* William Lee Irwin III <wli@xxxxxxxxxxxxxx> wrote: >> [...] sched_yield() semantics are yet another twist. On Wed, May 23, 2007 at 08:40:35PM +0200, Ingo Molnar wrote: > that's nonsense, sched_yield() semantics are totally uninteresting. It > is a fundamentally broken interface. They're not totally uninteresting. People will complain when apps break or perform poorly due to changes in its semantics. As an interface it may be poor and worse yet poorly specified, but it has non-negligible effects on performance issues that can't be ignored and that will remain the case for the foreseeable future. The content of my comment was that the patch does something to sched_yield() semantics, so it raises the question of what will happen in benchmarks and other performance affairs that are sensitive to sched_yield() semantics changes. -- wli _______________________________________________ Containers mailing list Containers@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/containers
