Tear and I spent some time tonight mapping the output of "TurionPowerControl -temp" to the physical cpu sockets on SuperMicro and Tyan quad g34 motherboards.
The Tyan S8812WGM3NR mapping is different than you might expect. I verified the IPMI sensor readings map correctly to the physical CPU socket labels. This data was collected and is valid for 6100 series processors. The TPC output maps as follows:
Tear tested the SuperMicro H8QGi+-F. The results are what I would have expected. This data was collected and is valid for 6100 series processors. The mapping is as follows:
UPDATE 2/7/14: brilong has tested a SuperMicro GL board with 6200 series processors. His results are:
** I used TurionPowerControl v0.40a in my testing.
The Tyan S8812WGM3NR mapping is different than you might expect. I verified the IPMI sensor readings map correctly to the physical CPU socket labels. This data was collected and is valid for 6100 series processors. The TPC output maps as follows:
Code:
[22:40:23] <&firedfly> cpu0 is node 0-1
[22:40:31] <&firedfly> cpu1 is node 4-5
[22:40:38] <&firedfly> cpu2 is node 6-7
[22:40:46] <&firedfly> cpu3 is node 2-3
Tear tested the SuperMicro H8QGi+-F. The results are what I would have expected. This data was collected and is valid for 6100 series processors. The mapping is as follows:
Code:
[22:47:15] <&tear> cpu1 is node 0-1
[22:47:19] <&tear> cpu2 is node 2-3
[22:47:25] <&tear> cpu3 is node 4-5
[22:47:32] <&tear> cpu4 is node 6-7
UPDATE 2/7/14: brilong has tested a SuperMicro GL board with 6200 series processors. His results are:
Code:
CPU1: node0/1
CPU2: node4/5
CPU3: node2/3
CPU4: node6/7
** I used TurionPowerControl v0.40a in my testing.
Last edited: