diff options
Diffstat (limited to 'libraries/fftw/README')
| -rw-r--r-- | libraries/fftw/README | 11 |
1 files changed, 9 insertions, 2 deletions
diff --git a/libraries/fftw/README b/libraries/fftw/README index c357014303ff..10b6bff214fc 100644 --- a/libraries/fftw/README +++ b/libraries/fftw/README @@ -4,5 +4,12 @@ symmetric, and parallel transforms, and can handle arbitrary array sizes efficiently. FFTW is typically faster than other publically-available FFT implementations, and is even competitive with vendor-tuned libraries. -There are several machine-dependent optimizations that can be enabled in order -to generate SIMD code; take a look in fftw.SlackBuild to enable them. +By default non-portable binaries will be created. If you need to create +a package that is shared among various machines, pass PORTABLE=yes to the +script. + +To enable SSE or SSE2 SIMD optimizations pass SSE=yes or SSE2=yes respectively +to the script. Enabling these will still create a portable package as fftw falls +back to the standard code, if the optimized one isn't supported on the cpu. +However, to be able to build them your cpu has to actually support those +extensions. |