[R] core dump during make check when building 64-bit R on Solaris8/9

[Prof Brian Ripley]

On Fri, 3 Dec 2004, CHAN Chee Seng wrote:

> Hi All,
>
> Thanks for all the replies.  As mentioned in first mail I am using Sun
> Studio 8, which is giving me the problem.  I am glad that there are
> people who got this to work.  However SS8 is all I have right now and
> since there is a problem I would like to get this fixed somehow.

You already have a fix: do not use libsunperf if it is buggy.

> And Peter,
>
> Yes, that's looks like the problem that I am facing (SEGV and the stack
> trace).
> My cc -V looks like this:
>
> cc: Sun C 5.5 2003/03/12
>
> I am using /opt/SUNWspro/lib/v9/libsunperf.so.4.  And I am not aware of
> the difference between v9, v9a and v9b.

They are variations on the Sparc architecture.

> Can you write/send me the standalone program to show the problem so I
> can send to some Sun people that I know?
>
> BTW, how did you run R under dbx?  (Sorry but I am not familiar with
> using dbx.)  I get this error:
>
> Fatal error: R home directory is not defined
>
> Thanks.
>
> Regards,
> Chee Seng
>
> -----Original Message-----
> From: Prof Brian Ripley [mailto:ripley at stats.ox.ac.uk]
> Sent: Wednesday, December 01, 2004 7:08 PM
> To: Peter Dalgaard
> Cc: CHAN Chee Seng; r-help at stat.math.ethz.ch
> Subject: Re: [R] core dump during make check when building 64-bit R on
> Solaris8/9
>
> On Wed, 1 Dec 2004, Peter Dalgaard wrote:
>
>> Prof Brian Ripley <ripley at stats.ox.ac.uk> writes:
>>
>>> We know of at least two builds which successfully used libsunperf,
> one
>>> (mine) with Studio ONE 7 and one with 9.  So this may well be a
>>> problem with your particular compiler suite.
>>
>> Or maybe libsunperf version issues. The one that caused me trouble was
>
> libsunperf *is* part of the compiler suite, as your path below shows.
>
>> this one
>>
>> /opt/FD8/SUNWspro/lib/v9b/libsunperf.so.4
>>
>> (on the DTU machines)
>
> I am using
>
> /opt/SunONE7/SUNWspro/lib/v9/libsunperf.so.4
>
> The precise architecture selected comes into play, too (v9 vs v9b).
>
> My point was that we do know of working examples.
>
> -- 
> Brian D. Ripley,                  ripley at stats.ox.ac.uk
> Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
> University of Oxford,             Tel:  +44 1865 272861 (self)
> 1 South Parks Road,                     +44 1865 272866 (PA)
> Oxford OX1 3TG, UK                Fax:  +44 1865 272595
>
>

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595