Dear All,
recently I posted a probably difficult question concerning the installation and implementation of R-2.1.0 on the IBM p690 cluster "Jump" at the research centre in Jülich, Germany.
We identified several modifications for an installation such as:
(i) the installation of the most recent iconflib in the local user directory
(ii) the configuration command
OBJECT_MODE=64 MAKE=gmake CC="cc -DSTDC" SHLIB_LDFLAGS="-Wl,-G"
F77="xlf" PREFIX=/home3/zdvex/ufzaloe ./configure --without-x
--without-blas --without-readline --with-libiconv-prefix=
=/home3/zdvex/ufzaloe/lib
(iii) and the compilation command OBJECT_MODE=64 gmake
(thanks to Christoph Pospiech for help).
The really severe problem is, that the compilation now aboarts when making the survival R-library. This is what we get:
gmake[3]: Leaving directory `/work/.j36/R.INSTALL.3207208/survival/src'
** R ** data ** moving datasets to lazyload DB ** inst ** preparing package for lazy loading
(see attchment, zip archive of log-file).
Besides the OS-problems (like missing readline implementation) it seems as if we need good help from the R developers community, as this most recent bug (compilation of survival) seems to an R problem.
However, we are still strongly interested in getting this stuff running, any good suggestions very welcome!
Cheers
Ralf
This archive was generated by hypermail 2.1.8 : Mon 24 Oct 2005 - 22:26:55 GMT