El lun, 20-06-2005 a las 06:12, Sampo Savolainen
escribió:
Quoting Christoph Eckert
<mchristoph.eckert(a)t-online.de>de>:
We all would be glad if we didn't need to
patch and recompile
the kernel. BUt there's hope that we get realtime
capabilities in the vanilla kernel soon. Let's stay tuned.
You are a bit late actually. :)
From 2.6.12 changelog:
"commit e43379f10b42194b8a6e1de342cfb44463c0f6da
Author: Matt Mackall <mpm(a)selenic.com>
Date: Sun May 1 08:59:00 2005 -0700
[PATCH] nice and rt-prio rlimits
Add a pair of rlimits for allowing non-root tasks to raise nice and rt
priorities. Defaults to traditional behavior. Originally written by
Chris Wright.
The patch implements a simple rlimit ceiling for the RT (and nice)
priorities
a task can set. The rlimit defaults to 0, meaning no change in behavior by
default. A value of 50 means RT priority levels 1-50 are allowed. A
value of
100 means all 99 privilege levels from 1 to 99 are allowed. CAP_SYS_NICE is
blanket permission.
(akpm: see
http://www.uwsg.iu.edu/hypermail/linux/kernel/0503.1/1921.html for
tips on integrating this with PAM).
Signed-off-by: Matt Mackall <mpm(a)selenic.com>
Acked-by: Ingo Molnar <mingo(a)elte.hu>
Signed-off-by: Andrew Morton <akpm(a)osdl.org>
Signed-off-by: Linus Torvalds <torvalds(a)osdl.org>"
Sampo Savolainen
Does this means that I can download the kernel source, compile it
without _any_ patch and have a "multimedia kernel" (aka a suitable
kernel for doing realtime audio) in my machine?
If so, damn, I'm downloading it!
Hmmm, not really. This patch gives _access_ to realtime scheduling to
non-root users, provided that PAM in your distribution (Pluggable
Authentication Modules) is patched to recognize the extra options, or
you have a small program that would allow you to use this if PAM is not
use by your distro.
This (access to realtime scheduling) is of course vital, you don't need
to do things as root. It serves the same purpose as and "officially"
replaces the realtime lsm kernel module.
This patch by itself does not provide for good low latency behavior.
2.6.12 should be pretty decent in that respect but to really get good
latencies you would have to add the realtime preempt patch by Ingo
Molnar[1]
-- Fernando
[1]