Re: Epic regression in throughput since v2.6.23

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
From: Nikos Chantziaras
Date: Tuesday, September 8, 2009 - 11:37 am

On 09/08/2009 03:57 PM, Serge Belyshev wrote:

It seems more people are starting to confirm this issue:

   http://foldingforum.org/viewtopic.php?f=44&t=11336

IMHO it's not *that* dramatic as some people there describe it ("Is it 
the holy grail?") but if something makes your desktop "smooth as silk" 
just like that, it might seem as a holy grail ;)  In any case, there 
clearly seems to be a performance problem with the mainline scheduler on 
many people's desktops that are being solved by BFS.
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Sun Sep 6, 8:38 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Mon Sep 7, 3:12 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Frederic Weisbecker, (Mon Sep 7, 4:01 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Arjan van de Ven, (Mon Sep 7, 7:40 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Michael Buesch, (Mon Sep 7, 11:51 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Pekka Pietikainen, (Mon Sep 7, 4:24 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Thomas Fjellstrom, (Mon Sep 7, 4:54 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Tue Sep 8, 12:19 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Tue Sep 8, 1:13 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Arjan van de Ven, (Tue Sep 8, 1:31 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Arjan van de Ven, (Tue Sep 8, 1:38 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Benjamin Herrenschmidt, (Tue Sep 8, 2:50 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Tue Sep 8, 3:13 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Tue Sep 8, 3:40 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Tue Sep 8, 4:30 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Juergen Beisert, (Tue Sep 8, 4:32 am)
Epic regression in throughput since v2.6.23, Serge Belyshev, (Tue Sep 8, 5:57 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Arjan van de Ven, (Tue Sep 8, 7:20 am)
Re: Epic regression in throughput since v2.6.23, Jesse Brandeburg, (Tue Sep 8, 10:47 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Tue Sep 8, 11:15 am)
Re: Epic regression in throughput since v2.6.23, Nikos Chantziaras, (Tue Sep 8, 11:20 am)
Re: Epic regression in throughput since v2.6.23, Nikos Chantziaras, (Tue Sep 8, 11:37 am)
Re: Epic regression in throughput since v2.6.23, Jeff Garzik, (Tue Sep 8, 12:00 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Tue Sep 8, 12:06 pm)
Re: Epic regression in throughput since v2.6.23, Serge Belyshev, (Tue Sep 8, 12:20 pm)
Re: Epic regression in throughput since v2.6.23, Jeff Garzik, (Tue Sep 8, 12:26 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Tue Sep 8, 2:28 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Tue Sep 8, 3:00 pm)
Re: Epic regression in throughput since v2.6.23, Serge Belyshev, (Tue Sep 8, 3:15 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Tue Sep 8, 3:53 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Tue Sep 8, 4:38 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Benjamin Herrenschmidt, (Tue Sep 8, 5:28 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Wed Sep 9, 1:34 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Wed Sep 9, 2:05 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Wed Sep 9, 2:40 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Benjamin Herrenschmidt, (Wed Sep 9, 2:53 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Wed Sep 9, 3:17 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Benjamin Herrenschmidt, (Wed Sep 9, 4:32 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Wed Sep 9, 4:52 am)
[tip:sched/core] sched: Turn off child_runs_first, tip-bot for Mike Gal ..., (Wed Sep 9, 8:37 am)
[tip:sched/core] sched: Re-tune the scheduler latency defa ..., tip-bot for Mike Gal ..., (Wed Sep 9, 8:37 am)
[tip:sched/core] sched: Keep kthreads at default priority, tip-bot for Mike Gal ..., (Wed Sep 9, 8:37 am)
Re: Epic regression in throughput since v2.6.23, Ingo Molnar, (Wed Sep 9, 8:52 am)
Re: [tip:sched/core] sched: Turn off child_runs_first, Theodore Tso, (Wed Sep 9, 10:57 am)
