topic Re:Vtune top-down report fails in Analyzers

Vtune top-down report fails

blednotnik — Thu, 28 Sep 2023 17:50:29 GMT

Dear Vtune experts,

I'm trying to generate csv report from the data collected using -collect=hotspots options.

And running on that error:

vtune: Executing actions 75 % Generating a report exception maximum recursion depth exceeded Traceback (most recent call last): File "/localdisk/jenkins/workspace/Official/VTune/Build/vtune/build/build_release_posix-x86_64_icx_2023.0.0_gnutools_9.2_binutils_2.28/reporter.python/src/python/tmpf2p9n04h/wrapper.py", line 110, in run File "/cvmfs/projects.cern.ch/intelsw/oneAPI/linux/x86_64/2023/vtune/2023.2.0/config/report/top-down.tmpl", line 82, in <module> table_formatter.format_table(tt) File "/localdisk/jenkins/workspace/Official/VTune/Build/vtune/build/build_release_posix-x86_64_icx_2023.0.0_gnutools_9.2_binutils_2.28/dpihelpers.python/python/dpihelpers1/tmpnuv7q07u/table_formatter.py", line 191, in format_table File "/localdisk/jenkins/workspace/Official/VTune/Build/vtune/build/build_release_posix-x86_64_icx_2023.0.0_gnutools_9.2_binutils_2.28/dpihelpers.python/python/dpihelpers1/tmpnuv7q07u/table_formatter.py", line 242, in _prepare_table File "/localdisk/jenkins/workspace/Official/VTune/Build/vtune/build/build_release_posix-x86_64_icx_2023.0.0_gnutools_9.2_binutils_2.28/dpihelpers.python/python/dpihelpers1/tmpnuv7q07u/table_formatter.py", line 232, in _deep_row_iterator File "/localdisk/jenkins/workspace/Official/VTune/Build/vtune/build/build_release_posix-x86_64_icx_2023.0.0_gnutools_9.2_binutils_2.28/dpihelpers.python/python/dpihelpers1/tmpnuv7q07u/table_formatter.py", line 228, in _deep_row_iterator_impl File "/localdisk/jenkins/workspace/Official/VTune/Build/vtune/build/build_release_posix-x86_64_icx_2023.0.0_gnutools_9.2_binutils_2.28/dpihelpers.python/python/dpihelpers1/tmpnuv7q07u/table_formatter.py", line 228, in _deep_row_iterator_impl File "/localdisk/jenkins/workspace/Official/VTune/Build/vtune/build/build_release_posix-x86_64_icx_2023.0.0_gnutools_9.2_binutils_2.28/dpihelpers.python/python/dpihelpers1/tmpnuv7q07u/table_formatter.py", line 228, in _deep_row_iterator_impl [Previous line repeated 985 more times] File "/localdisk/jenkins/workspace/Official/VTune/Build/vtune/build/build_release_posix-x86_64_icx_2023.0.0_gnutools_9.2_binutils_2.28/dpihelpers.python/python/dpihelpers1/tmpnuv7q07u/table_formatter.py", line 226, in _deep_row_iterator_impl File "/localdisk/jenkins/workspace/Official/VTune/Build/vtune/build/build_release_posix-x86_64_icx_2023.0.0_gnutools_9.2_binutils_2.28/dpihelpers.python/python/dpihelpers1/tmpnuv7q07u/utils.py", line 123, in __iter__ File "/localdisk/jenkins/workspace/Official/VTune/Build/vtune/build/build_release_posix-x86_64_icx_2023.0.0_gnutools_9.2_binutils_2.28/dicerhelpers.python/python/tmpy5_v2ck6/qlibrary.py", line 665, in __init__ RecursionError: maximum recursion depth exceeded

Here is the command I'm using to create a report:

vtune -report top-down -call-stack-mode all -column="CPU Time:Self","Module" -report-out ouput.csv -filter "Function Stack" -format csv -csv-delimiter comma;

Could you give me a hint of what can be wrong?

Re: Vtune top-down report fails

VaradJ_Intel — Fri, 29 Sep 2023 10:56:38 GMT

Good day to you!

Thanks for posting on Intel Communities.

Please can you share the following information:

1. O.S details

2. Processor details

3. Kernel version

4. Vtune version

5. The exact steps you followed

6. A sample reproducer of your application that can help us to reproduce the issue you are facing

Thank You!

Regards

Varad

Re:Vtune top-down report fails

VaradJ_Intel — Mon, 09 Oct 2023 05:43:52 GMT

Good day to you!

