On Sun, Apr 4, 2010 at 5:35 PM, Mark Mielke <m...@mark.mielke.cc> wrote:
> Optimally - both counts should be available. How many tasks can I run > simultaneously at near 100% CPU vs how many tasks can I run simultaneously > at any % of the CPU. > > A task doing video encoding or anything that would monopolize much of the > CPU, would probably use the first count, whereas something doing heavy I/O > and quick transfers of data from one thread to the other might choose the > latter count. group scheduling makes this *much* more complex than even your description alludes to. add in threads running SCHED_FIFO or SCHED_RR and simple-minded mapping of *any* possible CPU count (virtual, physical, group size, other) to a thread count becomes incredibly hard to do, and very application specific. naturally, if you have a fairly simply application, and always run on the whole machine, then the issues are not so great. but as a general solution a lot of what has been proposed is way too simplistic. --p _______________________________________________ gtk-devel-list mailing list gtk-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-devel-list
