For some reason the P-Core data is not showing on my 12900k on the matrix example.
I'm using V-Tune straight "out of the box" just running the example program as shown. However, in Threading it does show that all 16 (physical) cores are being being targeted as well as saying that on the console when using the example.
I have used V-Tune on my laptop which is also alderlake and cannot replicate the error.
Could there be a security feature stopping the drivers from being able to see the cpu event counters soemwhere?
Link Copied
Hi Jaideep,
I can confirm that my machine is not runnign in a virtual machine.
I have tried from another application I've designed and i still face the same issues. Below I have added a screenshot fromk within VS2022 and as you can see it's the same.
I have ran the self checker and it all seems to be okay, I have attached a screenshot below of the results.
As a note, I did have this issue many months ago but after reinstalling Windows it went away and reported fine on both P and E cores. Now however it has come back, and after reattempting to reinstall windows the issue has persisted. None of my other Alderlake machines experiance the same issues as this however. Here is a screenshot of my system info also:
Oliver
I have exactly the same issue! And looks like this post is also the same issue
My system:
CPU: i5 12400
OS: RHEL8
When looking at "Event Count" tab, it seems that most counters are present/correctly incremented, but some counters like "Clockticks" are stuck at 0. I'm including an image of the Event Count tab, and also I'm including a CSV file with Event counts.
When running the demo matrix example from command line:
[kevin@localhost bin64]$ ./vtune -collect uarch-exploration -r '/home/kevin/Documents/vtune_test' /home/kevin/intel/vtune/samples/matrix/matrix
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
vtune: Collection started. To stop the collection, either press CTRL-C or enter from another console window: vtune -r /home/kevin/Documents/vtune_test -command stop.
Addr of buf1 = 0x7fe3f4f61010
Offs of buf1 = 0x7fe3f4f61180
Addr of buf2 = 0x7fe3f2f60010
Offs of buf2 = 0x7fe3f2f601c0
Addr of buf3 = 0x7fe3f0f5f010
Offs of buf3 = 0x7fe3f0f5f100
Addr of buf4 = 0x7fe3eef5e010
Offs of buf4 = 0x7fe3eef5e140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 15.122 seconds
vtune: Collection stopped.
vtune: Using result path `/home/kevin/Documents/vtune_test'
vtune: Executing actions 19 % Resolving information for `ld-2.28.so'
vtune: Warning: Cannot locate file `nft_counter.ko'.
vtune: Executing actions 19 % Resolving information for `nft_counter'
vtune: Warning: Cannot locate file `bridge.ko'.
vtune: Executing actions 19 % Resolving information for `matrix'
vtune: Warning: Cannot locate file `r8169.ko'.
vtune: Executing actions 19 % Resolving information for `r8169'
vtune: Warning: Cannot locate file `nf_tables.ko'.
vtune: Executing actions 19 % Resolving information for `nf_tables'
vtune: Warning: Cannot locate file `nf_conntrack.ko'.
vtune: Executing actions 21 % Resolving information for `libc-2.28.so'
vtune: Warning: Cannot locate file `nft_ct.ko'.
vtune: Executing actions 21 % Resolving information for `nft_ct'
vtune: Warning: Cannot locate debugging information for file `/usr/lib64/libc-2.28.so'.
vtune: Warning: Cannot locate file `drm.ko'.
vtune: Executing actions 21 % Resolving information for `drm'
vtune: Warning: Cannot locate file `kvm.ko'.
vtune: Executing actions 22 % Resolving information for `kvm'
vtune: Warning: Cannot locate file `xfs.ko'.
vtune: Executing actions 22 % Resolving information for `xfs'
vtune: Warning: Cannot locate file `i915.ko'.
vtune: Executing actions 22 % Resolving information for `i915'
vtune: Warning: Cannot locate file `sep5.ko'.
vtune: Executing actions 22 % Resolving information for `vmlinux'
vtune: Warning: Function and source-level analysis for the Linux kernel will not be possible since neither debug version of the kernel nor kernel symbol tables are found. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
vtune: Executing actions 75 % Generating a report Elapsed Time: 15.135s
Clockticks: 0
P-Core: 0
E-Core: 0
Instructions Retired: 69,798,000,000
P-Core: 0
E-Core: 0
CPI Rate: 0.000
P-Core: 0.000
E-Core: 0.000
MUX Reliability: 0.000
| Precision of collected HW event data is not enough. Metrics data may be
| unreliable. Consider increasing your application execution time, using
| the multiple runs mode instead of event multiplexing, or creating a
| custom analysis with a limited subset of HW events. If you are using a
| driverless collection, consider reducing the value of
| /sys/bus/event_source/devices/cpu/perf_event_mux_interval_ms file.
