This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] c-cppbuiltin.c bits for libstdc++/31518
Joseph S. Myers wrote:
We've previously concluded that allowing diagnostic options to affect code
generation through preprocessor macros or otherwise is a bad idea (see PR
2677 and <http://gcc.gnu.org/ml/gcc/2001-05/msg00204.html>, for example),
and -fmessage-length is essentially a diagnostic (formatting) option.Frankly, I'm not sure I would consider *code generation* the application
in question: the idea is just outputting *diagnostic* in the debug-mode
library consistently, as regards formatting, with the compiler. Do you
have any comment about that?