On 29 November 2008 at 13:04, Martin Zobel-Helas wrote: | Hi, | | On Sat Nov 29, 2008 at 05:44:52 -0600, Dirk Eddelbuettel wrote: | > | > make[1]: Entering directory `/build/buildd/fasianoptions-270.74/src' | > | > gfortran -fpic -g -O2 -c EBMAsianOptions.f -o EBMAsianOptions.o | > | > gfortran -fpic -g -O2 -c GammaFunctions.f -o GammaFunctions.o | > | > gfortran -fpic -g -O2 -c HypergeometricFunctions.f -o HypergeometricFunctions.o | > | > gcc -std=gnu99 -shared -o fAsianOptions.so EBMAsianOptions.o GammaFunctions.o HypergeometricFunctions.o -llapack -lf77blas -latlas -lgfortran -lm -lgfortran -lm -L/usr/lib/R/lib -lR | > | > /usr/bin/ld: cannot find -lf77blas | > | > collect2: ld returned 1 exit status | > | > make[1]: *** [fAsianOptions.so] Error 1 | > | > Very strange. These compile/link options are determined when the main | > r-base-core package is built on the architecture. The Fortran version of the | > Blas would then be missing for _all_ other R package builds involving | > Fortran. Did you see any others fail? The buildd matrix shows fnonlinear | > failing in a similar way. | > | > However, looking at the build log of r-base 2.8.0-1 on the same arch, f77blas | > is used just fine: | > | > gfortran -fpic -O3 -pipe -g -c xxxpr.f -o xxxpr.o | > gcc -std=gnu99 -shared -Wl,-O1 -o libR.so CConverters.o CommandLineArgs.o Rdynload.o Renviron.o RNG.o apply.o apse.o arithmetic.o array.o attrib.o base.o bind.o builtin.o character.o coerce.o colors.o complex.o connections.o context.o cov.o cum.o dcf.o datetime.o debug.o deparse.o deriv.o devices.o dotcode.o dounzip.o dstruct.o duplicate.o engine.o envir.o errors.o eval.o format.o fourier.o gevents.o gram.o gram-ex.o graphics.o identical.o inlined.o internet.o iosupport.o lapack.o list.o localecharset.o logic.o main.o mapply.o match.o memory.o model.o names.o objects.o optim.o optimize.o options.o par.o paste.o pcre.o platform.o plot.o plot3d.o plotmath.o print.o printarray.o printvector.o printutils.o qsort.o random.o regex.o registration.o relop.o rlocale.o saveload.o scan.o seq.o serialize.o size.o sort.o source.o split.o sprintf.o startup.o subassign.o subscript.o subset.o summary.o sysutils.o unique.o util.o version.o vfonts.o xxxpr.o ../unix/Rembedded.o ../unix/libunix.a ../appl/libappl.a ../nmath/libnmath.a -lf77blas -latlas -lgfortran -lm -lreadline -lpcre -lbz2 -lz -ldl -lm | > make[5]: Entering directory `/build/buildd/r-base-2.8.0/src/main' | > mkdir -p -- /build/buildd/r-base-2.8.0/bin/exec | | Strange.... | | > Did anything change on your arch? | | Well... I took over the mipsel buildd two days ago so i can't tell you. | I just started a major rebuild of all packages which hade been in state | != installed. | | I am currently rebuilding mayer's unstable chroot, and will give | fasianoptions a try i the newly created chroot. Lets see if this works | out.
Did you get a chance to re-try it? Here is another cut and paste showing that the libf77blas which tripped the builder up is/was available before: * Installing *source* package 'fGarch' ... ** libs make[1]: Entering directory `/build/buildd/fgarch-280.75/src' gfortran -w -fpic -g -O2 -c dist.f -o dist.o gfortran -w -fpic -g -O2 -c llhGarch.f -o llhGarch.o gfortran -w -fpic -g -O2 -c math.f -o math.o gcc -std=gnu99 -shared -o fGarch.so dist.o llhGarch.o math.o -llapack -lf77blas -latlas -lgfortran -lm -lgfortran -lm -L/usr/lib/R/lib -lR make[1]: Leaving directory `/build/buildd/fgarch-280.75/src' make[1]: Entering directory `/build/buildd/fgarch-280.75/src' make[1]: Leaving directory `/build/buildd/fgarch-280.75/src' (that's from http://buildd.debian.org/fetch.cgi?pkg=fgarch;ver=280.75-1;arch=mipsel;stamp=1227832903 ) Dirk, still puzzled -- Three out of two people have difficulties with fractions.