|
P-Core
Retiring: 0.0% of Pipeline Slots
Light Operations: 0.0% of Pipeline Slots
FP Arithmetic: 0.0% of uOps
FP x87: 0.0% of uOps
FP Scalar: 0.0% of uOps
FP Vector: 0.0% of uOps
Memory Operations: 0.0% of Pipeline Slots
Fused Instructions: 0.0% of Pipeline Slots
Non Fused Branches: 0.0% of Pipeline Slots
Nop Instructions: 0.0% of Pipeline Slots
Other: 0.0% of Pipeline Slots
Heavy Operations: 0.0% of Pipeline Slots
Microcode Sequencer: 0.0% of Pipeline Slots
Assists: 0.0% of Pipeline Slots
CISC: 0.0% of Pipeline Slots
Front-End Bound: 0.0% of Pipeline Slots
Front-End Latency: 0.0% of Pipeline Slots
ICache Misses: 0.0% of Clockticks
ITLB Overhead: 0.0% of Clockticks
Branch Resteers: 0.0% of Clockticks
Mispredicts Resteers: 0.0% of Clockticks
Clears Resteers: 0.0% of Clockticks
Unknown Branches: 0.0% of Clockticks
DSB Switches: 0.0% of Clockticks
Length Changing Prefixes: 0.0% of Clockticks
MS Switches: 0.0% of Clockticks
Front-End Bandwidth: 0.0% of Pipeline Slots
Front-End Bandwidth MITE: 0.0% of Pipeline Slots
Decoder-0 Alone: 0.0% of Clockticks
Front-End Bandwidth DSB: 0.0% of Pipeline Slots
Front-End Bandwidth LSD: 0.0% of Pipeline Slots
(Info) DSB Coverage: 0.0%
(Info) LSD Coverage: 0.0%
(Info) DSB Misses Cost: 0.0% of Pipeline Slots
Bad Speculation: 100.0% of Pipeline Slots
| A significant proportion of pipeline slots containing useful work are
| being cancelled. This can be caused by mispredicting branches or by
| machine clears. Note that this metric value may be highlighted due to
| Branch Resteers issue.
|
Branch Mispredict: 0.0% of Pipeline Slots
Machine Clears: 100.0% of Pipeline Slots
| Issue: A significant portion of execution time is spent handling
| machine clears.
|
| Tips: See the "Memory Disambiguation" section in the Intel 64 and
| IA-32 Architectures Optimization Reference Manual.
|
Back-End Bound: 0.0% of Pipeline Slots
Memory Bound: 0.0% of Pipeline Slots
L1 Bound: 0.0% of Clockticks
DTLB Overhead: 0.0% of Clockticks
Load STLB Hit: 0.0% of Clockticks
Load STLB Miss: 0.0% of Clockticks
Loads Blocked by Store Forwarding: 0.0% of Clockticks
Lock Latency: 0.0% of Clockticks
Split Loads: 0.0% of Clockticks
FB Full: 0.0% of Clockticks
L2 Bound: 0.0% of Clockticks
L3 Bound: 0.0% of Clockticks
Contested Accesses: 0.0% of Clockticks
Data Sharing: 0.0% of Clockticks
L3 Latency: 0.0% of Clockticks
SQ Full: 0.0% of Clockticks
DRAM Bound: 0.0% of Clockticks
Memory Bandwidth: 0.0% of Clockticks
Memory Latency: 0.0% of Clockticks
Store Bound: 0.0% of Clockticks
Store Latency: 0.0% of Clockticks
False Sharing: 0.0% of Clockticks
Split Stores: 0.0% of Clockticks
Streaming Stores: 0.0% of Clockticks
DTLB Store Overhead: 0.0% of Clockticks
Store STLB Hit: 0.0% of Clockticks
Store STLB Hit: 0.0% of Clockticks
Core Bound: 0.0% of Pipeline Slots
Divider: 0.0% of Clockticks
Port Utilization: 0.0% of Clockticks
Cycles of 0 Ports Utilized: 0.0% of Clockticks
Serializing Operations: 0.0% of Clockticks
Slow Pause: 0.0% of Clockticks
Memory Fence: 0.0% of Clockticks
Mixing Vectors: 0.0% of Clockticks
Cycles of 1 Port Utilized: 0.0% of Clockticks
Cycles of 2 Ports Utilized: 0.0% of Clockticks
Cycles of 3+ Ports Utilized: 0.0% of Clockticks
ALU Operation Utilization: 0.0% of Clockticks
Port 0: 0.0% of Clockticks
Port 1: 0.0% of Clockticks
Port 6: 0.0% of Clockticks
Load Operation Utilization: 0.0% of Clockticks
Store Operation Utilization: 0.0% of Clockticks
E-Core
Retiring: 0.0% of Pipeline Slots
General Retirement: 0.0% of Pipeline Slots
FP Arithmetic: 0.0% of Pipeline Slots
Other: 0.0% of Pipeline Slots
Microcode Sequencer: 0.0% of Pipeline Slots
Front-End Bound: 0.0% of Pipeline Slots
Front-End Latency: 0.0% of Pipeline Slots
ICache Misses: 0.0% of Pipeline Slots
ITLB Overhead: 0.0% of Pipeline Slots
BACLEARS: 0.0% of Pipeline Slots
Branch Resteers: 0.0% of Pipeline Slots
Front-End Bandwidth: 0.0% of Pipeline Slots
Cisc: 0.0% of Pipeline Slots
