Future powertop development?

[DagB]

Hi.

I was wondering about the future of powertop.
I thought 2.0 was expected 8 months or so ago.
The development tree pointed to at lesswatts.org is no more (moblin.org
serves a 502), and the relevant git tree at kernel.org is empty.

Seeing how useful powertop is, I`d find it seriously odd if this tool
has been killed by management.
Good to see that Arjan still posts to the list from an intel mailaddress.


Dag B

[Kok, Auke-jan H]

Post by DagB
I was wondering about the future of powertop.
I thought 2.0 was expected 8 months or so ago.
The development tree pointed to at lesswatts.org is no more (moblin.org
serves a 502), and the relevant git tree at kernel.org is empty.
Seeing how useful powertop is, I`d find it seriously odd if this tool
has been killed by management.
Good to see that Arjan still posts to the list from an intel mailaddress.

on the contrary, things are being worked on, just read the list,

the git tree moved to github, just track back on the mailinglist archive.

Cheers,

Auke

[Arjan van de Ven]

Post by DagB
Hi.
I was wondering about the future of powertop.
I thought 2.0 was expected 8 months or so ago.
The development tree pointed to at lesswatts.org is no more (moblin.org
serves a 502), and the relevant git tree at kernel.org is empty.
Seeing how useful powertop is, I`d find it seriously odd if this tool
has been killed by management.
Good to see that Arjan still posts to the list from an intel mailaddress.

we moved for now to github;
http://github.com/fenrus75/powertop


and I've recently managed to convinced "management" to get a few more
resources on the project ....

[DagB]

Post by Arjan van de Ven

Post by DagB
Hi.
I was wondering about the future of powertop.
I thought 2.0 was expected 8 months or so ago.
The development tree pointed to at lesswatts.org is no more (moblin.org
serves a 502), and the relevant git tree at kernel.org is empty.
Seeing how useful powertop is, I`d find it seriously odd if this tool
has been killed by management.
Good to see that Arjan still posts to the list from an intel mailaddress.

we moved for now to github;
http://github.com/fenrus75/powertop
and I've recently managed to convinced "management" to get a few more
resources on the project ....

Great news!

I just tested the code in git on my macbook air. (4,1)
A few remarks on usability and some questions:


On startup for the first time, powertop says:
Cannot load from file /var/cache/powertop/saved_results.powertop
Cannot load from file /var/cache/powertop/saved_parameters.powertop
which is expected, I guess. I'd suggest adding: 'Creating it.'
Although saved_parameters.powertop *isn't* created, unless I 'touch' it
first?


'powertop --calibrate' requires access to :0, and powertop does not
complain if/when a laptop is plugged into AC.
The intended users of powertop easily understands these things and
remedies them. Others may be slightly confused I guess.


'powertop --calibrate' dims the panel. And I waited a looong while for
it to come back on. I finally gave up and increased brightness, just to
see powertop running in its regular mode. I assume DPMS kicked in
somewhere. :-)
Depending on what you want to measure, it may be an idea to disable
DPMS, or maybe just wake up the panel when the calibration loop is done?
Not sure what is doable, and how difficult it is. Perhaps this works
exactly as you want it?


The calibrating backlight thing... how is that affected if DPMS is
active? As I said, my panel went dark. I can't tell if powertop actually
adjusted the backlight during calibration.


At the end of the calibration run, powertop states:
Score: 90.9 (1950924.6)
Guess: 100.2
Actual: 6.8
Care to enlighten me as to what this means?


dmesg tells me:
ieee80211 phy0: brcms_ops_config: change power-save mode: false (implement)

This is clearly unrelated to powertop, but I wonder if this actually
means that the driver for my wifi card is missing power management?



Thanks,

Dag B