My company is reviewing upgrading from Intel Fortran Compiler XE 13.0 to Intel® Fortran Compiler 16.0 and noticed an anomaly that is blocking us. This issue can be duplicated using the following program:
X = 8.3141200E-2
Using version 13.0 to compile a binary, copying and running that binary on all test machines produces: 0.9965457.
Using version 16.0 to compile a binary, copying and running that binary on test machines produces different results : 0.9965457, and 0.9965458 on others.
What external factors are impacting the binary to cause this unexpected behavior, and what can I do to ensure a stable testing environment? Let me know what additional information can I provide to assist in reviewing this.
Thanks for your time,
Thanks for the help! In your document you mention:
"-fimf-arch-consistency=true – Math library gives same results across processors"
I would like to give this a test, but I do not know where/how to set this switch. I could not find it in the "Property Pages | Configuration Properties | Fortran". Also tried to manually add it to the "Property Pages | Configuration Properties | Fortran | Command Line | Additional Options:" as "-fimf-arch-consistency=true". But the compiler reported:
"1>ifort: command line warning #10006: ignoring unknown option '/fimf-arch-consistency=true'"
Your proposal did not work but it got me on the right track. The mentioned
"/Qfimf-arch-consistency=true" did not work. However, "/Qimf-arch-consistency:true" did.
I am happy to say that in my small test set the results are now consistent!
Thanks Steve and mecej4,