Decode: 0.0% of Pipeline Slots
Pre-Decode Wrong: 0.0% of Pipeline Slots
Front-End Other: 0.0% of Pipeline Slots
Bad Speculation: 0.0% of Pipeline Slots
Branch Mispredict: 0.0% of Pipeline Slots
Machine Clears: 0.0% of Pipeline Slots
Machine Clear: 0.0% of Pipeline Slots
MO Machine Clear Overhead: 0.0% of Pipeline Slots
Back-End Bound: 0.0% of Pipeline Slots
Resource Bound: 0.0% of Pipeline Slots
Memory Scheduler: 0.0% of Pipeline Slots
Non-memory Scheduler: 0.0% of Pipeline Slots
Register: 0.0% of Pipeline Slots
Full Re-order Buffer (ROB): 0.0% of Pipeline Slots
Allocation Restriction: 0.0% of Pipeline Slots
Serializing Operations: 0.0% of Pipeline Slots
Alternative Back-End Bound: 0.0% of Pipeline Slots
Core Bound: 0.0%
Memory Bound: 0.0%
L2 Bound: 0.0%
L3 Bound: 0.0%
DRAM Bound: 0.0%
Average CPU Frequency: 0.000 MHz
Total Thread Count: 9
Paused Time: 0s
Effective Physical Core Utilization: 0.0% (0.000 out of 6)
| The metric value is low, which may signal a poor physical CPU cores
| utilization caused by:
| - load imbalance
| - threading runtime overhead
| - contended synchronization
| - thread/process underutilization
| - incorrect affinity that utilizes logical cores instead of physical
| cores
| Explore sub-metrics to estimate the efficiency of MPI and OpenMP parallelism
| or run the Locks and Waits analysis to identify parallel bottlenecks for
| other parallel runtimes.
|
Effective Logical Core Utilization: 63.8% (7.653 out of 12)
| The metric value is low, which may signal a poor logical CPU cores
| utilization. Consider improving physical core utilization as the first
| step and then look at opportunities to utilize logical cores, which in
| some cases can improve processor throughput and overall performance of
| multi-threaded applications.
|
Collection and Platform Info
Application Command Line: /home/kevin/intel/vtune/samples/matrix/matrix
User Name: kevin
Operating System: 4.18.0-372.19.1.el8_6.x86_64 Red Hat Enterprise Linux release 8.6 (Ootpa)
Computer Name: localhost.localdomain
Result Size: 305.1 MB
Collection start time: 12:13:29 15/10/2022 UTC
Collection stop time: 12:13:44 15/10/2022 UTC
Collector Type: Event-based sampling driver
CPU
Name: Intel(R) microarchitecture code named Alderlake-S
Frequency: 2.995 GHz
Logical CPU Count: 12
Cache Allocation Technology
Level 2 capability: available
Level 3 capability: not detected
If you want to skip descriptions of detected performance issues in the report,
enter: vtune -report summary -report-knob show-issues=false -r <my_result_dir>.
Alternatively, you may view the report in the csv format: vtune -report
<report_name> -format=csv.
vtune: Executing actions 100 % done
And this is the output of self_checker:
[kevin@localhost bin64]$ sudo ./vtune-self-checker.sh
[sudo] password for kevin:
Intel(R) VTune(TM) Profiler Self Check Utility
Copyright (C) 2009 Intel Corporation. All rights reserved.
Build Number: 624050
HW event-based analysis (counting mode) (Intel driver)
Example of analysis types: Performance Snapshot
Collection: Ok
vtune: Warning: Cannot collect GPU hardware metrics because libmd.so was not found. Make sure you have installed Metrics Discovery API library from https://github.com/intel/metrics-discovery. See Error Message: Cannot Collect GPU Hardware Metrics help topic for more details.
Finalization: Ok...
vtune: Warning: Cannot collect GPU hardware metrics because libmd.so was not found. Make sure you have installed Metrics Discovery API library from https://github.com/intel/metrics-discovery. See Error Message: Cannot Collect GPU Hardware Metrics help topic for more details.
Report: Ok
Instrumentation based analysis check
Example of analysis types: Hotspots and Threading with user-mode sampling
Collection: Ok
Finalization: Ok...
Report: Ok
HW event-based analysis check (Intel driver)
Example of analysis types: Hotspots with HW event-based sampling, HPC Performance Characterization, etc.
