Strange FID & VID - Values

If you have trouble with PHC you can ask and hope for help here.
Ian
Posts: 1
Joined: Sun 23. Nov 2008, 05:42

Re: Strange FID & VID - Values

Post by Ian » Sun 23. Nov 2008, 05:50

Hey,

I'm running a C2D T8100 and getting similar values:

Code: Select all

$ cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies
2101000 2100000 1600000 1200000 800000
$ cat /sys/devices/system/cpu/cpu0/cpufreq/phc_controls
75:40 74:34 8:28 6:23 136:19  
And phctool agrees with scaling_available_frequencies.

Powertop lists my speeds as 2.11/2.10/1.6/1.2/0.8GHz

Let me know if i can be of any assistance testing.

the-fallen
Administrator
Posts: 346
Joined: Wed 9. Jul 2008, 19:57

Re: Strange FID & VID - Values

Post by the-fallen » Sun 23. Nov 2008, 08:29

Can one of you manually switch to one of the frequencies with a "strange value" and send me the read_msr output?

Spindoctor
Posts: 13
Joined: Thu 20. Nov 2008, 03:41

Re: Strange FID & VID - Values

Post by Spindoctor » Sun 23. Nov 2008, 14:00

I can only select 3 of the 4 frequencies (at least in the visual frequency-selection tool of gnome there are onle 3 frequencies listed).

These frequencies are 2.534Ghz, 1.6Ghz and 800Ghz.

Which of them would you like to see?

the-fallen
Administrator
Posts: 346
Joined: Wed 9. Jul 2008, 19:57

Re: Strange FID & VID - Values

Post by the-fallen » Sun 23. Nov 2008, 16:53

Highest and lowest please. Thanks for the support :)

Spindoctor
Posts: 13
Joined: Thu 20. Nov 2008, 03:41

Re: Strange FID & VID - Values

Post by Spindoctor » Sun 23. Nov 2008, 19:49

okay but one more question:

how do I get the read_msr output?

Sorry, I'm not very familiar with that kind of thing.

Spindoctor
Posts: 13
Joined: Thu 20. Nov 2008, 03:41

Re: Strange FID & VID - Values

Post by Spindoctor » Sun 23. Nov 2008, 20:17

Sorry, got it already.

800 Mhz:

Code: Select all

Displaying data for CPU cpu0
/---------------------------------------
| Current VID: 15
| Current FID: 134
| Target VID:  15
| Target FID:  134
| Highest VID: 36
| Highest FID: 73
| Lowest VID:  23
| Lowest FID:  6
\---------------------------------------



Displaying data for CPU cpu1
/---------------------------------------
| Current VID: 15
| Current FID: 134
| Target VID:  15
| Target FID:  134
| Highest VID: 43
| Highest FID: 10
| Lowest VID:  23
| Lowest FID:  6
\---------------------------------------
1600 Mhz:

Code: Select all

Displaying data for CPU cpu0
/---------------------------------------
| Current VID: 23
| Current FID: 6
| Target VID:  23
| Target FID:  6
| Highest VID: 36
| Highest FID: 73
| Lowest VID:  23
| Lowest FID:  6
\---------------------------------------



Displaying data for CPU cpu1
/---------------------------------------
| Current VID: 23
| Current FID: 6
| Target VID:  23
| Target FID:  6
| Highest VID: 36
| Highest FID: 73
| Lowest VID:  23
| Lowest FID:  6
\---------------------------------------
2534 Mhz:

Code: Select all

Displaying data for CPU cpu0
/---------------------------------------
| Current VID: 36
| Current FID: 73
| Target VID:  36
| Target FID:  73
| Highest VID: 36
| Highest FID: 73
| Lowest VID:  23
| Lowest FID:  6
\---------------------------------------



Displaying data for CPU cpu1
/---------------------------------------
| Current VID: 36
| Current FID: 73
| Target VID:  36
| Target FID:  73
| Highest VID: 36
| Highest FID: 73
| Lowest VID:  23
| Lowest FID:  6
\---------------------------------------

the-fallen
Administrator
Posts: 346
Joined: Wed 9. Jul 2008, 19:57

Re: Strange FID & VID - Values

Post by the-fallen » Sun 23. Nov 2008, 21:10

Okay so those "strange" Values are really in the MSR's. This means it is not an bug in the phc patches (thats good).
I do not know the reason why they look that strange since those FIDs should be some kind of multiplier to the FSB.

I hope we will find out what's the meaning of those "jumps" but until then ... I do not think it will be a problem using phc on those CPUs. I just need to remember that when writing on PHCTool2


Thanks for your help.

Post Reply