(In reply to Tobias Klausmann from comment #248)
> I for one have a stable system *without* disabling C6, but just
> rcu_nocbs=0-15 on the kernel command line. If C6 was hard-disabled, it
> wouldn't help me, but likely make the power consumption a lot worse.
I tested power consumption with C6 package enabled (default) or disabled. I disabled it via patch for zenstates.py.
I ran two tests over 2 hours with disabled VMs and no KDE session (just the login screen started, but active was text console 0).
Based on the measuring tolerance of the power meter, I couldn't find any difference between enabled or disabled C6 package.
The voltages I saw in sensors have been the same (most of the time 0.39 V) - no difference. The load of the machine was 0 0 0.
Maybe the patch for zenstates.py to disable C6 package doesn't work? It would have been nice to have the possibility to re enable it.
Another nice finding was: the consumption of my 4 idle(!) VMs is constantly about 5 Watt!
(In reply to Tobias Klausmann from comment #248)
> I for one have a stable system *without* disabling C6, but just
> rcu_nocbs=0-15 on the kernel command line. If C6 was hard-disabled, it
> wouldn't help me, but likely make the power consumption a lot worse.
I tested power consumption with C6 package enabled (default) or disabled. I disabled it via patch for zenstates.py.
I ran two tests over 2 hours with disabled VMs and no KDE session (just the login screen started, but active was text console 0).
Based on the measuring tolerance of the power meter, I couldn't find any difference between enabled or disabled C6 package.
The voltages I saw in sensors have been the same (most of the time 0.39 V) - no difference. The load of the machine was 0 0 0.
Maybe the patch for zenstates.py to disable C6 package doesn't work? It would have been nice to have the possibility to re enable it.
Another nice finding was: the consumption of my 4 idle(!) VMs is constantly about 5 Watt!