Hi, I'm trying to set up DCI OOB with Intel System Studio 2020 (target CPU i9-9900K, host CPU i5-1035G1, both running Linux). I can establish a DCI connection both via System Studio and via ipccli (green light from the CCA, the CPU microarchitecture and stepping is recognised correctly, and I can enumerate a range of devices with ipccli), but in both cases I get an error that there are "no available CPU threads" when I try to use System Trace or the System Debugger. When I try to run an ipccli command affecting the CPU cores like ipc.halt() I get the error message "There are no GPC cores available to halt". Does anyone have any ideas what could be causing this - is there perhaps another BIOS setting I need to change, or is it more likely to be an OpenIPC configuration problem? I'm using the default 9th Gen Intel Core Processor (Coffee Lake S Refresh) / Cannon Lake PCH-H configuration.
These are the changes I've made so far to enable DCI: on the target, IA32_DEBUG_INTERFACE reports it's enabled, and HDCIEN is also set. I've set PCH Trace Hub Enable Mode to Host/Target Debugger respectively on the two systems, and both have Platform Debug Consent set to 0x1 (DCI OOB+[DbC]). I've also enabled CPU Run Control and Intel Enhanced Debug on the target, and disabled CPU Run Control Lock. I haven't set the CPU Trace Hub Enable Mode for the i9-9900K, as it wasn't present in the UEFI Setup variables.
This is perhaps because you enumerate some devices in CPU, but not "cores". Your processor must be powered on during establishing DCI connection in System Studio. What devices are shown in System Studio Console after "Found TAPs:"?