On 03/11/2015 12:08 PM, Juri Lelli wrote:
Hi Daniel,
On 11/03/15 10:08, Daniel Lezcano wrote:
On 03/09/2015 03:06 PM, Juri Lelli wrote:
From: Mike Turquette mturquette@linaro.org
Building on top of the scale invariant capacity patches and earlier patches in this series that prepare CFS for scaling cpu frequency, this patch implements a simple, naive ondemand-like cpu frequency scaling policy that is driven by enqueue_task_fair and dequeue_tassk_fair. This new policy is named "energy_model" as an homage to the on-going work in that area. It is NOT an actual energy model.
Well, I would change the name to something more trivial, eg: load-driven or whatever. Introducing 'energy model' here has no sense from my pov except than buzzing :)
Or something like cpufreq_scheduler with "scheduler" as the new policy name. This thing has to be driven by the scheduler after all, and we might change our minds in the future about which signal we use to actually drive it.
Yes, why not.
[ ... ]
Not-signed-off-by: Mike Turquette mturquette@linaro.org
Signed-off-by: Michael Turquette mturquette@linaro.org
Please remove the Not-signed-off which means "I don't testify this is open source code" [1].
[1] https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/Documen...
And Juri, as you are in the delivery path, you should add your signed-off-by.
Oh, sure. Didn't added/removed anything here as this is simply a re-post with a little delta of Mike's original patchset. This wasn't mean for LKML. It's just a way to restart discussion. Thanks for your feedback anyway. I guess I'll let Mike answer your points as he might already completely reworked this code :).
Ok, thanks
-- Daniel
-- http://www.linaro.org/ Linaro.org │ Open source software for ARM SoCs
Follow Linaro: http://www.facebook.com/pages/Linaro Facebook | http://twitter.com/#!/linaroorg Twitter | http://www.linaro.org/linaro-blog/ Blog