[Rd] Comments on the R-2.0.0 release (PR#7351)
stefano iacus
stefano.iacus at unimi.it
Mon Nov 8 20:43:52 CET 2004
On Nov 8, 2004, at 7:41 PM, beebe at math.utah.edu wrote:
> Builds of the R-2.0.0 release have been considerably more successful
> at my site than previous releases. I now have it installed on these
> platforms:
>
> Apple PowerPC G3 267MHz GNU/Linux 2.4.19-4a (Yellow Dog Linux release
> 2.3 (Dayton))
> Compaq Alpha Sierra OSF/1 5.1
> Compaq/DEC Alpha OSF/1 4.0F
> Intel Itanium-2 GNU/Linux Red Hat Linux Advanced Server release
> 2.1AS (Derry)
> Intel Pentium II FreeBSD 4.4-RELEASE #0
> Intel Pentium III FreeBSD 5.0
> Intel Pentium III GNU/Linux 2.4.18-26.8.0smp (Red Hat 8.0 (Psyche))
> Intel Pentium III GNU/Linux 2.4.21-20.ELsmp (Red Hat Enterprise Linux
> WS release 3 (Taroon Update 3))
> Intel Pentium III NetBSD 1.6
> Intel Pentium III OpenBSD 3.2
> Intel Pentium III Solaris 9 x86
> SGI Origin 200 IRIX 6.5
> Sun SPARC GNU/Linux 2.2.19-6.2.1 (Red Hat 6.2)
> Sun SPARC Solaris 2.7
> Sun SPARC Solaris 2.8
> Sun SPARC Solaris 2.9
>
> However, builds initially failed on several of these systems because
> of errors like this:
>
> /usr/bin/ld: cannot find -lX11
>
> On these systems, I was able to restart the build with
>
> make LDFLAGS=-L/usr/X11R6/lib
>
> and have it complete successfully. I've recently seen some other
> packages that failed to compile for the same reason: the list of
> standard locations that is searched at configure time for X11
> libraries is out-of-date. The R-2.0.0 configure.ac script has this
> snippet:
>
> for dir in /usr/bin/X11 /usr/X11R6/bin /usr/openwin/bin; do
> r_xtra_path="${r_xtra_path}${PATH_SEPARATOR}${dir}"
> done
>
> However, it does not seem to have an analogue for the corresponding
> lib directory.
>
> Also, on some systems, I had to switch from using the native C/C++
> compilers to GNU gcc/g++.
>
> R-2.0.0 has failed to build on these systems:
>
> Apple PowerPC G4 1.4GHz Darwin 7.3.0 (MacOS 10.3.2 (7D26))
> SGI Indy-200 GNU/Linux Gentoo 1.4.16
> IBM PowerPC AIX 4.2
>
> On the MacOS 10 system, builds with the native cc (== gcc 3.3
> 20030304) and with the latest gcc-3.4.3 fail like this:
>
can you please tell us if you gcc-3.4.3 comes from? -framework Carbon,
which causes the error apparently, should work (and it works for
gcc-3.5).
gcc --v ?
would be nice to know.
stefano
> gcc -dynamiclib -L/usr/local/lib -dynamiclib -install_name libR.dylib
> \
> -compatibility_version 2.0.0 -current_version 2.0.0 \
> -headerpad_max_install_names -framework Carbon -o libR.dylib \
> Rembedded.lo CConverters.lo CommandLineArgs.lo Rdynload.lo
> Renviron.lo \
> RNG.lo apply.lo arithmetic.lo apse.lo array.lo attrib.lo base.lo \
> bind.lo builtin.lo character.lo coerce.lo colors.lo complex.lo \
> connections.lo context.lo cov.lo cum.lo dcf.lo datetime.lo debug.lo \
> devPS.lo devPicTeX.lo deparse.lo deriv.lo devices.lo dotcode.lo \
> dounzip.lo dstruct.lo duplicate.lo engine.lo envir.lo errors.lo \
> eval.lo format.lo fourier.lo gram.lo gram-ex.lo graphics.lo \
> identical.lo internet.lo iosupport.lo lapack.lo list.lo logic.lo \
> main.lo mapply.lo match.lo memory.lo model.lo names.lo objects.lo \
> optim.lo optimize.lo options.lo par.lo paste.lo pcre.lo platform.lo \
> plot.lo plot3d.lo plotmath.lo print.lo printarray.lo printvector.lo \
> printutils.lo qsort.lo random.lo regex.lo registration.lo relop.lo \
> saveload.lo scan.lo seq.lo serialize.lo size.lo sort.lo source.lo \
> split.lo sprintf.lo startup.lo subassign.lo subscript.lo subset.lo \
> summary.lo sysutils.lo unique.lo util.lo version.lo vfonts.lo
> xxxpr.lo \
> `ls ../appl/*.lo ../nmath/*.lo ../unix/*.lo 2>/dev/null` -lblas \
> -L/usr/local/lib \
> -L/export/local/powerpc-macosx/bin/../lib/gcc/powerpc-apple-
> darwin7.5.0/3.4.2 \
> -L/export/local/powerpc-macosx/bin/../lib/gcc \
> -L/usr/local/lib/gcc/powerpc-apple-darwin7.5.0/3.4.2 \
> -L/export/local/powerpc-macosx/bin/../lib/gcc/powerpc-apple-
> darwin7.5.0/3.4.2/../../.. \
> -L/usr/local/lib/gcc/powerpc-apple-darwin7.5.0/3.4.2/../../..
