R-1.6.0; BLAS and friends

Peter Dalgaard BSA p.dalgaard@biostat.ku.dk
21 Sep 2002 10:50:04 +0200


Göran Broström <gb@stat.umu.se> writes:

> On Fri, 20 Sep 2002, Dirk Eddelbuettel wrote:
> 
> > On Fri, Sep 20, 2002 at 10:45:47PM +0200, Peter Dalgaard BSA wrote:
> > > G?ran Brostr?m <gb@stat.umu.se> writes:
> > >
> > > > R-1.6.0 beta on Debian Linux:
> >
> > Did you build it yourself, or did you grab the (currently weekly) .deb?
> 
> I built it myself. I checked with my Debian system at home, and,
> interestingly, there 'dtrmm_' is loaded. Then I reinstalled R-1.5.1, and
> 'dtrmm_' is still missing! So I get different results on different
> _machines_, not different versions of R. I have different setups of atlas
> on the two machines. Can that be the cause? Note that not all BLAS
> routines are missing, only a selected few. I configure with
> 
> ./configure --with-gnome --enable-R-shlib

Hmm. On Franz (alias cvs.r-project.org) I seem to be linking against
this guy

[pd@franz pd]$ file /usr/lib/3dnow/atlas/libblas.so.2.3
/usr/lib/3dnow/atlas/libblas.so.2.3: ELF 32-bit LSB shared object,
Intel 80386, version 1 (SYSV), stripped
                               ********

Could that have something to do with it? The corresponding .a file
does have various incarnations of dtrmm inside.


[pd@franz BUILD]$ ldd bin/R.bin
        libblas.so.2 => /usr/lib/3dnow/atlas/libblas.so.2 (0x40021000)
        libreadline.so.4 => /lib/libreadline.so.4 (0x403ba000)
        libdl.so.2 => /lib/libdl.so.2 (0x403e1000)
        libncurses.so.5 => /lib/libncurses.so.5 (0x403e4000)
        libm.so.6 => /lib/libm.so.6 (0x40422000)
        libz.so.1 => /usr/lib/libz.so.1 (0x40443000)
        libc.so.6 => /lib/libc.so.6 (0x40452000)
        /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)


-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._