On Thu, 8 Apr 2004, Greg Hosler wrote: > Hi All, > > I've got a large C++ application that compiles/runs just fine on RH7.3 - I'm in > the process of porting it to RHEL3, and I am having some issues. > > My program has some usage of iostreams / fstream / etc C++ constructs. > > Some of the questions I have come across are: > > - I noticed that there are 2 sets of C++ header files. > one is "compat-libstdc++-devel-7.3-2.96.123", and the > other is "libstdc++-devel-3.2.3-24" > > Is there any rule of thumb as to which set of header files I should > use ? The compat headers are designed to be used with the compat-gcc-c++ package, and they are intended for programs compiled with g++296. The others are for the default g++, which is 3.3.2. > > - It seems that by default, libstdc++-devel-3.2.3-24 is being used. > > As I have a lot of pre ISO C99 constructs, this set of header files > seems to be giving me fits. If I use -I /usr/lib/g++-3, most of my > compilation errors go away (but leave me w/ link errors). Question > is there a non-default c++ library that I need to specify ? Use g++296 as your compiler instead of g++. > > - What are the implications, limitations, issues, with linking together > object files created with -I /usr/lib/g++-3, and without (i.e. with > "standard" g++ header files. i.e. should this work ? Is this known to > not work ? etc. C++ ABIs have been changing a lot over the recent history of the compiler. I think they are stable going forward now, but for older versions, the entire app must be compiled and linked with the same generation of compiler and linker. > > any thoughts, or tips (or pointers) would be much appreciated. > > thanks you, and best regards, > > -Greg Hosler > > +---------------------------------------------------------------------+ > You can release software that's good, software that's inexpensive, or > software that's available on time. You can usually release software > that has 2 of these 3 attributes -- but not all 3. > | Greg Hosler greg@xxxxxxxxxxxxx | > +---------------------------------------------------------------------+ > > > > -- Matthew Saltzman Clemson University Math Sciences mjs AT clemson DOT edu http://www.math.clemson.edu/~mjs -- redhat-list mailing list unsubscribe mailto:redhat-list-request@xxxxxxxxxx?subject=unsubscribe https://www.redhat.com/mailman/listinfo/redhat-list