Collection: Ok
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
Finalization: Ok...
vtune: Warning: Cannot locate debugging information for the Linux kernel. Source-level analysis will not be possible. Function-level analysis will be limited to kernel symbol tables. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
Report: Ok
HW event-based analysis check (Intel driver)
Example of analysis types: Microarchitecture Exploration
Collection: Ok
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
Finalization: Ok...
vtune: Warning: Cannot locate debugging information for the Linux kernel. Source-level analysis will not be possible. Function-level analysis will be limited to kernel symbol tables. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
Report: Ok
HW event-based analysis with uncore events (Intel driver)
Example of analysis types: Memory Access
Collection: Ok
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
Finalization: Ok...
vtune: Warning: Cannot locate debugging information for the Linux kernel. Source-level analysis will not be possible. Function-level analysis will be limited to kernel symbol tables. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
Report: Ok
HW event-based analysis with stacks (Intel driver)
Example of analysis types: Hotspots with HW event-based sampling and call stacks
Collection: Ok
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
Finalization: Ok...
vtune: Warning: Cannot locate debugging information for the Linux kernel. Source-level analysis will not be possible. Function-level analysis will be limited to kernel symbol tables. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
Report: Ok
HW event-based analysis with context switches (Intel driver)
Example of analysis types: Threading with HW event-based sampling
Collection: Ok
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
Finalization: Ok...
vtune: Warning: Cannot locate debugging information for the Linux kernel. Source-level analysis will not be possible. Function-level analysis will be limited to kernel symbol tables. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
Report: Ok
Checking DPC++ application as prerequisite for GPU analyses: Fail
Unable to run DPC++ application on GPU connected to this system. If you are using an Intel GPU and want to verify profiling support for DPC++ applications, check these requirements:
* Install Intel(R) GPU driver.
* Install Intel(R) Level Zero GPU runtime.
* Install Intel(R) oneAPI DPC++ Runtime and set the environment.
The system is ready to be used for performance analysis with Intel VTune Profiler.
Review warnings in the output above to find product limitations, if any.
The system is ready for the following analyses:
* Performance Snapshot
* Hotspots and Threading with user-mode sampling
* Hotspots with HW event-based sampling, HPC Performance Characterization, etc.
* Microarchitecture Exploration
* Memory Access
* Hotspots with HW event-based sampling and call stacks
* Threading with HW event-based sampling
The following analyses have failed on the system:
* GPU Compute/Media Hotspots (characterization mode)
* GPU Compute/Media Hotspots (source analysis mode)
Log location: /tmp/vtune-tmp-root/self-checker-2022.10.15_14.27.41/log.txt
Hi Jaideep,
I'm seeing the same issue. Please find answers to the questions you ask below.
Any help or next steps are appreciated.
Thanks,
Max
I'm not running in a VM.
I have tried three different applications and see the same behavior in each. Task manager/Resource Monitors shows all 16 cores being utilized. Vtune suggests only the E-cores are active.
One other tidbit - In the Collection and Platform info - the report claims I'm running Windows 10, but as you can see from the copy and paste from windows device info, I'm running Windows 11.
Device name xxx
Processor 12th Gen Intel(R) Core(TM) i7-1270P 2.20 GHz
Installed RAM 16.0 GB (15.6 GB usable)
System type 64-bit operating system, x64-based processor
Pen and touch No pen or touch input is available for this display
Edition Windows 11 Pro
Version 21H2
Installed on 10/6/2022
OS build 22000.1098
Experience Windows Feature Experience Pack 1000.22000.1098.0
C:\Windows\System32>"C:\Program Files (x86)\Intel\oneAPI\vtune\2022.4.0\bin64\vtune-self-checker.bat"
Intel(R) VTune(TM) Profiler Self Check Utility
Copyright (C) 2009 Intel Corporation. All rights reserved.
Build Number: 624343
HW event-based analysis (counting mode) (Intel driver)
Example of analysis types: Performance Snapshot
Collection: Ok
Finalization: Ok...
Report: Ok
Instrumentation based analysis check
Example of analysis types: Hotspots and Threading with user-mode sampling
Collection: Ok
Finalization: Ok...
Report: Ok
HW event-based analysis check (Intel driver)
Example of analysis types: Hotspots with HW event-based sampling, HPC Performance Characterization, etc.
Collection: Ok
Finalization: Ok...
Report: Ok
HW event-based analysis check (Intel driver)
Example of analysis types: Microarchitecture Exploration
Collection: Ok
Finalization: Ok...
Report: Ok
HW event-based analysis with uncore events (Intel driver)
Example of analysis types: Memory Access
Collection: Ok
Finalization: Ok...
vtune: Warning: The result contains a lot of raw data. Finalization may take a long time to complete.
Report: Ok
HW event-based analysis with stacks (Intel driver)
Example of analysis types: Hotspots with HW event-based sampling and call stacks
Collection: Ok
Finalization: Ok...
