Matthew Burgess wrote: > Hi guys, > > Archaic and I have put our heads together to try and come up with a more > reasonable set of Udev rules. These are based on the following criteria: > > 1) If a device needs packages outside those installed by LFS then don't > include a rule for it. (e.g. audio devices) > 2) If hardware is specific to a non-LFS targeted architecture then don't > include a rule for it (e.g. /dev/dasd - s/390 specific, apparently!) > 3) For other devices, do the absolute minimum necessary to configure the > device node sanely, bearing in mind that *every* device that udev finds > will have a device node created for it albeit with defaults of > NAME="%k", GROUP="root", MODE="0660" > > With that in mind, we'd appreciate feedback on the attached config file > especially if you've tested it "in the field" and found that we broke > something! Errors and omissions expected :)
Sorry to go against everything my BLFS editorial colleagues have said, but I rather like this proposal. I think it adds to the educational nature of both books. If a device exists and has no rule in the lfs set, then a node is created with the default permissions. This is good, since the user will learn, indeed be taught, how to write a rule to achieve their real objectives. This is far better than giving a rote answer, that may only be satisfactory in the context of writing BLFS. In the cases where the device node must be 'forced' because there is no /sys entry, then I say we should teach why, not just baldly tel the answer. I think the effort to put this into BLFS will be worth it in the long run. R. -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
