After getting VTUNE more often than not to simply hang when finalizing results, I went the conservative way:
- short runtime
- single thread
voila -- everything fine -- finalizing was a snap.
Setting the number of threads used to 8, 4 or even 2 makes finalizing hang (and me having wasted plenty of time).
- runtime of application: 9min
- CentOS Linux release 7.6.1810 (Core)
- CPU: Xenon Skylake, 2.9GHz
- #CPUs: 32
- result size: 30MB (87MB)
- user-mode sampling and tracing
- "Duration time estimate" setting: between 15min and 3hours
- "Result size from collection start, MB": 500000
- "CPU sampling internal, ms": 10
- "analyze child processes": not selected
- threading package: openMP (I did not decide this!)
- executable size: 1.5GByte
- output of Linux tool file: ELF 64-bit LSB executable, x86-64, version 1 (GNU/Linux), dynamically linked (uses shared libs), for GNU/Linux 2.6.32, not stripped
Please no suggestions to download and install a newer version, as this should have worked already in 2018 since decades.
I cannot be root currently, but I'm working on getting this permission (in order to install device drivers).
Thank you for posting in Intel Communities.
- Please, can you share the exact steps you followed and a sample reproducer (a sample application that is similar to the application you are trying to analyze)?
- Can you please run hardware event base sampling with the same application and let us know if it is working or not?
- If Possible Could you please upgrade your Vtune to the latest version and run the same sample again and let us know?
Glad to know 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. Have a great day ahead.