R help - Jan 2003 - dyn.load warning message in R1.6.2 on Windows XP

Hi folks:

I used to load a DLL fine up until R 1.5.1.  The same DLL dyn.loaded under R
1.6.2 prints the following warning message:
> dyn.load("C:/Rajiv/Bin/rpdate.dll")
Warning message: 
DLL attempted to change FPU control word from 8001f to 9001f
>
Any indication as to what the warning means, and whether I should be
concerned about it?  The DLL is made from a pure C code, compiled with MS
Visual C++ 6.0.
> version
         _              
platform i386-pc-mingw32
arch     i386           
os       mingw32        
system   i386, mingw32  
status                  
major    1              
minor    6.2            
year     2003           
month    01             
day      10             
language R              
>
Thanks.

Rajiv
--------
Rajiv Prasad
Postdoctoral Research Associate, Hydrology Group
Pacific Northwest National Laboratory, P.O. Box 999, MSIN K9-33
Richland, WA 99352
Voice: (509) 375-2096  Fax: (509) 372-6089  Email: rajiv.prasad at pnl.gov

"Prasad, Rajiv" <Rajiv.Prasad at pnl.gov> writes:
> Hi folks:
> 
> I used to load a DLL fine up until R 1.5.1.  The same DLL dyn.loaded under
R
> 1.6.2 prints the following warning message:
> 
> > dyn.load("C:/Rajiv/Bin/rpdate.dll")
> Warning message: 
> DLL attempted to change FPU control word from 8001f to 9001f 
> >
> 
> Any indication as to what the warning means, and whether I should be
> concerned about it?  The DLL is made from a pure C code, compiled with MS
> Visual C++ 6.0.
The check is new, but the problem is likely older. Before the check,
it would seem that your DLL was meddling with the FPU and not putting
the settings back as it found them. Now, R detects the change and sets
things back, which is probably better (I think Duncan M. told us that
that particular bit in the control word reduces floating point
precision from 64 to 53 bits), but your DLL shouldn't do that in the
first place. As far as I know, we can only detect the change upon
loading the DLL, not when running the functions inside of it, so you
should be somewhat worried.

-- 
   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