#9040: fatal relocation error installing R on OpenSolaris
------------------------+---------------------------------------------------
   Reporter:  drkirkby  |       Owner:  drkirkby                                
    
       Type:  defect    |      Status:  new                                     
    
   Priority:  major     |   Milestone:  sage-4.5.2                              
    
  Component:  solaris   |    Keywords:                                          
    
     Author:            |    Upstream:  None of the above - read trac for 
reasoning.
   Reviewer:            |      Merged:                                          
    
Work_issues:            |  
------------------------+---------------------------------------------------

Comment(by drkirkby):

 Replying to [comment:4 jhpalmieri]:
 > This also seems like an issue on Solaris on x64 (not just !OpenSolaris).
 At least, I can't build R on fulvia, and I get a similar error message.

 You do not surprise me there. Page 49 of the
 [http://cran.r-project.org/doc/manuals/R-admin.pdf R Installation and
 Administration Guide] says in the section on Solaris:

 ''Tests with gcc4 on x86 and amd64 have been less successful: x86 builds
 have failed on tests using complex arithmetic, whereas on amd64 the builds
 have failed to complete in several different ways, most recently with
 relocation errors for libRblas.so'' I'm not sure if R is built with the
 -fPIC option, but if not, that is likely to be the cause of relocation
 errors. I finally convinced the Pari developers that one should use -fPIC
 on shared libraries.

 Since R runs as a stand-alone program, building with the recommended Sun
 compilers should not cause any linking problems.

 It means one needs to have two sets of compilers installed, which is a bit
 annoying, but there may be no way around it. I guess if we had a patch
 that worked with gcc, we could submit that to the R developers. I thought
 I'd worry about the R issue when the rest of Sage was building on this
 hardware, since R is not a critical component to getting something
 working.

 An older version of the Sun compilers is installed on 't2' in the
 directory {{{/opt/SUNWspro}}}. The C compiler is
 {{{/opt/SUNWspro/bin/cc}}} and the C++ compiler is
 {{{/opt/SUNWspro/bin/CC}}}. That might be sufficient. It is not the latest
 !SunStudio, but I doubt one needs to have the latest version. If it is
 needed, !SunStudio is a free download, though there's a lack of disk space
 on t2, so I am not keen to install it now.

 BTW, there's several errors in the R guide about Solaris, which I have
 raised, but I've had no positive response about. First I sent to the
 r-help, but was told they should go to r-devel or the bug database. Then
 after sending them to r-devel it was ignored. Then a reminder got ignored
 too. I guess I should submit them to the bug database.

 Dave

-- 
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/9040#comment:5>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica, 
and MATLAB

-- 
You received this message because you are subscribed to the Google Groups 
"sage-trac" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/sage-trac?hl=en.

Reply via email to