On Wed, Jul 25, 2001 at 08:40:43AM -0500, Steve Snyder wrote: > I have learned the hard way that the "-fomit-frame-pointer" compiler switch > cannot be used when compiling C++ source files. Has anyone seen any > problems when using this switch to compile C files? > > Maybe I should give some background. With either the compiler that shipped > with Seawolf or the subsequently-released update, C++ code does not build > correctly when this switch is used. The source may not compile (e.g. > internal compiler error when building Mozilla 0.9.2) or it may compile but > not run correctly (runtime faults when running in KDE after rebuilding > kdelibs). Yes, I've submitted this to Bugzilla (#41794, submitted in May). > > Anyway, I have learned to have zero confidence in RedHat's compiler when > "-fomit-frame-pointer" is used on C++ code, and I've been too spooked by my > experiences to try it on plain old C code. Can anyone give a testimonial > that "-fomit-frame-pointer" will correctly compile C code? I use -fomit-frame-pointer all the time for C with no problems observed. There is also some C++ code in this project, also compiled with -fomit-frame-pointer. However, this C++ code is little used at present so I can't say whether or not there are any issues with it due to -fomit-frame-pointer. _______________________________________________ Seawolf-list mailing list [EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/seawolf-list
