This version contains the following changes:
1. upgrade the staircase scheduler to version 11.6,
2. major reorganization of the spa_no_frills and zaphod code to clean up
the mechanism by which they share code. This includes the definition of
a simple interface which can be used to extend spa_no_frills to create
new schedulers relatively easily, and
3. a new scheduler, spa_ws, which is an extension of spa_no_frills
intended for use on work stations.
The spa_ws aspires to provide good interactive responsiveness and glitch
free audio/video playing by giving tasks that are forking or waking from
a sleep a "latency bonus". The latency bonus is composed of a small
fixed bonus plus a component based on the tasks sleepiness. Certain
tasks (e.g. those with a "nice" value greater than zero or tagged with
the TASK_NONINTERACTIVE flag do not receive any bonus as it is assumed
that latency is not an issue for these tasks). This scheduler tries to
identify audio/video tasks based on the regularity of their wake
intervals and their length (currently between 10 and 100 milliseconds)
and if it does so gives them a bonus one greater than the maximum that
could be attained for other tasks subject to their cpu usage rate being
less than a specified minimum. The mechanisms involved are very simple
in an attempt to not degrade the compute performance of the scheduler.
A patch for 2.6.12 is available at:
<http://prdownloads.sourceforge.net/cpuse/plugsched-6.0-for-2.6.12.patch?download>
and for 2.6.13-rc6 at:
<http://prdownloads.sourceforge.net/cpuse/plugsched-6.0-for-2.6.13-rc6.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 [email protected]
"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 [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
| |
 |