On Mon, Nov 18, 2013 at 03:21:49PM -0500, Daniel Hollocher wrote: > So again, how could this be "fixed" in Ubuntu? If there are hardware > specific optimal values for sampling_down_factor, how would that get > included into Ubuntu?
Am I right in understanding that this value can be changed at runtime? Or is it a kernel compile-time option? If runtime, then presumably there would have to be some userspace logic that detects what your system is and maps that to a preferred value. Perhaps it would examine /proc/cpuinfo or run dmidecode or something. Is this accurate? If so, then would a sensible first step be to implement this command in a userspace package? AFAICS this command would be the hard part, but doesn't require thought about integration, since it would be a separate piece anyway. Then the command might be extended to adjust the tunable itself with an option, and then the packaging might be extended to add an init script that does this on startup. -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
