R-beta: R-0.61.3 and __setfpucw.

Peter Dalgaard BSA p.dalgaard at biostat.ku.dk
Tue May 12 11:46:14 CEST 1998

Goran Brostrom <gb at stat.umu.se> writes:

> I have Linux, Redhat 4.2, on two machines, an i586 and a Sparc.
> When I compile R-0.61.3 on them, everything is fine on i586,
> but on the Sparc I get, in the last linking stage: 
> ../lib/libunix.a(system.o): In function `main':
> /usr/local/R-0.61.3/src/unix/system.c:353: undefined reference to
> `__setfpucw'
> ../lib/libunix.a(system.o): In function `R_CleanUp':
> /usr/local/R-0.61.3/src/unix/system.c:430: undefined reference to
> `__setfpucw'

Using __setfpucw is a horrible hack in the first place... 

You could try simply commenting it out. Or study <fpu_control.h> and
guess what the mechanism is supposed to be on Sparc Linux.

On i386 with RH4.2, _FPU_DEFAULT == _FPU_IEEE, so the __setfpucw() is
a no-op anyway. (Earlier Linuxen would unmask exceptions invalid-op,
zero-div, and overflow - halting programs at inconvenient times).

   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907

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