On Wed, Nov 15, 2017 at 2:02 PM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote: >> Internally, we use the priority map that the kernel >> uses, e.g. [0..99] for real-time policies and [100..139] >> for the SCHED_NORMAL/SCHED_BATCH policies. > > I will break that without consideration if I have to. That really isn't > something anybody outside of the core code should rely on. I mostly used these because it makes it easier to debug, since that range is used in other places in the kernel (and in trace events). All priority calculations use things that are in sched headers, like NICE_TO_PRIO, PRIO_TO_NICE, and MAX_USER_RT_PRIO. So things wouldn't necessarily break if you just changed the value ranges. If you inverted the range, that would be a problem... _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
