Thanks, that worked. Make finished without any errors on the R-patched_2006-05-10.
To get R-patched_2006-05-10 to pass make check on the AIX 5.2 here I added $(LIBM) as you suggested in the other posting for R-2.2.1 to the end of R_X11_la_LIBADD and vfonts_la_LIBADD in the X11 and vfonts modules makefiles respectively before running make. So those makefiles, namely .../src/modules/X11/Makefile and .../src/modules/vfonts/Makefile, have R_X11_la_LIBADD = $(ALL_X_LIBS) $(BITMAP_LIBS) $(LIBR) $(LIBM) vfonts_la_LIBADD = $(LIBR) $(LIBM) Similar editing for the R-2.2.1 version over here allows it to pass make check. Although for some reason the axis labels and tick marks for both R-2.2.1 and R-2.3.0 still seem darker and bold-faced on the x11 device compared to R-2.0.1. Don't know what to make of that, but looks like I can live with everything so far. Thanks, Jagat -----Original Message----- From: Prof Brian Ripley [mailto:[EMAIL PROTECTED] Sent: Friday, May 12, 2006 8:06 AM To: Sheth, Jagat K Cc: r-devel@stat.math.ethz.ch Subject: Re: [Rd] R-2.3.0 make error on AIX 5.2 We've seen something almost identical before. Looks like fcntl is redefining open (and not open()). At the top of the file is #ifdef HAVE_FCNTL_H # include <fcntl.h> #endif Insert #undef open in there. On Fri, 12 May 2006, Sheth, Jagat K wrote: > I am encountering a make error on an AIX 5.2 with R-2.3.0 and > R-patched_2006-05-10 after successfully configuring them using gcc-4.1.0, gnu > make, and the following configure options > > CC=/usr/local/bin/gcc > F77=/usr/local/bin/gfortran > CXX=/usr/local/bin/g++ > MAIN_LDFLAGS=-Wl,-brtl > SHLIB_LDFLAGS=-Wl,-G > CFLAGS='-g -O' > FFLAGS='-O' > CXXFLAGS='-g -O' > > bash-3.00$ ./configure --disable-nls > > The gcc specs on this machine are > > bash-3.00$ /usr/local/bin/gcc -v > Using built-in specs. > Target: powerpc-ibm-aix5.2.0.0 > Configured with: ../gcc-4.1.0/configure --disable-aix64 --disable-nls > Thread model: aix > gcc version 4.1.0 > > The make error from R-patched_2006-05-10 is > > make[4]: Entering directory > `/sas/data/cmi/ma/data/js36954/src/R-patched/src/mod > ules/internet' > /usr/local/bin/gcc -I. -I../../../src/include -I../../../src/include > -I/usr/loca > l/include -DHAVE_CONFIG_H -mno-fp-in-toc -g -O -c internet.c -o internet.o > internet.c: In function 'in_R_newurl': > internet.c:174: error: 'struct Rconn' has no member named 'open64' > make[4]: *** [internet.o] Error 1 > make[4]: Leaving directory > `/sas/data/cmi/ma/data/js36954/src/R-patched/src/modu > les/internet' > make[3]: *** [R] Error 2 > make[3]: Leaving directory > `/sas/data/cmi/ma/data/js36954/src/R-patched/src/modu > les/internet' > make[2]: *** [R] Error 1 > make[2]: Leaving directory > `/sas/data/cmi/ma/data/js36954/src/R-patched/src/modu > les' > make[1]: *** [R] Error 1 > make[1]: Leaving directory `/sas/data/cmi/ma/data/js36954/src/R-patched/src' > make: *** [R] Error 1 > > I haven't encountered any make errors on this machine when building R-2.0.1 > or R-2.2.1 (although the latter failed make check and didn't load the X11 and > vfonts modules unless $(LIBS) was added to their makefiles as mentioned in a > separate posting from this). > > Any advice on the above would be greatly appreciated. I will be happy to > provide other additional details. > > Thanks, > Jagat > > Jagat K. Sheth 1000 Technology Drive >> Mortgage Analytics Third Floor, Mail Station 55 >> CitiMortgage, Inc. O'Fallon, MO 63368-2240 >> Tel:(636)261-1407 FAX:(636)261-1312 >> >> Email: [EMAIL PROTECTED] >> >> >> >> >> >> > > ______________________________________________ > R-devel@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-devel > > -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595 ______________________________________________ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
