I used the following command for profiling:
vtune -collect memory-access -knob analyze-mem-objects=true -knob analyze-openmp=true -data-limit=30000 -result-dir <my_home_dir>/gapbs/vtune_results/sssp/sssp_road_memAcc_cl12_t8 -app-working-dir <my_home_dir>/gapbs -- numactl --interleave=all <my_home_dir>/gapbs/sssp -f <my_home_dir>/gapbs/inputs/GAP-road.mtx -n 1
Trying to execute the above command with the help of a SLURM job results in the following error:
vtune: Error: [Instrumentation Engine]: Source/pin/elfio/img_elf.cpp: ProcessSectionHeaders: 809: unknown section type 0x6fff4c04 for sec[123,.deplibs] in /rwthfs/rz/SW/UTIL/clang/12.0.0/lib64/libc++.so.1 vtune: Collection failed. vtune: Internal Error
Intel VTune Profile 2020 Update 1
Product Build 607630
(using command $hostnamectl)
Static hostname: <my_host_name>
Icon name: computer-server
Machine ID: ****************************************
Boot ID: ********************************************
Operating System: CentOS Linux 7 (Core)
CPE OS Name: cpe:/o:centos:centos:7
Kernel: Linux 3.10.0-1160.45.1.el7.x86_64
(using command $lsb_release -a)
LSB Version: :core-4.1-amd64:core-4.1-noarch
Distributor ID: CentOS
Description: CentOS Linux release 7.9.2009 (Core)
Thank you for posting in Intel Communities. We are trying to reproduce your issue from our end. And we observed that you're using a older version of VTune. So we suggest you to upgrade to the latest version and try to reproduce the sample on the updated version of vtune.
And could you please send us the result directory of the failed analysis & the self-checker logs? For self checker logs, please follow the below steps:
- In Vtune installed directory(path: /intel/oneapi/vtune/latest/bin64), run the below command:
- VTune Profiler Self Check Utility starts and the log location will be given in the end of the analysis(Log location: /tmp/vtune-tmp-administrator/self-checker-xxxxxx/log.txt
We tried to reproduce your issue in Intel Devcloud for oneAPI, but we were unable to. So we would like to provide you the steps that worked for us. If you are not a devcloud user, please enroll for a new devcloud account here- https://devcloud.intel.com/oneapi/
1)We recommend that you create a new environment when installing. To install the core python3 environment, do:
conda create -n <env_name> intelpython3_core python=3.x
Please note that "x" in "python=3.x" should signify which version of Python* you would like to install.
After creating, you can activate that respective environment using
conda activate <env-name>
We created a new Intel Aikit Pytorch environment and activated it.
2) After cloning the Gap benchmark suite, we installed clang-12 compiler with the below command:
git clone https://github.com/sbeamer/gapbs.git conda install -c conda-forge clang-12*
3) We compiled the sssp code with the below command:
clang++ sssp.cc -o output-file
4) We used the same command you used for profiling:
vtune -collect memory-access -knob analyze-mem-objects=true -knob analyze-openmp=true -data-limit=30000 -result-dir ~/gapbs/vtune_results/sssp/sssp_road_memAcc_cl12_t8 -app-working-dir ~/gapbs -- <sssp-code-output-file> --interleave=all ~/gapbs/sssp -f ~/gapbs/test/graphs/4.mtx -n 1
If this resolves your issue, make sure you accept this as a solution. This would help others with the similar issue. Thank you!
We assume that your issue is resolved. If you need any additional information, please post a new question as this thread will no longer be monitored by Intel.