It is not clear to me if this is a dup of bug 985661 or not. Some discussion goes beyond just reported load averages. As for the reported load averages part someone above said: "So somebody introduced a patch because the computed load averages were too low - and now they are way too high." Actually, the reported load averages, under many conditions were massively too low, actually 0.0. Now, apparently, under "idle" conditions on desktops reported load averages can be somewhat too high. Compounding the issue is that people are comparing to kernels without the patch, which would report load averages that are too low. As someone noted above proper comparitvie results can be obatined by compiling the kernel with CONFIG_NO_HZ=no. Two people above listed output from the "top" command which can also be used to estimate the real load average. For the first one (the OP) I got 0.17 and for the second one (post # 40) I got 0.53 (I realize it is rediculous to state them to two decimal places). Over the entire operational space of all loads and all idle enter/exit frequencies and all number of processes, I still think the reported load averages are better than they were. It is not clear to me that there is a better solution within the current context of how load averages are calculated with tickless kernels.
-- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/995284 Title: High idle load on 64bit Ubuntu 12.04 with 3.2.0-24-generic kernel To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/995284/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
