7:36 a.m.
The #2 power consumer on my laptop according to powertop is
nic:virbr0 which seems unlikely. That's more than Firefox, and a bit
less than the display backlight. And it's nearly 300x more power than
used by the wifi driver.
https://pastebin.com/GuxA4Dsg
862 mW 0.0 pkts/s Device nic:virbr0
If it's wrong, it's a powertop bug. If it's real, then it's a bug
somewhere maybe in libvirt or maybe the kernel? There has been no VM
running during this boot, so nic:virbr0 shouldn't even be in use.
Is this the actual wifi radio, but somehow virb0 is just getting dinged?
--
Chris Murphy
7:48 a.m.
New subject: [Powertop] 700+mW for nic:virbr0?
On 8/30/2017 2:36 PM, Chris Murphy wrote:
The #2 power consumer on my laptop according to powertop is
nic:virbr0 which seems unlikely. That's more than Firefox, and a bit
less than the display backlight. And it's nearly 300x more power than
used by the wifi driver.
so powertop seems to correlate virbr activity with power consumption...
so while it may not be the virtual bridge itself to cause power consumption, the usage of
it very likely is
9:47 a.m.
New subject: [Powertop] 700+mW for nic:virbr0?
On Wed, Aug 30, 2017 at 3:48 PM, Arjan van de Ven <arjan(a)linux.intel.com> wrote:
On 8/30/2017 2:36 PM, Chris Murphy wrote:
>
> The #2 power consumer on my laptop according to powertop is
> nic:virbr0 which seems unlikely. That's more than Firefox, and a bit
> less than the display backlight. And it's nearly 300x more power than
> used by the wifi driver.
>
so powertop seems to correlate virbr activity with power consumption...
so while it may not be the virtual bridge itself to cause power consumption,
the usage of it very likely is
It varies a lot, even while connected to the same AP. 190mW up to 2W.
--
Chris Murphy
1845
days inactive
1845
days old
2 comments
2 participants
participants (2)
-
Arjan van de Ven -
Chris Murphy