Re: [tip:sched/core] sched: Turn off child_runs_first, Ingo Molnar, (Wed Sep 9, 11:08 am)
Re: [tip:sched/core] sched: Turn off child_runs_first, Chris Friesen, (Wed Sep 9, 11:59 am)
Re: [tip:sched/core] sched: Turn off child_runs_first, Pavel Machek, (Wed Sep 9, 12:48 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Wed Sep 9, 1:12 pm)
Re: Epic regression in throughput since v2.6.23, Serge Belyshev, (Wed Sep 9, 1:49 pm)
Re: Epic regression in throughput since v2.6.23, Cory Fields, (Wed Sep 9, 2:23 pm)
Re: Epic regression in throughput since v2.6.23, Ingo Molnar, (Wed Sep 9, 11:53 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Peter Zijlstra, (Wed Sep 9, 11:55 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Mike Galbraith, (Thu Sep 10, 3:57 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Mike Galbraith, (Thu Sep 10, 4:21 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Mike Galbraith, (Thu Sep 10, 4:28 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Mike Galbraith, (Thu Sep 10, 4:42 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Steven Rostedt, (Thu Sep 10, 6:53 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Peter Zijlstra, (Thu Sep 10, 9:05 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Nikos Chantziaras, (Thu Sep 10, 10:53 am)
[tip:sched/core] sched: Disable NEW_FAIR_SLEEPERS for now, tip-bot for Ingo Molnar, (Thu Sep 10, 11:51 am)
[tip:sched/core] sched: Fix sched::sched_stat_wait tracepo ..., tip-bot for Ingo Molnar, (Thu Sep 10, 11:57 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Martin Steigerwald, (Thu Sep 10, 12:45 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Frederic Weisbecker, (Thu Sep 10, 1:25 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Martin Steigerwald, (Thu Sep 10, 1:39 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Martin Steigerwald, (Thu Sep 10, 2:19 pm)
Re: Epic regression in throughput since v2.6.23, Serge Belyshev, (Thu Sep 10, 4:23 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Benjamin Herrenschmidt, (Thu Sep 10, 6:36 pm)
Re: Epic regression in throughput since v2.6.23, Ingo Molnar, (Thu Sep 10, 11:10 pm)
Re: [crash, bisected] Re: clocksource: Resolve cpu hotplug ..., Martin Schwidefsky, (Fri Sep 11, 12:48 am)
Re: Epic regression in throughput since v2.6.23, Serge Belyshev, (Fri Sep 11, 1:55 am)
Re: [crash, bisected] Re: clocksource: Resolve cpu hotplug ..., Martin Schwidefsky, (Fri Sep 11, 6:33 am)
[tip:timers/core] clocksource: Resolve cpu hotplug dead lo ..., tip-bot for Martin S ..., (Fri Sep 11, 11:22 am)
Re: BFS vs. mainline scheduler benchmarks and measurements, Martin Steigerwald, (Sat Sep 12, 4:26 am)
Re: [tip:sched/core] sched: Re-tune the scheduler latency ..., Martin Steigerwald, (Sat Sep 12, 4:45 am)
Re: [tip:sched/core] sched: Keep kthreads at default priority, Martin Steigerwald, (Sat Sep 12, 4:48 am)
Re: Epic regression in throughput since v2.6.23, Serge Belyshev, (Sun Sep 13, 8:27 am)
Re: Epic regression in throughput since v2.6.23, Ingo Molnar, (Sun Sep 13, 8:47 am)
Re: Epic regression in throughput since v2.6.23, Mike Galbraith, (Sun Sep 13, 12:17 pm)
Re: Epic regression in throughput since v2.6.23, Mike Galbraith, (Sun Sep 13, 11:15 pm)
Phoronix CFS vs BFS bencharks, Nikos Chantziaras, (Mon Sep 14, 2:46 am)
Re: Phoronix CFS vs BFS bencharks, Mike Galbraith, (Mon Sep 14, 4:35 am)
Re: Phoronix CFS vs BFS bencharks, Mike Galbraith, (Mon Sep 14, 8:32 am)
Re: [crash, bisected] Re: clocksource: Resolve cpu hotplug ..., Martin Schwidefsky, (Mon Sep 14, 8:37 am)
Re: [crash, bisected] Re: clocksource: Resolve cpu hotplug ..., Martin Schwidefsky, (Mon Sep 14, 10:59 am)
Re: Phoronix CFS vs BFS bencharks, Marcin Letyns, (Mon Sep 14, 12:14 pm)
Re: Phoronix CFS vs BFS bencharks, Willy Tarreau, (Mon Sep 14, 1:49 pm)
Re: Phoronix CFS vs BFS bencharks, Mike Galbraith, (Tue Sep 15, 1:37 am)
Re: Epic regression in throughput since v2.6.23, Ingo Molnar, (Wed Sep 16, 12:45 pm)
Re: Epic regression in throughput since v2.6.23, Serge Belyshev, (Wed Sep 16, 4:18 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Benjamin Herrenschmidt, (Wed Sep 16, 6:29 pm)
[patchlet] Re: Epic regression in throughput since v2.6.23, Mike Galbraith, (Wed Sep 16, 9:55 pm)
Re: BFS vs. mainline scheduler benchmarks and measurements, Felix Fietkau, (Sat Sep 19, 11:43 am)