CHALLENGEcomplib is a comprehensive collection of high-
performance mathematical libraries that provide support for
numerical techniques used in scientific and technical
computing. These libraries include public-domain software
such as LAPACK, LINPACK and EISPACK, which are provided by
SGI for your convenience.
CHALLENGEcomplib 3.0 is available in several modes in single
and multiple processor versions (----mmmmpppp).:
+o 64-bit mode for R8000 and R10000 architectures (----mmmmiiiippppssss4444)
+o high performance 32-bit (n32) mode (----mmmmiiiippppssss4444) and (----
mmmmiiiippppssss3333) for all platforms
+o 32-bit mode (----mmmmiiiippppssss2222) for R4* platforms
The CCCCHHHHAAAALLLLLLLLEEEENNNNGGGGEEEEccccoooommmmpppplllliiiibbbb library is available on Silicon
Graphics, Inc., systems via the compilation flag
----llllccccoooommmmpppplllliiiibbbb....ssssggggiiiimmmmaaaatttthhhh or ----llllccccoooommmmpppplllliiiibbbb....ssssggggiiiimmmmaaaatttthhhh----mmmmpppp for
multiprocessing libraries.
The library ccccoooommmmpppplllliiiibbbb....ssssggggiiiimmmmaaaatttthhhh contains an extensive collection
of industry standard libraries such as Basic Linear Algebra
Subprograms (BLAS), the Extended BLAS (Level 2 and Level 3),
EISPACK, LINPACK, and LAPACK. Internally developed libraries
for calculating Fast Fourier Transforms (FFT's) and
Convolutions are also included, as well as direct linear
equation solvers for sparse symmetric linear systems of
equations.
Documentation is available per routine via individual
reference pages (man pages). A complete description of
ccccoooommmmpppplllliiiibbbb....ssssggggiiiimmmmaaaatttthhhh can be accessed via the complib man page.
General man pages for the BLAS (mmmmaaaannnn bbbbllllaaaassss), fft routines (mmmmaaaannnn