[ANNOUNCE][RFC] PlugSched-6.1 for 2.6.13
From: Peter Williams (pwil3058_at_bigpond.net.au)
Date: 08/31/05
- Previous message: Bartlomiej Zolnierkiewicz: "Re: IDE HPA"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Wed, 31 Aug 2005 10:46:07 +1000 To: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
This version contains a modified spa_ws scheduler with a more persistent
bonus mechanism. Although the "bonus only at wake up" mechanism of the
original worked well on the first systems it was tested on (an old SMP
system and a 3GHz SMT system) subsequent tests on a 2GHz single
processor system were disappointing. (Apart from a bug in the original
implementation) the primary reason for this was that the X server was
not always able to complete its work in the first time slice after
waking and had to try and complete it without the benefit of the bonus
which causes obvious delays when the system is loaded. The new
mechanism is a simplification of the persistent interactive bonus
mechanism in zaphod.
A patch for 2.6.13 is available at:
<http://prdownloads.sourceforge.net/cpuse/plugsched-6.1-for-2.6.13.patch?download>
Very Brief Documentation:
You can select a default scheduler at kernel build time. If you wish to
boot with a scheduler other than the default it can be selected at boot
time by adding:
cpusched=<scheduler>
to the boot command line where <scheduler> is one of: ingosched,
nicksched, staircase, spa_no_frills, spa_ws or zaphod. If you don't
change the default when you build the kernel the default scheduler will
be ingosched (which is the normal scheduler).
The scheduler in force on a running system can be determined by the
contents of:
/proc/scheduler
Control parameters for the scheduler can be read/set via files in:
/sys/cpusched/<scheduler>/
Peter
-- Peter Williams pwil3058@bigpond.net.au "Learning, n. The kind of ignorance distinguishing the studious." -- Ambrose Bierce - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
- Previous message: Bartlomiej Zolnierkiewicz: "Re: IDE HPA"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
- Major 2.4 / 2.{5|6}-smp Scheduler Slow-ness
... Ok, first of all, I have never been able to run a 2.4 kernel on this ... much
so that this is a scheduler issue. ... When i finally let the 2.6.0-test2-mm1 boot
with smp support, ... send the line "unsubscribe linux-kernel" in ... (Linux-Kernel) - [ANNOUNCE][RFC] PlugSched-5.1 for 2.6.12-rc6 and 2.6.12-rc6-mm1
... The patch of PlugSched-5.1 for 2.6.12-rc5 applies cleanly to 2.6.12-rc6 ...
You can select a default scheduler at kernel build time. ... boot with a scheduler
other than the default it can be selected at boot ... send the line "unsubscribe linux-kernel"
in ... (Linux-Kernel) - Re: Strange IDE performance change in 2.6.1-rc1 (again)
... > Do you get different numbers if you boot with: ... Changing io scheduler
doesn't seem to affect performance too much... ... send the line "unsubscribe linux-kernel"
in ... Please read the FAQ at http://www.tux.org/lkml/ ... (Linux-Kernel) - Re: [RFC] CPU scheduler: Simplified interactive bonus mechanism
... >> So, as Con seems to imply, it is JASW (just another scheduler ... >>
rewrite). ... Or, for that matter, even the bonus mechanism itself ... >
other than the calculation of the bonus as the way the bonus is ... (Linux-Kernel) - [ANNOUNCE][RFC] PlugSched-5.2.1 for 2.6.11 and 2.6.12
... You can select a default scheduler at kernel build time. ... boot
with a scheduler other than the default it can be selected at boot ... send the line "unsubscribe
linux-kernel" in ... (Linux-Kernel)