On Wed, Jul 11, 2007 at 12:44:42PM -0700, Paul Menage wrote:
> >I'm inclined to take the cautious route here - I don't think people will be
> >dying for the CFS thingy (which I didn't even know about?) in .23, and it's
> >rather a lot of infrastructure to add for a CPU scheduler configurator
>
> S
On 7/10/07, Andrew Morton <[EMAIL PROTECTED]> wrote:
I'm inclined to take the cautious route here - I don't think people will be
dying for the CFS thingy (which I didn't even know about?) in .23, and it's
rather a lot of infrastructure to add for a CPU scheduler configurator
Selecting the rele
On Wed, 2007-07-11 at 10:03 -0700, Paul Jackson wrote:
> Peter wrote:
> > I'd very much like to have it, but that is just me.
>
> Why? [linear combinations of uid, container, pid, pgrp weighting]
Good question, and I really have no other answer than that it seems
usefull and not impossible (or ev
Peter wrote:
> I'd very much like to have it, but that is just me.
Why? [linear combinations of uid, container, pid, pgrp weighting]
You provide some implementation details and complications, but no
motivation that I noticed.
Well ... a little motivation ... "just me", which would go a long
way
On Wed, Jul 11, 2007 at 01:30:40PM +0200, Peter Zijlstra wrote:
> > One thing to think on though, we cannot have per process,uid,gid,pgrp
> > scheduling for one release only. So we'd have to manage interaction with
> > process containers. It might be that a simple weight multiplication
> > scheme i
On Wed, Jul 11, 2007 at 05:09:53PM +0530, Srivatsa Vaddagiri wrote:
> > (btw., if this goes into 2.6.23 then we cannot possibly turn it off in
> > 2.6.24,
>
> The fact that we will have two interface for group scheduler in 2.6.24
> is what worries me a bit (one user-id based and other container b
On Wed, 2007-07-11 at 04:42 -0700, Paul Jackson wrote:
> Srivatsa wrote:
> > The fact that we will have two interface for group scheduler in 2.6.24
> > is what worries me a bit (one user-id based and other container based).
>
> Yeah.
>
> One -could- take linear combinations, as Peter drew in his
Srivatsa wrote:
> The fact that we will have two interface for group scheduler in 2.6.24
> is what worries me a bit (one user-id based and other container based).
Yeah.
One -could- take linear combinations, as Peter drew in his ascii art,
but would one -want- to do that?
I imagine some future ti
On Wed, 2007-07-11 at 13:24 +0200, Peter Zijlstra wrote:
> On Wed, 2007-07-11 at 04:10 -0700, Paul Jackson wrote:
> > Srivatsa wrote:
> > > So Ingo was proposing we use cpuset as that user interface to manage
> > > task-groups. This will be only for 2.6.23.
> >
> > Good explanation - thanks.
> >
On Wed, Jul 11, 2007 at 12:19:58PM +0200, Ingo Molnar wrote:
> * Srivatsa Vaddagiri <[EMAIL PROTECTED]> wrote:
>
> > The other alternative is to hook up group scheduler with user-id's
> > (again only for 2.6.23).
>
> could you just try this and send an as simple patch as possible? This is
> act
On Wed, 2007-07-11 at 04:10 -0700, Paul Jackson wrote:
> Srivatsa wrote:
> > So Ingo was proposing we use cpuset as that user interface to manage
> > task-groups. This will be only for 2.6.23.
>
> Good explanation - thanks.
>
> In short, the proposal was to use the task partition defined by cpuse
Srivatsa wrote:
> So Ingo was proposing we use cpuset as that user interface to manage
> task-groups. This will be only for 2.6.23.
Good explanation - thanks.
In short, the proposal was to use the task partition defined by cpusets
to define CFS task-groups, until the real process containers are
a
* Srivatsa Vaddagiri <[EMAIL PROTECTED]> wrote:
> The other alternative is to hook up group scheduler with user-id's
> (again only for 2.6.23).
could you just try this and send an as simple patch as possible? This is
actually something that non-container people would be interested in as
well.
