opcm

Thanks a lot for the suggestion. We are looking into it

I am still trying to find a reviewer for your patch

this should be possible by writing a script that checks `lspci -n` to find `IMC_DEV_IDS` (from cpucounter.cpp) and then read the [tcdbp, tcrap, tcrwp, tcothp, tcrfp registers](https://www.intel.com/content/dam/www/public/us/en/documents/datasheets/xeon-e5-v2-datasheet-vol-2.pdf) from bus/device/function offset...

@ilia-kurakin any idea if pcm-iio can be extended to support 8+ socket systems?

thanks for the patch, this is useful. But to reduce the overhead for non-interactive mode when sampling with a higher frequency the flushing should be avoided in csv mode. The...

that is a good idea. Please let me know if you would like to send a pull request making this change Roman

The instruction count (156.010) does not match the order of the iteration count (10.000.000). Likely the thread was migrated? Could you try to pin the thread using https://man7.org/linux/man-pages/man3/pthread_setaffinity_np.3.html

I think the difference is noise-level here.. E.g. for instructions retired it is 0.06%

cc @jakub-d

Generally running two pcm tools in parallel is not supported. There are exceptions but they are not documented (and not guaranteed to be supported in future). If you have a...