On Tue, 2009-06-23 at 18:44 -0400, Paul Davis wrote:
On Tue, Jun 23, 2009 at 6:19 PM, Fernando
Lopez-Lezcano<nando(a)ccrma.stanford.edu> wrote:
that is hilarious :)
and
http://bugzilla.kernel.org/show_bug.cgi?id=10361
(referenced inside the previous ticket)
this is when it was apparently added to the Fedora kernels:
--------
* Fri Apr 18 2008 Kyle McMartin <kmcmartin(a)redhat.com> -
Enable CONFIG_RT_GROUP_SCHED (#442959)
--------
There's something else I'm missing because I'm still able to run rt as a
non-root user even though I have all this stuff. Is Fedora configuring
this somewhere else? Is this overriden by something else?
this seems deeply odd. i am also in that same situation. it seems that
something *must* be overriding it, but what?
I have no idea yet. Or maybe there is some other configuration that
needs to be done before it is activated.
Or.
Maybe it is not working at all and we are not getting rt, we have no
error message and have not noticed it :-) ;-P
the API looks entirely usable to me, in the sense that
i could easily
imagine a GUI control panel for this that would be comprehensible to
most users.
I think the idea is no control panel :-)
yeah, i note that in the second bug report there, peter z. makes some
comments about RT scheduling that are not entirely accurate for our
use case.
we are aiming at deterministic behaviour by each client in a JACK
graph, but we don't actually expect determinism across time (e.g. as
clients are started and stopped). we're happy to eat 95% of the CPU
time, even if we generally expect the typical figure to be more like
10%. this tends to force JACK towards a cgroup that has a perhaps
unreasonably large allocation, and a misleading one. AFAICT, the
Sum(all cgroup runtime) cannot exceed the rt sched period. So you
cannot set things up to allow "JACK" to get 95% and <other RT stuff>
to get 95% and just have stuff break if you try to do both. this is
good and bad - its an accurate reflection of resource reservation
policy, but it doesn't actually reflect what a regular user may want -
running two RT subsystems at different times without reconfiguring
their cgroups. of course, i have no idea what those two RT subsystems
might be, so its a bit moot.
I imagine configuring this so that all rt usage is in one bag if
possible (both root and nonroot processes are counted in the same bin).
There is no need in our usage case to differentiate, and in fact it is
impossible.
-- Fernando