pcm-memory on an 8-socket machine

[wy140408]

"We have compiled the Windows version of PCM for performance monitoring, but when running pcm-memory on an 8-socket machine, it always fails. The same tool works fine on a 2-socket Windows system. Please help analyze where the issue might be."

Detailed logs are as follows:
DEBUG: Setting Ctrl+C done.

 Intel(r) Performance Counter Monitor: Memory Bandwidth Monitoring Utility (2025-11-01 14:21:33 +0100 ID=0f3bdfcf)

 This utility measures memory bandwidth per channel or per DIMM rank in real-time

===== Processor information =====

Hybrid processor : no

IBRS and IBPB supported : yes

STIBP supported : yes

Spec arch caps supported : yes

Max CPUID level : 32

CPU family : 6

CPU model number : 143

Number of physical cores: 80

Number of logical cores: 80

Number of online logical cores: 80

Threads (logical cores) per physical core: 1

Num sockets: 8

Physical cores per socket: 10

Last level cache slices per socket: 60

Core PMU (perfmon) version: 5

Number of core PMU generic (programmable) counters: 8

Width of generic (programmable) counters: 48 bits

Number of core PMU fixed counters: 4

Width of fixed counters: 48 bits

Nominal core frequency: 1900000000 Hz

IBRS enabled in the kernel : no

STIBP enabled in the kernel : no

The processor is not susceptible to Rogue Data Cache Load: yes

The processor supports enhanced IBRS : yes

Package thermal spec power: 350 Watt; Package minimum power: 206 Watt; Package maximum power: 764 Watt;

PCM warning: total_sockets_ 8 does not match socket2M2Mbus.size() 4

PCM error: Did not find UPI perfmon device on every socket in a multisocket system.

ERROR: UPI LL monitoring device (0:ffffffff:1:1) is missing. The UPI statistics will be incomplete or missing.

ERROR: UPI LL monitoring device (0:ffffffff:2:1) is missing. The UPI statistics will be incomplete or missing.

Socket 0: 4 memory controllers detected with total number of 8 channels. 0 UPI ports detected. 4 M2M (mesh to memory)/B2CMI blocks detected. 0 HBM M2M blocks detected. 0 EDC/HBM channels detected. 0 Home Agents detected. 0 M3UPI/B2UPI blocks detected.

PCM warning: total_sockets_ 8 does not match socket2M2Mbus.size() 4

PCM error: Did not find UPI perfmon device on every socket in a multisocket system.

ERROR: UPI LL monitoring device (0:ffffffff:1:1) is missing. The UPI statistics will be incomplete or missing.

ERROR: UPI LL monitoring device (0:ffffffff:2:1) is missing. The UPI statistics will be incomplete or missing.

Socket 1: 4 memory controllers detected with total number of 8 channels. 0 UPI ports detected. 4 M2M (mesh to memory)/B2CMI blocks detected. 0 HBM M2M blocks detected. 0 EDC/HBM channels detected. 0 Home Agents detected. 0 M3UPI/B2UPI blocks detected.

PCM warning: total_sockets_ 8 does not match socket2M2Mbus.size() 4

PCM error: Did not find UPI perfmon device on every socket in a multisocket system.

ERROR: UPI LL monitoring device (0:ffffffff:1:1) is missing. The UPI statistics will be incomplete or missing.

ERROR: UPI LL monitoring device (0:ffffffff:2:1) is missing. The UPI statistics will be incomplete or missing.

Socket 2: 4 memory controllers detected with total number of 8 channels. 0 UPI ports detected. 4 M2M (mesh to memory)/B2CMI blocks detected. 0 HBM M2M blocks detected. 0 EDC/HBM channels detected. 0 Home Agents detected. 0 M3UPI/B2UPI blocks detected.

PCM warning: total_sockets_ 8 does not match socket2M2Mbus.size() 4

PCM error: Did not find UPI perfmon device on every socket in a multisocket system.

ERROR: UPI LL monitoring device (0:ffffffff:1:1) is missing. The UPI statistics will be incomplete or missing.

ERROR: UPI LL monitoring device (0:ffffffff:2:1) is missing. The UPI statistics will be incomplete or missing.

Socket 3: 4 memory controllers detected with total number of 8 channels. 0 UPI ports detected. 4 M2M (mesh to memory)/B2CMI blocks detected. 0 HBM M2M blocks detected. 0 EDC/HBM channels detected. 0 Home Agents detected. 0 M3UPI/B2UPI blocks detected.

PCM error: socket_ 4 >= socket2M2Mbus.size() 4

PCM warning: total_sockets_ 8 does not match socket2M2Mbus.size() 4

PCM error: Did not find UPI perfmon device on every socket in a multisocket system.

PCM error: no memory controllers found.

