desperately trying to build Maxima



>>>>> "Valerij" == Valerij Pipin <vpipin@irk.ru> writes:

    Valerij> Hi,
    Valerij> On 07 May 2002 11:09:12 -0500
    Valerij> James Amundson <amundson@fnal.gov> wrote:

    >> The cvs version of Maxima will compile, install and run cleanly with
    >> Clisp 2.28 or CMUCL 18d. 
    Valerij> I also would like to have cvs-maxima running cleanly on my linux box. It was indeed compiled without problem
    Valerij>  with clisp-2.28. However, I cannot say it is running cleanly. Some problems I found are
    Valerij> 1) online help does not work, e.g., 
    Valerij> Maxima 5.9.0pre-cvs (with enhancements by W. Schelter).
    Valerij> Licensed under the GNU Public License (see file COPYING)
    Valerij> (C1) describe(matrix);

    Valerij>  0: AUGCOEFMATRIX :(maxima.info)Definitions for Matrices and Linear Algebra.
    Valerij>  1: COEFMATRIX :Definitions for Matrices and Linear Algebra.
    Valerij>  2: COPYMATRIX :Definitions for Matrices and Linear Algebra.
    Valerij>  3: DIAGMATRIX :Definitions for Matrices and Linear Algebra.
    Valerij>  4: EMATRIX :Definitions for Matrices and Linear Algebra.
    Valerij>  5: ENTERMATRIX :Definitions for Matrices and Linear Algebra.
    Valerij>  6: GENMATRIX :Definitions for Matrices and Linear Algebra.
    Valerij>  7: MATRIX :Definitions for Matrices and Linear Algebra.
    Valerij>  8: MATRIX_ELEMENT_ADD :Definitions for Matrices and Linear Algebra.  9: MATRIX_ELEMENT_MULT :Definitions for Matrices and Linear Algebra. 10: MATRIX_ELEMENT_TRANSPOSE :Definitions for Matrices and Linear Algebra.
    Valerij>  11: MATRIXMAP :Definitions for Matrices and Linear Algebra.
    Valerij>  12: MATRIXP :Definitions for Matrices and Linear Algebra.
    Valerij>  13: SCALARMATRIXP :Definitions for Matrices and Linear Algebra.
    Valerij>  14: SUBMATRIX :Definitions for Matrices and Linear Algebra.
    Valerij>  15: ZEROMATRIX :Definitions for Matrices and Linear Algebra.
    Valerij> Enter n, all, none, or multiple choices eg 1 3 : 7;

    Valerij> Info from file /usr/local/info/maxima.info:
    Valerij> *** - SBIT: index 1060 for #*0000000000001000000000000000000100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 is out of range

This is weird.  It works fine for me with clisp on Solaris.  Don't
know what the problem could be.  Perhaps a mangled info file?

    Valerij> I did not try cmucl. People said it is fast. I can
    Valerij> compare only gcl and clisp.  On some tensor manipulations
    Valerij> (with itensor) gcl is factor 5 or more faster than clisp.

If the factor of 5 is really important to you, then CMUCL may help.
Don't know if it will be faster or slower.

Ray