brat on darwin

From: Stephen Sanders (ssanders@ku.edu)
Date: Mon Sep 09 2002 - 11:11:37 EDT

  • Next message: Truls Martin Larsen: "Brat 2-4-6"

    Hi,
    
    For the Mac fans out there, a quick rundown on what is needed to
    get brat going [this is not a detailed, blow-by-blow description, I did
    have to iterate on a couple of these items, but no more so than
    a new linux installation usually takes...]:
    
    Use OS10.1 with the April 2002 developer tools (available from the
    apple developer website--you need to register, but the tools are free).
    
    Get fink  (http://fink.sourceforge.net).  I downloaded the full system 
    with the
    exception of the ssh stuff and some of the mutually exclusive parts.  
    Everything
    is installed by default in a /sw directory, so you need to make some
    changes to your path.  This is described on the fink site.
    
    Fink also tells you why you DO NOT YET want to switch to 10.2.
    
    Get CERNLIB by following the links given for the Macintosh build of root.
    Fink supplies all of the non-cernlib related programs that this site 
    tells you
    that you need.  I downloaded the compiled libraries.  I did not attempt
    the build from source.
    
    Get the latest version of the root source and make sure you can get it to
    build.  If root fails, brat almost certainly will fail.
    
    Get the latest brat and build it. THE MAKE STEP WILL FAIL UNLESS YOU SET
    THE CONFIGURE FLAG:  --disable-dependency-tracking
    
    Enjoy.
    
    All of the brat tests that work on my linuxppc system give the same 
    results
    using darwin.  The eventmanager test segfaults on both systems, but I 
    haven't
    spent any time trying to figure out the problem.  The calibrations 
    script that I
    use for the si, tiles, and bb arrays works fine.
    
    Regards,
    Steve
    



    This archive was generated by hypermail 2b30 : Mon Sep 09 2002 - 11:12:28 EDT