Report: Ok
HW event-based analysis with context switches (Intel driver)
Example of analysis types: Threading with HW event-based sampling
Collection: Ok
Finalization: Ok...
Report: Ok
Checking DPC++ application as prerequisite for GPU analyses: Fail
Unable to run DPC++ application on GPU connected to this system. If you are using an Intel GPU and want to verify profiling support for DPC++ applications, check these requirements:
* Install Intel(R) GPU driver.
* Install Intel(R) Level Zero GPU runtime.
* Install Intel(R) oneAPI DPC++ Runtime and set the environment.
The system is ready to be used for performance analysis with Intel VTune Profiler.
Review warnings in the output above to find product limitations, if any.
The system is ready for the following analyses:
* Performance Snapshot
* Hotspots and Threading with user-mode sampling
* Hotspots with HW event-based sampling, HPC Performance Characterization, etc.
* Microarchitecture Exploration
* Memory Access
* Hotspots with HW event-based sampling and call stacks
* Threading with HW event-based sampling
The following analyses have failed on the system:
* GPU Compute/Media Hotspots (characterization mode)
* GPU Compute/Media Hotspots (source analysis mode)
Log location: C:\Users\optane\AppData\Local\Temp\vtune-tmp-optane\self-checker-2022.10.17_09.15.15\log.txt
C:\Program Files (x86)\Intel\oneAPI\vtune\2022.4.0\bin64>
Not sure why my comment got deleted, but I've got exactly the same issue on i5 12400 and RHEL8.
Although when I look at Event Count tab, I can see that most counters look like they are working correctly, but for example "Clockticks" column is always zero:
I'm including a CSV file with exported counter values as attachment.
This is the output of self checker:
[kevin@localhost bin64]$ sudo ./vtune-self-checker.sh
[sudo] password for kevin:
Intel(R) VTune(TM) Profiler Self Check Utility
Copyright (C) 2009 Intel Corporation. All rights reserved.
Build Number: 624050
HW event-based analysis (counting mode) (Intel driver)
Example of analysis types: Performance Snapshot
Collection: Ok
vtune: Warning: Cannot collect GPU hardware metrics because libmd.so was not found. Make sure you have installed Metrics Discovery API library from https://github.com/intel/metrics-discovery. See Error Message: Cannot Collect GPU Hardware Metrics help topic for more details.
Finalization: Ok...
vtune: Warning: Cannot collect GPU hardware metrics because libmd.so was not found. Make sure you have installed Metrics Discovery API library from https://github.com/intel/metrics-discovery. See Error Message: Cannot Collect GPU Hardware Metrics help topic for more details.
Report: Ok
Instrumentation based analysis check
Example of analysis types: Hotspots and Threading with user-mode sampling
Collection: Ok
Finalization: Ok...
Report: Ok
HW event-based analysis check (Intel driver)
Example of analysis types: Hotspots with HW event-based sampling, HPC Performance Characterization, etc.
Collection: Ok
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
Finalization: Ok...
vtune: Warning: Cannot locate debugging information for the Linux kernel. Source-level analysis will not be possible. Function-level analysis will be limited to kernel symbol tables. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
Report: Ok
HW event-based analysis check (Intel driver)
Example of analysis types: Microarchitecture Exploration
Collection: Ok
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
Finalization: Ok...
vtune: Warning: Cannot locate debugging information for the Linux kernel. Source-level analysis will not be possible. Function-level analysis will be limited to kernel symbol tables. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
Report: Ok
HW event-based analysis with uncore events (Intel driver)
Example of analysis types: Memory Access
Collection: Ok
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
Finalization: Ok...
vtune: Warning: Cannot locate debugging information for the Linux kernel. Source-level analysis will not be possible. Function-level analysis will be limited to kernel symbol tables. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
Report: Ok
HW event-based analysis with stacks (Intel driver)
Example of analysis types: Hotspots with HW event-based sampling and call stacks
Collection: Ok
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
Finalization: Ok...
vtune: Warning: Cannot locate debugging information for the Linux kernel. Source-level analysis will not be possible. Function-level analysis will be limited to kernel symbol tables. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
Report: Ok
HW event-based analysis with context switches (Intel driver)
Example of analysis types: Threading with HW event-based sampling
Collection: Ok
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
Finalization: Ok...
vtune: Warning: Cannot locate debugging information for the Linux kernel. Source-level analysis will not be possible. Function-level analysis will be limited to kernel symbol tables. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
Report: Ok
Checking DPC++ application as prerequisite for GPU analyses: Fail
Unable to run DPC++ application on GPU connected to this system. If you are using an Intel GPU and want to verify profiling support for DPC++ applications, check these requirements:
* Install Intel(R) GPU driver.
* Install Intel(R) Level Zero GPU runtime.
