× If you run into troubles, it is always a good habit to report the following information:
  • the way build.mrcc was invoked
  • the output of build.mrcc
  • compiler version (for example: ifort -V, gfortran -v)
  • blas/lapack versions
  • as well as gcc and glibc versions

as well as the value of the relevant environmental variables, such OMP_NUM_THREADS, etc.

This information helps us a lot when figuring out what is going on with your compilation :)

Intel compiler warning

More
4 years 9 months ago #11 by rkanters
Hi,

I just compiled mrcc.2014-07-10 with the Intel 11 compiler (./build.mrcc Intel -pOMP) and I get the following warning message (the only one in the list):

Compiling ovirt with options -fpp -O3 -i8 -assume byterecl -DIntel -DINT64 -DOMP
An internal threshold was exceeded for routine MAIN__ and optimization level may be reduced. See software.intel.com/en-us/articles/intern...reshold-was-exceeded for more information and advice.

When following that link, I see this warning suggests that it is possible that threading is not really used for the code and that a compiler option -override-limits could fix the issue.

Is this something to worry about? Should I play around with the build program to see whether I can force this additional option to be passed on the the compiler?

Please Log in or Create an account to join the conversation.

More
4 years 9 months ago #12 by kallay
Replied by kallay on topic Intel compiler warning
Hello,
I don't think you should worry about that. Only the optimization level was reduced, and in the worst case the ovirt program will run slower. To avoid this you can try the -override-limits compiler option, but the best solution is to upgrade to the last Intel version (14.0.2).

Best regards,
Mihaly Kallay
The following user(s) said Thank You: rkanters

Please Log in or Create an account to join the conversation.

More
4 years 9 months ago #21 by rkanters
Replied by rkanters on topic Intel compiler warning
Dear Mihaly,

The best solution wasn't an option (because of lack of budget), so I used the -override-limits. After running MRCC for a bit, it seems that my worry about ovirt not possibly using multiple threads because of the compiler limits was indeed not something to worry about. I have not seen ovirt at over 100% cpu usage, so I assume it wasn't multithreaded to begin with.

Please Log in or Create an account to join the conversation.

Time to create page: 0.021 seconds
Powered by Kunena Forum