LAPACK Archives

[Lapack] Hi, Dear LAPACK Team

Dear Xiangyu,
It seems that CLAPACK is having some problem with complex precision on AMD 
machines.
We did not manage to isolate the problem at the moment.
I would advice to maybe switch to the FORTRAN LAPACK and if you prefer 
programming in C, just use the C Interface to LAPACK. Both are available on the 
LAPACK website: http://www.netlib.org/lapack/
Regards,
Julie
On Dec 10, 2010, at 10:22 AM, Xiangyu Liu wrote:

Hi, Dear LAPACK Team,

    Excuse me, I'm from  ShenZhen.China(Mainland), and I've download 
CLAPACK3.2.1 package from netlib.org to my home PC (Debian in 
Chinese[simplified]) for my personal interest ,  and read the 
readme.installed and tried to install clapck , but I met trouble.

    First, I used ATLAS3.9.32 (from sourceforge.net) to generate a tuned 
optimized BLAS, and I tested atlas library(libcblas.a, libf77blas.a, 
libatlas.a) successly in ATLAS test procedure. But when I set make.inc by 
"BLASLIB = -lf77blas -latlas -lgfortran" and "CC = gcc-4.5 -DNO_BLAS_WRAP", I 
got a error message in blas_testing as "???" in Chinese (it means 
"segmentation fault") in ./xblat1c and ./xblat1z run time.

    Then, (after make cleanall) I reset make.inc by "CC = gcc-4.5", and 
"BLASLIB = ../../libfblaswr.a -lf77blas -latlas -lgfortran" , but there is 
still error messages as "???" (segmentation fault).

    So, I reset make.inc by "CC = gcc-4.5 -DNO_BLAS_WRAP" and "BLASLIB = 
../../blas$(PLAT).a", and I built blas library within clapack. There was no 
any error message in link and run time, but I checked the blas testing 
result(c/d/s/z-blat-1/2/3.out) and found there were some "FATAL ERROR"s in 
these .out file (espetially in sblat*.out, single precision complex number). 
I've send these .out file to you within patch (tar.gz format).

    Some of my friends have install clapack successly with optimized BLAS 
(ATLAS, ACML,etc) in different system (Ubuntu, Fedora, Debian), and now they 
are successful and enjoying this matrix library. But I, (set my make.in as 
same as theirs) failed again and again. I know I must miss something , but I 
don't know what it exactly is. 

   Would you mind to tell me some advices or check my makefile and test 
result ? Or my compiler environment and configuration has some bugs?



Appendix : My PC&gcc configuration

        Processor        : AMD Sempron(tm)   2200+ (K6 or K7 ?)
        Memory        : 1034MB
        Operating System        : Debian GNU/Linux squeeze/sid
        -Version-
            Kernel        : Linux 2.6.33.7 (i686)
            Compiled        : #1 SMP Wed Nov 3 13:20:26 CST 2010
        C Library        : GNU C Library version 2.11.2 (stable)
        Distribution        : Debian GNU/Linux squeeze/sid
        Desktop Environment        : GNOME 2.20 (session name: Default)
        Compiler            : gcc 
        - Version- 4.5
            Using built-in specs.
            COLLECT_GCC=gcc-4.5
            COLLECT_LTO_WRAPPER=/usr/lib/gcc/i486-linux-gnu/4.5.1/lto-wrapper
            Target: i486-linux-gnu
            Configured with: ../src/configure -v --with-pkgversion='Debian 
4.5.1-9' --with-bugurl=file:///usr/share/doc/gcc-4.5/README.Bugs 
--enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr 
--program-suffix=-4.5 --enable-shared --enable-multiarch 
--enable-linker-build-id --with-system-zlib --libexecdir=/usr/lib 
--without-included-gettext --enable-threads=posix 
--with-gxx-include-dir=/usr/include/c++/4.5 --libdir=/usr/lib --enable-nls 
--enable-clocale=gnu --enable-libstdcxx-debug --enable-libstdcxx-time=yes 
--enable-plugin --enable-gold --with-plugin-ld=ld.gold --enable-objc-gc 
--enable-targets=all --with-arch-32=i586 --with-tune=generic 
--enable-checking=release --build=i486-linux-gnu --host=i486-linux-gnu 
--target=i486-linux-gnu
Thread model: posix
gcc version 4.5.1 (Debian 4.5.1-9) 



  Thanks,
  Best Regards,
  Xiangyu Liu 
<clapack_record.tar.bz2>_______________________________________________
Lapack mailing list
Lapack@Domain.Removed
http://lists.eecs.utk.edu/mailman/listinfo/lapack

**********************************************
Julie Langou; Research Associate in Computer Science
Innovative Computing Laboratory;
University of Tennessee from Denver, Colorado ;-)
julie@Domain.Removed; http://www.cs.utk.edu/~julie/







-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
http://lists.eecs.utk.edu/mailman/private/lapack/attachments/20101214/b571b5c2/attachment.html
 

<Prev in Thread] Current Thread [Next in Thread>


For additional information you may use the LAPACK/ScaLAPACK Forum.
Or one of the mailing lists, or