* Install Intel(R) oneAPI DPC++ Runtime and set the environment.
The system is ready to be used for performance analysis with Intel VTune Profiler.
Review warnings in the output above to find product limitations, if any.
The system is ready for the following analyses:
* Performance Snapshot
* Hotspots and Threading with user-mode sampling
* Hotspots with HW event-based sampling, HPC Performance Characterization, etc.
* Microarchitecture Exploration
* Memory Access
* Hotspots with HW event-based sampling and call stacks
* Threading with HW event-based sampling
The following analyses have failed on the system:
* GPU Compute/Media Hotspots (characterization mode)
* GPU Compute/Media Hotspots (source analysis mode)
Log location: /tmp/vtune-tmp-root/self-checker-2022.10.15_14.27.41/log.txt
And this is the output when I try the default matrix demo on CLI:
[kevin@localhost bin64]$ ./vtune -collect uarch-exploration -r '/home/kevin/Documents/vtune_test' /home/kevin/intel/vtune/samples/matrix/matrix
vtune: Warning: To enable hardware event-based sampling, VTune Profiler has disabled the NMI watchdog timer. The watchdog timer will be re-enabled after collection completes.
vtune: Collection started. To stop the collection, either press CTRL-C or enter from another console window: vtune -r /home/kevin/Documents/vtune_test -command stop.
Addr of buf1 = 0x7fe3f4f61010
Offs of buf1 = 0x7fe3f4f61180
Addr of buf2 = 0x7fe3f2f60010
Offs of buf2 = 0x7fe3f2f601c0
Addr of buf3 = 0x7fe3f0f5f010
Offs of buf3 = 0x7fe3f0f5f100
Addr of buf4 = 0x7fe3eef5e010
Offs of buf4 = 0x7fe3eef5e140
Threads #: 8 Pthreads
Matrix size: 2048
Using multiply kernel: multiply1
Execution time = 15.122 seconds
vtune: Collection stopped.
vtune: Using result path `/home/kevin/Documents/vtune_test'
vtune: Executing actions 19 % Resolving information for `ld-2.28.so'
vtune: Warning: Cannot locate file `nft_counter.ko'.
vtune: Executing actions 19 % Resolving information for `nft_counter'
vtune: Warning: Cannot locate file `bridge.ko'.
vtune: Executing actions 19 % Resolving information for `matrix'
vtune: Warning: Cannot locate file `r8169.ko'.
vtune: Executing actions 19 % Resolving information for `r8169'
vtune: Warning: Cannot locate file `nf_tables.ko'.
vtune: Executing actions 19 % Resolving information for `nf_tables'
vtune: Warning: Cannot locate file `nf_conntrack.ko'.
vtune: Executing actions 21 % Resolving information for `libc-2.28.so'
vtune: Warning: Cannot locate file `nft_ct.ko'.
vtune: Executing actions 21 % Resolving information for `nft_ct'
vtune: Warning: Cannot locate debugging information for file `/usr/lib64/libc-2.28.so'.
vtune: Warning: Cannot locate file `drm.ko'.
vtune: Executing actions 21 % Resolving information for `drm'
vtune: Warning: Cannot locate file `kvm.ko'.
vtune: Executing actions 22 % Resolving information for `kvm'
vtune: Warning: Cannot locate file `xfs.ko'.
vtune: Executing actions 22 % Resolving information for `xfs'
vtune: Warning: Cannot locate file `i915.ko'.
vtune: Executing actions 22 % Resolving information for `i915'
vtune: Warning: Cannot locate file `sep5.ko'.
vtune: Executing actions 22 % Resolving information for `vmlinux'
vtune: Warning: Function and source-level analysis for the Linux kernel will not be possible since neither debug version of the kernel nor kernel symbol tables are found. See the Enabling Linux Kernel Analysis topic in the product online help for instructions.
vtune: Executing actions 75 % Generating a report Elapsed Time: 15.135s
Clockticks: 0
P-Core: 0
E-Core: 0
Instructions Retired: 69,798,000,000
P-Core: 0
E-Core: 0
CPI Rate: 0.000
P-Core: 0.000
E-Core: 0.000
MUX Reliability: 0.000
| Precision of collected HW event data is not enough. Metrics data may be
| unreliable. Consider increasing your application execution time, using
| the multiple runs mode instead of event multiplexing, or creating a
| custom analysis with a limited subset of HW events. If you are using a
| driverless collection, consider reducing the value of
| /sys/bus/event_source/devices/cpu/perf_event_mux_interval_ms file.
