Skip to content

shared ATLAS library?

4 messages · Paul Gilbert, Roger D. Peng, Dirk Eddelbuettel

Original
Paul Gilbert
# 
Note that the BLAS library will be used for several add-on packages as 
   well as for R itself. This means that it is better to use a
   shared BLAS library, as most of a static library will be compiled 
   into the R executable and each BLAS-using package. 

Is there an idiots guide on how to do this (or should idiots just not try). I've
compiled atlas as a static (.a) library and successfully built R (although, when
it is static, I don't know how to tell for sure that it is being used). However,
their seems to be little indication of how to make a shared (.so) ATLAS library.
I've had even less success with blas from netlib. (I compiled .so files and
generated a file libblas.so, but there are a lot of unresolved symbols, so R
crashes.)

Any help would be appreciated. (Linux or Solaris).

Thanks,
Paul Gilbert
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Roger D. Peng
# 
I think you can use nm on the R.bin file -- if you compiled with static
ATLAS there should a bunch of ATL_* symbols.

-roger
_______________________________
UCLA Department of Statistics
rpeng at stat.ucla.edu
http://www.stat.ucla.edu/~rpeng
On Fri, 15 Nov 2002, Paul Gilbert wrote:

            

      
      
      
    

        
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

  
  


  


      

    

    

      
      

        


  

        

      Roger D. Peng
    

    

      
      #
    

  


  

      
You may want to check out

http://www.r-project.org/nocvs/mail/r-help/2002/7917.html

from the R-help archive.  It doesn't exactly look "straightforward" but
this may point you in the right direction.  Can't say I've done it myself.

-roger
_______________________________
UCLA Department of Statistics
rpeng at stat.ucla.edu
http://www.stat.ucla.edu/~rpeng

        
On Fri, 15 Nov 2002, Paul Gilbert wrote:

        

            

      
      
      
    

        
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

  
  


  


      

    

    

      
      

        


  

        

      Dirk Eddelbuettel
    

    

      
      #
    

  


  

      
On Fri, Nov 15, 2002 at 05:50:21PM -0500, Paul Gilbert wrote:

            

      
      
      
    

        
For Linux, I feel compelled to plug in Camm's great work for Atlas under
Debian. It simply works out of the box, and my R packages use it. No
configuration on the user part whatsoever. As for Solaris, can't help
there...

Dirk