On Wed, Jul 11, 2007 at 02:23:52AM -0700, Paul Jackson wrote:
> Ingo wrote:
> > another option would be to trivially hook up CONFIG_FAIR_GROUP_SCHED
> > with cpusets, ...
>
> ah ... you triggered my procmail filter for 'cpuset' ... ;).
:-)
> What would it mean to hook up CFS with cpusets?
CFS
Ingo wrote:
> another option would be to trivially hook up CONFIG_FAIR_GROUP_SCHED
> with cpusets, ...
ah ... you triggered my procmail filter for 'cpuset' ... ;).
What would it mean to hook up CFS with cpusets? I've a pretty
good idea what a cpuset is, but don't know what kind of purpose
you h
* Andrew Morton <[EMAIL PROTECTED]> wrote:
> > c. Enable group scheduling bits only in -mm for now (2.6.23-mmX), using
> >Paul's container patches. I can send you a short patch that hooks up
> >cfs group scheduler with Paul's container infrastructure.
> >
> > If a. is not po
On Tue, Jul 10, 2007 at 10:29:42PM -0700, Andrew Morton wrote:
> I'm inclined to take the cautious route here - I don't think people will be
> dying for the CFS thingy (which I didn't even know about?) in .23, and it's
> rather a lot of infrastructure to add for a CPU scheduler configurator
> gadge
On Wed, 11 Jul 2007 10:25:16 +0530 Srivatsa Vaddagiri <[EMAIL PROTECTED]> wrote:
> On Tue, Jul 10, 2007 at 11:53:19AM -0700, Andrew Morton wrote:
> > On Tue, 10 Jul 2007 11:34:38 -0700
> > "Paul Menage" <[EMAIL PROTECTED]> wrote:
> >
> > > Andrew, how about we merge enough of the container framew
On Tue, Jul 10, 2007 at 11:53:19AM -0700, Andrew Morton wrote:
> On Tue, 10 Jul 2007 11:34:38 -0700
> "Paul Menage" <[EMAIL PROTECTED]> wrote:
>
> > Andrew, how about we merge enough of the container framework to
> > support CFS? Bits we could leave out for now include container_clone()
> > suppor
On 7/10/07, Andrew Morton <[EMAIL PROTECTED]> wrote:
> Andrew, how about we merge enough of the container framework to
> support CFS? Bits we could leave out for now include container_clone()
> support and the nsproxy subsystem, fork/exit callback hooks, and
> possibly leave cpusets alone for now
On Tue, 10 Jul 2007 11:34:38 -0700
"Paul Menage" <[EMAIL PROTECTED]> wrote:
> Andrew, how about we merge enough of the container framework to
> support CFS? Bits we could leave out for now include container_clone()
> support and the nsproxy subsystem, fork/exit callback hooks, and
> possibly leave
On 7/10/07, Srivatsa Vaddagiri <[EMAIL PROTECTED]> wrote:
>
> Container stuff. Hold, I guess. I was expecting updates from Paul.
Paul,
Are you working on a new version? I thought it was mostly ready
for mainline.
There are definitely some big changes that I want to make internally
* Srivatsa Vaddagiri <[EMAIL PROTECTED]> wrote:
> On Tue, Jul 10, 2007 at 01:31:52AM -0700, Andrew Morton wrote:
> > cpuset-zero-malloc-revert-the-old-cpuset-fix.patch
> > containersv10-basic-container-framework.patch
> > containersv10-basic-container-framework-fix.patch
> > containersv10-basic-c
On Tue, Jul 10, 2007 at 01:31:52AM -0700, Andrew Morton wrote:
> cpuset-zero-malloc-revert-the-old-cpuset-fix.patch
> containersv10-basic-container-framework.patch
> containersv10-basic-container-framework-fix.patch
> containersv10-basic-container-framework-fix-2.patch
> containersv10-basic-contain
24 matches
Mail list logo