|
P-Core
Retiring: 0.0% of Pipeline Slots
Light Operations: 0.0% of Pipeline Slots
FP Arithmetic: 0.0% of uOps
FP x87: 0.0% of uOps
FP Scalar: 0.0% of uOps
FP Vector: 0.0% of uOps
Memory Operations: 0.0% of Pipeline Slots
Fused Instructions: 0.0% of Pipeline Slots
Non Fused Branches: 0.0% of Pipeline Slots
Nop Instructions: 0.0% of Pipeline Slots
Other: 0.0% of Pipeline Slots
Heavy Operations: 0.0% of Pipeline Slots
Microcode Sequencer: 0.0% of Pipeline Slots
Assists: 0.0% of Pipeline Slots
CISC: 0.0% of Pipeline Slots
Front-End Bound: 0.0% of Pipeline Slots
Front-End Latency: 0.0% of Pipeline Slots
ICache Misses: 0.0% of Clockticks
ITLB Overhead: 0.0% of Clockticks
Branch Resteers: 0.0% of Clockticks
Mispredicts Resteers: 0.0% of Clockticks
Clears Resteers: 0.0% of Clockticks
Unknown Branches: 0.0% of Clockticks
DSB Switches: 0.0% of Clockticks
Length Changing Prefixes: 0.0% of Clockticks
MS Switches: 0.0% of Clockticks
Front-End Bandwidth: 0.0% of Pipeline Slots
Front-End Bandwidth MITE: 0.0% of Pipeline Slots
Decoder-0 Alone: 0.0% of Clockticks
Front-End Bandwidth DSB: 0.0% of Pipeline Slots
Front-End Bandwidth LSD: 0.0% of Pipeline Slots
(Info) DSB Coverage: 0.0%
(Info) LSD Coverage: 0.0%
(Info) DSB Misses Cost: 0.0% of Pipeline Slots
Bad Speculation: 100.0% of Pipeline Slots
| A significant proportion of pipeline slots containing useful work are
| being cancelled. This can be caused by mispredicting branches or by
| machine clears. Note that this metric value may be highlighted due to
| Branch Resteers issue.
|
Branch Mispredict: 0.0% of Pipeline Slots
Machine Clears: 100.0% of Pipeline Slots
| Issue: A significant portion of execution time is spent handling
| machine clears.
|
| Tips: See the "Memory Disambiguation" section in the Intel 64 and
| IA-32 Architectures Optimization Reference Manual.
|
Back-End Bound: 0.0% of Pipeline Slots
Memory Bound: 0.0% of Pipeline Slots
L1 Bound: 0.0% of Clockticks
DTLB Overhead: 0.0% of Clockticks
Load STLB Hit: 0.0% of Clockticks
Load STLB Miss: 0.0% of Clockticks
Loads Blocked by Store Forwarding: 0.0% of Clockticks
Lock Latency: 0.0% of Clockticks
Split Loads: 0.0% of Clockticks
FB Full: 0.0% of Clockticks
L2 Bound: 0.0% of Clockticks
L3 Bound: 0.0% of Clockticks
Contested Accesses: 0.0% of Clockticks
Data Sharing: 0.0% of Clockticks
L3 Latency: 0.0% of Clockticks
SQ Full: 0.0% of Clockticks
DRAM Bound: 0.0% of Clockticks
Memory Bandwidth: 0.0% of Clockticks
Memory Latency: 0.0% of Clockticks
Store Bound: 0.0% of Clockticks
Store Latency: 0.0% of Clockticks
False Sharing: 0.0% of Clockticks
Split Stores: 0.0% of Clockticks
Streaming Stores: 0.0% of Clockticks
DTLB Store Overhead: 0.0% of Clockticks
Store STLB Hit: 0.0% of Clockticks
Store STLB Hit: 0.0% of Clockticks
Core Bound: 0.0% of Pipeline Slots
Divider: 0.0% of Clockticks
Port Utilization: 0.0% of Clockticks
Cycles of 0 Ports Utilized: 0.0% of Clockticks
Serializing Operations: 0.0% of Clockticks
Slow Pause: 0.0% of Clockticks
Memory Fence: 0.0% of Clockticks
Mixing Vectors: 0.0% of Clockticks
Cycles of 1 Port Utilized: 0.0% of Clockticks
Cycles of 2 Ports Utilized: 0.0% of Clockticks
Cycles of 3+ Ports Utilized: 0.0% of Clockticks
ALU Operation Utilization: 0.0% of Clockticks
Port 0: 0.0% of Clockticks
Port 1: 0.0% of Clockticks
Port 6: 0.0% of Clockticks
Load Operation Utilization: 0.0% of Clockticks
Store Operation Utilization: 0.0% of Clockticks
E-Core
Retiring: 0.0% of Pipeline Slots
General Retirement: 0.0% of Pipeline Slots
FP Arithmetic: 0.0% of Pipeline Slots
Other: 0.0% of Pipeline Slots
Microcode Sequencer: 0.0% of Pipeline Slots
Front-End Bound: 0.0% of Pipeline Slots
Front-End Latency: 0.0% of Pipeline Slots
ICache Misses: 0.0% of Pipeline Slots
ITLB Overhead: 0.0% of Pipeline Slots
BACLEARS: 0.0% of Pipeline Slots
Branch Resteers: 0.0% of Pipeline Slots
Front-End Bandwidth: 0.0% of Pipeline Slots
Cisc: 0.0% of Pipeline Slots
Decode: 0.0% of Pipeline Slots
Pre-Decode Wrong: 0.0% of Pipeline Slots
Front-End Other: 0.0% of Pipeline Slots
Bad Speculation: 0.0% of Pipeline Slots
Branch Mispredict: 0.0% of Pipeline Slots