Please can you provide us with the above details.

Thank You!

Regards

Varad

Re: Vtune top-down report fails

blednotnik — Thu, 12 Oct 2023 13:02:57 GMT

Hi,
Sorry for a quite slow reply.

The O.S. is CentOS 7.

The processor details:

Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 64 On-line CPU(s) list: 0-31 Off-line CPU(s) list: 32-63 Thread(s) per core: 1 Core(s) per socket: 16 Socket(s): 2 NUMA node(s): 2 Vendor ID: AuthenticAMD CPU family: 23 Model: 49 Model name: AMD EPYC 7302 16-Core Processor Stepping: 0 CPU MHz: 3000.000 CPU max MHz: 3000.0000 CPU min MHz: 1500.0000 BogoMIPS: 5999.54 Virtualization: AMD-V L1d cache: 32K L1i cache: 32K L2 cache: 512K L3 cache: 16384K NUMA node0 CPU(s): 0-15 NUMA node1 CPU(s): 16-31

kernel version:

Linux version 3.10.0-1160.42.2.el7.x86_64 (mockbuild@kbuilder.bsys.centos.org) (gcc version 4.8.5 20150623 (Red Hat 4.8.5-44) (GCC) ) #1 SMP Tue Sep 7 14:49:57 UTC 2021

packages:

Vtune version 2023.2.0

Python 3.9.12

GCC compiler GCC 1.1

The full version of the installed packages you can find here: https://lcginfo.cern.ch/release_packages/x86_64-centos7-gcc11-opt/102b_ATLAS_29/

Unfortunately, I don't think that I can easily provide the sample reproducer, cause those tests were run in the Athena framework with ATLAS experiment data. The code is open, but I'm not sure that you are willing to install it to check the issue. But if it's only a way, I will try to come up with some container.
I suspect that the output is quite large thus the table_formatter.py fails to produce a CVS file. I can share the r000hs with you on any suitable for you file exchanger, but I can't attach it here, because it's >1G uncompressed and over 71Mb compressed.

Thank you, Best regards, Tania.

Re: Vtune top-down report fails

VaradJ_Intel — Tue, 17 Oct 2023 11:46:28 GMT

Good day to you!

We can only offer direct support for Intel hardware platforms that the Intel® oneAPI product supports.

Intel provides instructions on how to compile oneAPI code for both CPU and a wide range of GPU accelerators.

https://intel.github.io/llvm-docs/GetStartedGuide.html

We can see that your processor and operating system don't meet the Intel Vtune system requirements.

We recommend you try using an Intel processor & supported operating system as mentioned in the below system requirements page:

https://www.intel.com/content/www/us/en/developer/articles/system-requirements/vtune-profiler-system-requirements.html

Also, you can try executing the same task on Intel DevCloud:

https://devcloud.intel.com/oneapi/get_started/

We tried creating csv report from the result file in Intel DevCloud and it is working fine for us. Please refer the screenshot below:

Thank You!

Regards

Varad

Re: Vtune top-down report fails

blednotnik — Wed, 18 Oct 2023 21:45:58 GMT

Hi, Thank you for your suggestion.

Could you please let me know how large the size of the r000hs is in your case?

Best regards, Tania.

Re:Vtune top-down report fails

VaradJ_Intel — Mon, 23 Oct 2023 11:44:57 GMT

Good day to you!

We tried it on multiple result folders ranging in few MB to GB. The largest one among them was close to 3GB

Thank You!

Regards

Varad

Re:Vtune top-down report fails

VaradJ_Intel — Thu, 02 Nov 2023 11:06:19 GMT

Good day to you!

Could you please give us an update?

Did you tried executing the same task on a supported Processor and Operating System?

Thank You!

Regards

Varad

Re: Re:Vtune top-down report fails

blednotnik — Thu, 02 Nov 2023 15:40:07 GMT

Hi,

Sorry not yet. I will explore the https://devcloud.intel.com/oneapi/get_started/ option and give you an update by the end of next week.

Thank you, Best regards, Tania.

Re:Vtune top-down report fails

VaradJ_Intel — Mon, 13 Nov 2023 17:38:44 GMT

Good day to you!

Could you please give us an update?

Thank You!

Regards

Varad

Re:Vtune top-down report fails

VaradJ_Intel — Mon, 20 Nov 2023 05:54:32 GMT

Good day to you!

We have not heard back from you.

This thread will no longer be monitored by Intel.

If you need further assistance, please post a new question.

Thanks and Regards

Varad