OK, this is really a new one for me. Just thought I'd try building R-devel
(2005-03-15) on our AIX power machines and see if there are any problems. I
used the settings in the Appendix B of R-admin, using the set of variables
that Tim Hoar used (i.e., trying to build 64-bit R using native compilers).
configure ran fine, but `gmake' gave me a syntax error when compiling
src/main/memory.c. The offending line seems to be:
Rboolean success = FALSE;
Can anyone provide pointers?
(Just to make sure, I tried building it on our SLES9-x86_64 box, and that
passes make check.)
TIA!
Andy
syntax error in memory.c when building R-devel on AIX
3 messages · Liaw, Andy, Brian Ripley, Luke Tierney
That statement was added yesterday (not by me) and needed moving up a line. Which version do you have? The current one has Rboolean success = FALSE; s = NULL; /* initialize to suppress warning */ and passes all my tests. The previous one had those two lines reversed and worked on gcc (with a warning) but failed on Solaris' cc (and was not legal ISO C90).
On Tue, 15 Mar 2005, Liaw, Andy wrote:
OK, this is really a new one for me. Just thought I'd try building R-devel (2005-03-15) on our AIX power machines and see if there are any problems. I used the settings in the Appendix B of R-admin, using the set of variables that Tim Hoar used (i.e., trying to build 64-bit R using native compilers). configure ran fine, but `gmake' gave me a syntax error when compiling src/main/memory.c. The offending line seems to be: Rboolean success = FALSE; Can anyone provide pointers? (Just to make sure, I tried building it on our SLES9-x86_64 box, and that passes make check.)
Brian D. Ripley, ripley@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
On Tue, 15 Mar 2005, Prof Brian Ripley wrote:
That statement was added yesterday (not by me) and needed moving up a line. Which version do you have? The current one has Rboolean success = FALSE; s = NULL; /* initialize to suppress warning */ and passes all my tests. The previous one had those two lines reversed and worked on gcc (with a warning) but failed on Solaris' cc (and was not legal ISO C90).
That was me--sorry. gcc -Wall does not warn -- I've added -pedantic to my .Rconf; that does warn. luke
On Tue, 15 Mar 2005, Liaw, Andy wrote:
OK, this is really a new one for me. Just thought I'd try building R-devel (2005-03-15) on our AIX power machines and see if there are any problems. I used the settings in the Appendix B of R-admin, using the set of variables that Tim Hoar used (i.e., trying to build 64-bit R using native compilers). configure ran fine, but `gmake' gave me a syntax error when compiling src/main/memory.c. The offending line seems to be: Rboolean success = FALSE; Can anyone provide pointers? (Just to make sure, I tried building it on our SLES9-x86_64 box, and that passes make check.)
Luke Tierney
Chair, Statistics and Actuarial Science
Ralph E. Wareham Professor of Mathematical Sciences
University of Iowa Phone: 319-335-3386
Department of Statistics and Fax: 319-335-3017
Actuarial Science
241 Schaeffer Hall email: luke@stat.uiowa.edu
Iowa City, IA 52242 WWW: http://www.stat.uiowa.edu