Machine Clears: 0.0% of Pipeline Slots
Machine Clear: 0.0% of Pipeline Slots
MO Machine Clear Overhead: 0.0% of Pipeline Slots
Back-End Bound: 0.0% of Pipeline Slots
Resource Bound: 0.0% of Pipeline Slots
Memory Scheduler: 0.0% of Pipeline Slots
Non-memory Scheduler: 0.0% of Pipeline Slots
Register: 0.0% of Pipeline Slots
Full Re-order Buffer (ROB): 0.0% of Pipeline Slots
Allocation Restriction: 0.0% of Pipeline Slots
Serializing Operations: 0.0% of Pipeline Slots
Alternative Back-End Bound: 0.0% of Pipeline Slots
Core Bound: 0.0%
Memory Bound: 0.0%
L2 Bound: 0.0%
L3 Bound: 0.0%
DRAM Bound: 0.0%
Average CPU Frequency: 0.000 MHz
Total Thread Count: 9
Paused Time: 0s
Effective Physical Core Utilization: 0.0% (0.000 out of 6)
| The metric value is low, which may signal a poor physical CPU cores
| utilization caused by:
| - load imbalance
| - threading runtime overhead
| - contended synchronization
| - thread/process underutilization
| - incorrect affinity that utilizes logical cores instead of physical
| cores
| Explore sub-metrics to estimate the efficiency of MPI and OpenMP parallelism
| or run the Locks and Waits analysis to identify parallel bottlenecks for
| other parallel runtimes.
|
Effective Logical Core Utilization: 63.8% (7.653 out of 12)
| The metric value is low, which may signal a poor logical CPU cores
| utilization. Consider improving physical core utilization as the first
| step and then look at opportunities to utilize logical cores, which in
| some cases can improve processor throughput and overall performance of
| multi-threaded applications.
|
Collection and Platform Info
Application Command Line: /home/kevin/intel/vtune/samples/matrix/matrix
User Name: kevin
Operating System: 4.18.0-372.19.1.el8_6.x86_64 Red Hat Enterprise Linux release 8.6 (Ootpa)
Computer Name: localhost.localdomain
Result Size: 305.1 MB
Collection start time: 12:13:29 15/10/2022 UTC
Collection stop time: 12:13:44 15/10/2022 UTC
Collector Type: Event-based sampling driver
CPU
Name: Intel(R) microarchitecture code named Alderlake-S
Frequency: 2.995 GHz
Logical CPU Count: 12
Cache Allocation Technology
Level 2 capability: available
Level 3 capability: not detected
If you want to skip descriptions of detected performance issues in the report,
enter: vtune -report summary -report-knob show-issues=false -r <my_result_dir>.
Alternatively, you may view the report in the csv format: vtune -report
<report_name> -format=csv.
vtune: Executing actions 100 % done
Hi All,
The issue is known issue on VTune product when the system runs on Hyper-V because Hyper-V blocks non-converged PMU feature on hetero platforms.
Can you please disable Hyper-V in Windows features and let us know if the issue still persists. Please attach the results after disabling Hyper-V.
Thanks,
Jaideep
Hi Jaideep,
I already thought that could be an issue and checked it wasn't on and it isn't, I also turned off virtulisation and it also didnt make a difference.
Thanks again,
Oliver
Hi Jaideep,
Looking at what you said I had to go one step further, I turned Intel Virtulisation (VMX) off in the BIOS and that has made me able to see P-core performance in Microarchitecture exploration now. (resolved the issue).
Maybe theres some virtulisation option that needs to be reset after setting hyper-v.
For any one else following this post I would suggest disabling virtulisation off in the BIOS and see if that works for you.
Regards,
Oliver
Hi Jaideep,
I have tried to follow the advice and I have disabled virtualization in BIOS (as can be seen in the screenshot below), but I still have the same issue, I'm attaching results of the microarchitecture analysis.
I'm not running Windows, I'm running Linux (RHEL8). Is there something I need to change in my OS settings?
Both "Intel Virtualization Tech" and "Intel VT-D Tech" is disabled:
Hello Jaideep, as I have previously stated, this did NOT solve my issue. As can be seen in attached screenshot, virtualization in BIOS is disabled, but still microarchitecture analysis does not show values for P cores.
OS: RHEL8
Is there anything I can try next?
