#10454: New scheduler: substantial performance drop -----------------------------+---------------------------- Reporter: stippi | Owner: pdziepak Type: task | Status: in-progress Priority: high | Milestone: R1 Component: System/Kernel | Version: R1/Development Resolution: | Keywords: Blocked By: 10487 | Blocking: Has a Patch: 0 | Platform: All -----------------------------+---------------------------- Comment (by stippi): It definitely improved things quite a bit (second and third runs): {{{ real 1m5.878s user 2m25.134s sys 0m41.105s real 1m5.652s user 2m25.156s sys 0m40.971s }}} The effect is that load is spread across all cores much more consistently. However, there are some curious effects still to be observed. When the compile test begins, after jam collected targets, load jumps to between 80% and 90% on all cores, but slowly decreases over the duration of the test. It could be dependent on what exactly is compiling, as there is a short dip of load at a consistent time during the compile. I am attaching a screenshot of the Activity Monitor graphs. -- Ticket URL: <https://dev.haiku-os.org/ticket/10454#comment:13> Haiku <https://dev.haiku-os.org> Haiku - the operating system.