> -lfrtbegin \
> -lg2c -lgcc_s -lSystem ../extra/zlib/libz.a ../extra/bzip2/libbz2.a \
> ../extra/pcre/libpcre.a -lreadline -lcc_dynamic -lncurses -lm \
> -lcc_dynamic \
> gcc: Carbon: No such file or directory
>
> On the GNU/Linux MIPS system, there is as yet no working Fortran
> compiler, and f2c fails to build there, so R cannot be expected to
> run.
>
> On the IBM AIX system, I get this failure:
>
> gcc -Wl,-bM:SRE -Wl,-H512 -Wl,-T512 -Wl,-bnoentry -Wl,-bexpall \
> -Wl,-bI:../../../etc/R.exp -Wl,-L/usr/local/lib \
> -Wl,-bE:../../../etc/Rlapack.exp -o libRlapack.so dlamc.lo \
> dlapack0.lo dlapack1.lo dlapack2.lo dlapack3.lo cmplx.lo \
> cmplxblas.lo \
> -L/usr/local/lib/gcc-lib/powerpc-ibm-aix4.2.1.0/2.95.3 \
> -L/usr/local/powerpc-ibm-aix4.2.1.0/lib -L/usr/local/lib -lg2c \
> -lm \
> /usr/local/lib/gcc-lib/powerpc-ibm-aix4.2.1.0/2.95.3/libgcc.a \
> -lg
> ld: 0711-224 WARNING: Duplicate symbol: .lsame_
> ld: 0711-224 WARNING: Duplicate symbol: lsame_
> ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain more
> information.
> make[4]: Leaving directory `/local/build/R-2.0.0/src/modules/lapack'
>
> ...
>
> gcc -Wl,-bM:SRE -Wl,-H512 -Wl,-T512 -Wl,-bnoentry -Wl,-bexpall \
> -Wl,-bI:../../../etc/R.exp -Wl,-L/usr/local/lib -o lapack.so \
> -Wl,-bI:../../../etc/Rlapack.exp Lapack.lo rgeev.lo rsyev.lo \
> -L../../../lib -lRlapack \
> -L/usr/local/lib/gcc-lib/powerpc-ibm-aix4.2.1.0/2.95.3 \
> -L/usr/local/powerpc-ibm-aix4.2.1.0/lib -L/usr/local/lib -lg2c -lm \
> /usr/local/lib/gcc-lib/powerpc-ibm-aix4.2.1.0/2.95.3/libgcc.a -lg \
> -lreadline -ldl -lncurses -lm -lc \
> ld: 0706-006 Cannot find or open library file: -l Rlapack \
> ld:open(): No such file or directory
> collect2: ld returned 255 exit status
> make[4]: Leaving directory `/local/build/R-2.0.0/src/modules/lapack'
>
> Shared libraries (*.so) are uncommon on AIX 4.2; there are none at all
> in /usr/lib or /lib or /usr/lpp/X11/lib/, and only a few locally-added
> ones in /usr/local/lib.
>
> -----------------------------------------------------------------------
> --------
> - Nelson H. F. Beebe Tel: +1 801 581 5254
> -
> - University of Utah FAX: +1 801 581 4148
> -
> - Department of Mathematics, 110 LCB Internet e-mail:
> beebe at math.utah.edu -
> - 155 S 1400 E RM 233 beebe at acm.org
> beebe at computer.org -
> - Salt Lake City, UT 84112-0090, USA URL:
> http://www.math.utah.edu/~beebe -
>
> ______________________________________________
> R-devel at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>
More information about the R-devel
mailing list