On Sun, Nov 18, 2012 at 07:06:04PM -0800, Greg KH wrote > Again, udev isn't the problem here. It hasn't broken the standalone > /usr issue at all.
systemd-udev supporters have an "interesting" definition of broken. "I plead not guilty to vandalism your honour. The complainant's window has actually been broken for several years. The stone I threw through it merely pointed out the existing brokeness." > There isn't anything in udev to change for this. I don't understand > why you are thinking that udev has anything to do with this issue > at all. Before version 181, udev booted with a separate /usr. As of 181, it doesn't. If anything, I would argue that udev 181 was deliberately broken. The fact is, udev made new - and insane - rules that are simply *invalid*. Modern udev is broken, and needs to be fixed. > It's other packages that are the problem here. You mean like systemd? When udev got rolled into the systemd tarball, and started sharing code with systemd, it also inherited its restrictions and separate-/usr-brokeness. And yes, I'm aware of "bluetooth keyboard drivers". As I said in my previous message, there have always been a few edge cases that require a pre-mounted /usr to boot properly. What I'm complaining about is that the other 99% of udev users are now gratuitously forced to share the pain of the edge case users. -- Walter Dnes <waltd...@waltdnes.org> We are apparently better off trying to avoid udev like the plague. Linus Torvalds; 2012/10/03 https://lkml.org/lkml/2012/10/3/349
