nvidia-settings multi-gpu temp query broken
I was using the command `nvidia-settings -q [gpu:$1]/GPUCoreTemp` where $1 was the 0, 1, or 2 depending on the GPU I was interested in to get the temperature of that gpu, and at some point, I think a few updates ago to be honest, it stopped working for gpus 1 and 2, and just displayed 0. The nvidia-settings gui still shows the temperatures ok. The attached .txt has the full output of nvidia-settings -q all (minus some data I thought was irrelevant to make the 100 KB size limit.) Grepping GPUCoreTemp will show you the 0 data. Interestingly, ThermalSensorReading (query-able by thermalsensor) seems to be working correctly, but I'm not positive that will always give me one reading per gpu or that the gpu and thermal sensor number will always match. Any help would be appreciated, I'm assuming this is a bug in the driver / nvidia-settings software.
Arch Linux 64 bit. Current X.org version 1.10.1, driver version 270.41.03.