ifort, icc, icpc - "Nice to know" options
by
Tor Johansen
—
last modified
Aug 23, 2010 01:03 PM
Compiler options for the Intel ifort, icc, icpc compilers that are "nice to know" about for many users.
Below is a list of compiler options for the Intel ifort compiler that we consider to be "nice to know" for many users. The list contains options that you typical should test out for
-
increased performance,
-
portability issues
-
and trouble shooting.
| -V |
display compiler version information |
| -openmp-stubs |
enables the user to compile OpenMP programs in sequential mode. The openmp directives are ignored and a stub OpenMP library is linked (sequential) |
| -parallel |
enable the auto-parallelizer to generate multi-threaded code for loops that can be safely executed in parallel |
| -fast |
enable -O3 -ipo -static |
| -O1 |
Faster compilation, disable some optimizations which increase code size for a small speed benefit. Also disables software pipelining and global code scheduling |
| -ftrapuv | trap uninitialized variables |
| -unroll0 | disable loop unrolling |
| -[no-]IPF-fltacc | enable/disable optimizations that affect floating point accuracy |
| -mp | maintain floating point precision (disables some optimizations) |
| -[no-]ftz | enable/disable flush denormal results to zero |
| -w |
disable all warnings |
| -g |
produce symbolic debug information in object file (implies -O0 when another optimization option is not explicitly set) |
Check the manual page, 'man icc/icpc', for more detailed information.
