On Sun, 2010-03-07 at 15:48 +0100, Daniel Lezcano wrote: : - Big snip...
> At this point, the udev events for the network devices should be > implemented to send events only for the container where the network > devices belong. > And then generalize the udev events for all the containers. > When udev per container will be implemented, I will be curious to see a > "good" lxc.cgroup.devices configuration :) But you've got a huge problem there anyways that I've recognized right from the very get go. Those are based on major and minor numbers and the whole precept of udev (and devfs before it) is that those should no longer be needed to be fixed. There are several drivers that are both in the primary kernel source tree, such as my own "ip2" multiport serial interface, and outside of the tree, such as x10dev, that use HUGE numbers of devices that really just do not NEED a fixed major number. The Intelliport II driver (ip2) has three entire major numbers assigned to it from the dark ancient days of the 1.x kernel! In a fixed allocation scheme, it needed them! It supports up to 256 high speed serial ports (all the minors that were available in a major) and needed both raw and cooked interfaces (there go two majors) and a couple of per driver and per board control interfaces (and there goes the third major). Even though I have never run into a single installation that was fully decked out with 4 boards (64 interfaces per board - I have seen ONE that had THREE and that was scary). Now days, I could reduce that to 1 major (move the controls to sysfs and the raw/cooked to an ioctl and/or sysfs) and float that. The majors are valuable resources. I've actually tried to give back one of those majors (the raw/cooked) and it really can't be done for legacy reasons even as late as this, now. But how will this LXC config with majors and minors play with drivers that may have different major numbers and, heavens, how do you deal with all these USB devices and potentially varying minor numbers. You can't, really. Not with the current configuration scheme, you just can't. Mike -- Michael H. Warfield (AI4NB) | (770) 985-6132 | m...@wittsend.com /\/\|=mhw=|\/\/ | (678) 463-0932 | http://www.wittsend.com/mhw/ NIC whois: MHW9 | An optimist believes we live in the best of all PGP Key: 0x674627FF | possible worlds. A pessimist is sure of it!
signature.asc
Description: This is a digitally signed message part
------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev
_______________________________________________ Lxc-users mailing list Lxc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lxc-users
