Re: [Announce] [patch] Modular Scheduler Core and Completely Fair Scheduler [CFS]

[Ingo Molnar]

* Bill Huey <billh@xxxxxxxxxxxxxxxxx> wrote:

> Con has been asking for module support for years if I understand your 
> patch corectly. [...]

Yeah. Note that there are some subtle but crutial differences between 
PlugSched (which Con used, and which i opposed in the past) and this 
approach.

PlugSched cuts the interfaces at a high level in a monolithic way and 
introduces kernel/scheduler.c that uses one pluggable scheduler 
(represented via the 'scheduler' global template) at a time.

while in this CFS patchset i'm using modularization ('scheduler 
classes') to simplify the _existing_ multi-policy implementation of the 
scheduler. These 'scheduler classes' are in a hierarchy and are stacked 
on top of each other. They are in use at once. Currently there's two of 
them: sched_ops_rt is stacked ontop of sched_ops_fair. Fortunately the 
performance impact is minimal.

So scheduler classes are mainly a simplification of the design of the 
scheduler - not just a mere facility to select multiple schedulers. 
Their ability to also facilitate easier experimentation with schedulers 
is 'just' a happy side-effect. So all in one: it's a fairly different 
model from PlugSched (and that's why i didnt reuse PlugSched) - but 
there's indeed overlap.

> [...] You'll also need this for -rt as well with regards to bandwidth 
> scheduling.

yeah.

scheduler classes are also useful for other purposes like containers and 
virtualization, hierarchical/group scheduling, security encapsulation, 
etc. - features that can be on-demand layered, and which we dont 
necessarily want to have enabled all the time.

> [...] Good to see that you're moving in this direction.

thanks! :)

	Ingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/