[R] Version 1.5.1 STILL failing make check.

Rolf Turner rolf at math.unb.ca
Thu Sep 5 22:11:28 CEST 2002


I tried setting CFLAGS="-g -O" in config.site and got the same errors
from running ``make check''.  Then I tried setting CC=cc in
config.site, and the damned thing failed again (but differently).
Now at least ``!is.nan(var(1))'' is TRUE, as it should be.  The
problem comes from lapack.  At the end of

	.../test/Examples/base-Ex.Rout.fail

we find:

	> loc <- cmdscale(eurodist)
	Error in La.eigen(-x/2, symmetric = TRUE) : 
		lapack routines cannot be loaded
	Execution halted

Which I guess I should've expected --- it couldn't be so simple as to
change compilers from gcc to cc, could it?  Have I done something
naive in making the change to config.site?  Should I have made some
other changes as well? In particular I left the Fortran compiler as
g77, rather than changing it to f77 --- is this the wrong thing to
do?

I notice that in config.log I get a lot of sequences of warning
messages the form

	/dev/null:0: malformed option `-A system=unix'
	/dev/null:0: malformed option `-A system=svr4'
	*Initialization*:1: warning: `__SIZE_TYPE__' redefined
	*Initialization*:1: warning: `__PTRDIFF_TYPE__' redefined
	*Initialization*:1: warning: `__WCHAR_TYPE__' redefined
	/dev/null:0: malformed option `-A cpu=sparc'
	/dev/null:0: malformed option `-A machine=sparc'

This does not encourage me, but I haven't a clue what's going on.

Why in hell won't our system build R???  I'm going nuts over this.

				cheers (but not very cheerfully),

					Rolf Turner
					rolf at math.unb.ca
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list