“In a continued thread about how the recently merged Completely
Fair Scheduler affects the nice command, Ingo Molnar offered a
history of nice levels in the Linux kernel. He began by describing
the three most frequent complaints he has received, first was ‘nice
levels were always so weak under Linux that people continuously
bugged me about making nice +19 tasks use up much less CPU time,’
second was ‘the fact that nice level behavior depended on the
_absolute_ nice level as well, while the nice API itself is
fundamentally ‘relative’,’ and third was ‘negative nice levels were
not ‘punchy enough,’ so lots of people had to resort to run audio
(and other multimedia) apps under RT priorities such as
SCHED_FIFO…'”
Related Story:
Linux: CFS
and Nice(Jul 19, 2007)