Can not access server uncore PCI configuration space. Access to uncore counters (memory and QPI bandwidth) is disabled.

You must have signed msr.sys driver in your current directory and have administrator rights to run this program.

WARNING: PCU PMU not found

WARNING: MDF PMU not found

WARNING: PCU PMU not found

WARNING: MDF PMU not found

WARNING: PCU PMU not found

WARNING: MDF PMU not found

WARNING: PCU PMU not found

WARNING: MDF PMU not found

Info: IDX - Detected 16 IAA devices, 16 DSA devices, 16 QAT devices.

Socket 0: 1 PCU units detected. 12 IIO units detected. 12 IRP units detected. 60 CHA/CBO units detected. 40 MDF units detected. 1 UBOX units detected. 0 CXL units detected. 0 PCIE_GEN5x16 units detected. 0 PCIE_GEN5x8 units detected.

Socket 1: 1 PCU units detected. 12 IIO units detected. 12 IRP units detected. 60 CHA/CBO units detected. 40 MDF units detected. 1 UBOX units detected. 0 CXL units detected. 0 PCIE_GEN5x16 units detected. 0 PCIE_GEN5x8 units detected.

Socket 2: 1 PCU units detected. 12 IIO units detected. 12 IRP units detected. 60 CHA/CBO units detected. 40 MDF units detected. 1 UBOX units detected. 0 CXL units detected. 0 PCIE_GEN5x16 units detected. 0 PCIE_GEN5x8 units detected.

Socket 3: 1 PCU units detected. 12 IIO units detected. 12 IRP units detected. 60 CHA/CBO units detected. 40 MDF units detected. 1 UBOX units detected. 0 CXL units detected. 0 PCIE_GEN5x16 units detected. 0 PCIE_GEN5x8 units detected.

Socket 4: 0 PCU units detected. 12 IIO units detected. 12 IRP units detected. 60 CHA/CBO units detected. 0 MDF units detected. 1 UBOX units detected. 0 CXL units detected. 0 PCIE_GEN5x16 units detected. 0 PCIE_GEN5x8 units detected.

Socket 5: 0 PCU units detected. 12 IIO units detected. 12 IRP units detected. 60 CHA/CBO units detected. 0 MDF units detected. 1 UBOX units detected. 0 CXL units detected. 0 PCIE_GEN5x16 units detected. 0 PCIE_GEN5x8 units detected.

Socket 6: 0 PCU units detected. 12 IIO units detected. 12 IRP units detected. 60 CHA/CBO units detected. 0 MDF units detected. 1 UBOX units detected. 0 CXL units detected. 0 PCIE_GEN5x16 units detected. 0 PCIE_GEN5x8 units detected.

Socket 7: 0 PCU units detected. 12 IIO units detected. 12 IRP units detected. 60 CHA/CBO units detected. 0 MDF units detected. 1 UBOX units detected. 0 CXL units detected. 0 PCIE_GEN5x16 units detected. 0 PCIE_GEN5x8 units detected.

Initializing RMIDs

Detected Intel(R) Xeon(R) Platinum 8490H "Intel(r) microarchitecture codename Sapphire Rapids-SP" stepping 6 microcode level 0x2b0004d0

Access to Intel(r) Performance Counter Monitor has denied (no MSR or PCI CFG space access).: operation not permitted

Cleaning up

[rdementi]

Hi, are you using msr.sys or Winring0 driver?

[wy140408]

I am using msr.sys

[rdementi]

thanks. Could you please try this version (https://github.com/intel/pcm/archive/refs/heads/SPR-8S-Windows.zip) and let us know if it works for you.

[wy140408]

I am using the tool SPR-8S-Windows.zip. After normal compilation, it always prompts the following error:

Intel(R) Performance Counter Monitor (2025-11-11 07:19:31 +0100 ID=2584ae27)
Starting MSR service failed with error 2. Trying to load winring0.dll/winring0.sys driver...
Failed to load winring0.dll/winring0.sys driver.
Access to Intel(r) Performance Counter Monitor has been denied (no MSR or PCI CFG space access).

After encountering the above error, I executed the command pcm --uninstallDriver and rebooted the machine, but the error still occurs

[rdementi]

Hi, error 2 means "The system cannot find the file specified". Meaning msr.sys driver cannot be found. Is msr.sys still in the current and the same directory as pcm exe files? The error cannot be due to the change in the "SPR-8S-Windows" branch as it did not change anything related to the driver. The issue is unrelated and it is strange that msr.sys worked before for you and now it cannot be loaded... Please check.

[wy140408]

After recompiling the tool and registering the msr.sys driver, the tool is able to monitor properly now. Thank you!

[rdementi]

you are welcome!

[rdementi]

the change has been merged into the master branch