1 |
On 24/08/2017 00:29, R0b0t1 wrote: |
2 |
> As an example, I am interested in characterizing the power consumption |
3 |
> of rendering a PDF document. I would hopefully only need to run the |
4 |
> renderer once. |
5 |
> |
6 |
> I can use PowerTOP, but it seems to be limited to rough measurements |
7 |
> on the order of tenths of a watt. This measurement can be divided |
8 |
> among the wakeup events in an attempt to calculate software power |
9 |
> consumption but it seems imperfect if I want to monitor a single |
10 |
> process that may be competing relatively equally for resources with |
11 |
> the kernel and other user processes. |
12 |
> |
13 |
> PMBus is a spinoff of SMBus which is a spinoff of I2C which is found |
14 |
> on many motherboards. PMBus is supposed to be the interface which |
15 |
> controls and reports power supply activity. Besides the main kW power |
16 |
> supply, there is usually a power supply near your processor that steps |
17 |
> down 3.3V or 5V to 2.8V, 1.8V, or lower (I've seen as low as 0.8V, but |
18 |
> not on a desktop). I was not aware these had a visible interface. |
19 |
> |
20 |
> Apparently you can talk to these, but my searches can only find code |
21 |
> which seems highly experimental. The other replies seem to be for |
22 |
> embedded Linux systems running on FPGAs and perhaps Cortex-A parts. |
23 |
> |
24 |
> If I were using a microcontroller I could get uA or nA draw per MHz |
25 |
> and I know my operating voltage and operating time. However, desktop |
26 |
> processors are much more complex, and I am not sure if they have been |
27 |
> entirely characterized. The most advanced tool I can find is PowerTOP |
28 |
> and it does not seem very accurate. |
29 |
> |
30 |
> Does anyone have any suggestions? Should I start reading source code |
31 |
> or post on the forums? |
32 |
|
33 |
Both of these sound good |
34 |
|
35 |
> Or perhaps someone has used PowerTOP and found |
36 |
> it to be reasonably accurate? |
37 |
|
38 |
No not this. PowerTOP was designed to find badly-behaving programs like |
39 |
pidgin that woke up and polled it's queue every 1ms or so. It's not for |
40 |
what you want at all, not even close. |
41 |
|
42 |
|
43 |
> |
44 |
> R0b0t1. |
45 |
> |
46 |
|
47 |
|
48 |
-- |
49 |
Alan McKinnon |
50 |
alan.mckinnon@×××××.com |