the
> book at udev-182) and see what happens.  Any preferences?

My first choice is to leave it at udev-182.  I'm not sure if my second 
choice is to use the make.sh script, create a custom Makefile by hand, 
or create a patch for configure.am.  In any of these other choices, we 
will need to create a patch to add back things like specifying the 
devices to be copied to /dev at boot time or for network/cdrom devices.

It strikes me that adjusting the patch for configure.ac/Makefile.am for 
each new systemd release is the most problematical choice, especially if 
we don't get cooperation from upstream.

   -- Bruce


-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page

Reply via email to