#7606: Stride scheduling for Haskell threads with priorities ---------------------------------+------------------------------------------ Reporter: ezyang | Owner: ezyang Type: feature request | Status: new Priority: normal | Milestone: 7.8.1 Component: Runtime System | Version: 7.7 Keywords: | Os: Unknown/Multiple Architecture: Unknown/Multiple | Failure: None/Unknown Difficulty: Unknown | Testcase: Blockedby: | Blocking: Related: | ---------------------------------+------------------------------------------ Comment(by simonmar): Looking good, I'm still excited about this work. I really hope you'll put some of your benchmarks into `nofib/smp` - the current set may well be giving us a skewed view of what's important. I definitely agree that we want latency benchmarks too, but I haven't thought about how to construct them. Have you thought about keeping the red-black tree nodes in the heap? That's what I would do. Heap allocation is much faster than malloc, and reclamation is free. You would get some advantages for free, such as when parts of the tree are in the old generation they won't get touched during a minor GC. -- Ticket URL: http://hackage.haskell.org/trac/ghc/ticket/7606#comment:28 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler