[Rd] Problems with Lapack's eigen() on 64-bit

[Martin Maechler]

>>>>> "BDR" == Prof Brian Ripley <ripley at stats.ox.ac.uk>
>>>>>     on Tue, 27 Jul 2004 13:17:48 +0100 (BST) writes:

    BDR> Our `public' Opteron (dual 248, FC2 + Goto BLAS) is
    BDR> right, as is Solaris 64-bit. FC2 is using gcc version
    BDR> 3.3.3 20040412 (Red Hat Linux 3.3.3-7), quite a lot
    BDR> later.

    BDR> I would try updating your compiler (to 3.4.1?) and
    BDR> perhaps try a different BLAS.

I've now hand-compiled gcc 3.4.1 and recompiled R-patched and
R-devel both configured "--without-blas" and the eigen()
problem is gone.

For the record, note that before the problem appeared both for
"--with-blas=goto" and "--without-blas".

I've also added a regression test to R-devel which makes sure
that "make check" only passes when the mentioned eigen()
computation is correct.

Martin

    BDR> We do have RHEL for Opteron, but AFAIK it is not on a
    BDR> system at present.  (There are others running SuSE
    BDR> 9.0/.1, I believe.)

    BDR> On Tue, 27 Jul 2004, Martin Maechler wrote:

    >> I'm only now realizing that we have severe problems with
    >> R on our AMD 'Opteron' and 'Athlon64' clients running
    >> Redhat Enterprise with all 64-bit libraries (AFAICS).
    >> 
    >> The Lapack problem happens for R-patched and R-devel both
    >> on the Opteron and the Athlon64.
    >> 
    >> Here are platform details:
    >> 
    >> o "gcc -v" and "g77 -v" both end with the line gcc
    >> version 3.2.3 20030502 (Red Hat Linux 3.2.3-34)
    >> 
    >> o I've used ...../configure --without-blas
    >> 
    >> 1) Opteron ("deb7", a dual-processor compute server):
    >> 
    >> - uname -a : Linux deb7 2.4.21-9.0.3.ELsmp #1 SMP Tue Apr
    >> 20 19:44:29 EDT 2004 x86_64 x86_64 x86_64 GNU/Linux -
    >> /proc/cpuinfo contains (among much more) vendor_id :
    >> AuthenticAMD cpu family : 15 model : 5 model name : AMD
    >> Opteron(tm) Processor 248
    >> 
    >> 2) Athlon64 (a simple "new generation" client - to become
    >> my desktop soon):
    >> 
    >> - uname -a : Linux setup-12 2.4.21-15.0.2.EL #1 Wed Jun
    >> 16 22:41:44 EDT 2004 x86_64 x86_64 x86_64 GNU/Linux
    >> 
    >> - /proc/cpuinfo contains vendor_id : AuthenticAMD cpu
    >> family : 15 model : 14 model name : AMD Athlon(tm) 64
    >> Processor 2800+
    >> 
    >> ----------------
    >> 
    >> Now the Lapack problem, easily seen from the base eigen()
    >> example:
    >> 
    >> > eigen(cbind(1, 3:1, 1:3)) $values [1] 5.7015621
    >> 1.0000000 -0.7015621
    >> 
    >> $vectors [,1] [,2] [,3] [1,] 0.4877939 -0.7181217
    >> -0.9576161 [2,] 0.6172751 -0.3893848 0.2036804 [3,]
    >> 0.6172751 0.5767849 0.2036804
    >> 
    >> which is simply plainly wrong and eigen(cbind(1, 3:1,
    >> 1:3), EISPACK=TRUE) gives the correct eigen values c(5,
    >> 1, 0) and corresponding eigenvectors.
    >> 
    >> IIRC, we've already dealt with a Lapack problem, and that
    >> workaround (built into R-devel's Makefiles) has been to
    >> use -ffloat-store for the compilation of
    >> src/modules/lapack/dlamc.f
    >> 
    >> ----------
    >> 
    >> Thank you for hints / suggestions.
    >> 
    >> Others with 64-bit platforms might also try to see what
    >> eigen(cbind(1, 3:1, 1:3)) gives there.

    BDR> -- Brian D. Ripley, ripley at stats.ox.ac.uk Professor of
    BDR> Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
    BDR> University of Oxford, Tel: +44 1865 272861 (self) 1
    BDR> South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG,
    BDR> UK Fax: +44 1865 272595

    BDR> ______________________________________________
    BDR> R-devel at stat.math.ethz.ch mailing list
    BDR> https://www.stat.math.ethz.ch/mailman/listinfo/r-devel