As I said before I experience difficulties to evaluate what configure has found.
The place in R's config.log, where 'cpow' fails, sais: ------------------------ configure:32656: checking for cpow configure:32721: gcc -o conftest -g -O2 -D__NO_MATH_INLINES -I/usr/local/include -L/usr/local/lib conftest.c -lm >&5 conftest.c:165: warning: conflicting types for built-in function 'cpow' /var/tmp//cc4mUk2D.o(.text+0x14): In function `main': /usr/local/R-beta/conftest.c:189: undefined reference to `cpow' /var/tmp//cc4mUk2D.o(.data+0x0):/usr/local/R-beta/conftest.c:188: undefined reference to `cpow' configure:32727: $? = 1 configure: failed program was: | /* confdefs.h. */ [...] ------------------------ It is almost the same for the other references. If desired, I can send config.log. In FreeBSD's '/usr/include' I found the header-file 'tgmath.h' which defines the following: #define pow(x, y) __tg_impl_full(x, x, y, pow, powf, powl,\ cpow, cpowf, cpowl, x, y) #define carg(x) __tg_simple(x, carg) Up to now I have no idea, where the rest (clog, cexp, csqrt ...) is defined in FreeBSD. I am not familiar with configure scripts, autoconf and automake. So I am afraid I need more guidance to progress. Rainer Hurling > Prof Brian Ripley wrote: > These were found by AC_CHECK_FUNCS (please confirm what configure said) > so most likely some macro needs to be set or header included. > > Could you please find out how configure managed to find cpow etc when > they appear not to be in libc/libm? ______________________________________________ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel