On Wed, 10 May 2023 at 13:23, Eli Zaretskii <e...@gnu.org> wrote: > > > From: Jonathan Wakely <jwakely....@gmail.com> > > Date: Wed, 10 May 2023 12:56:48 +0100 > > Cc: Arsen Arsenović <ar...@aarsen.me>, dje....@gmail.com, > > ja...@redhat.com, gcc@gcc.gnu.org > > > > On Wed, 10 May 2023 at 12:51, Eli Zaretskii wrote: > > > Once again, it is not GCC's business to clean up the packages which > > > use GCC as the compiler. GCC is a tool, and should allow any > > > legitimate use of it that could be useful to someone. Warning about > > > dubious usage is perfectly fine, as it helps those who do that > > > unintentionally or due to ignorance. But completely failing an > > > operation that could have produce valid code is too radical. > > > > Again (are you even reading the replies?) > > Please assume that I read everything, subject to email delivery times. > There's no reason for you to assume anything but good faith from my > side. > > > GCC will not force anybody to change code, at most it this change > > would force them to consciously and intentionally say "I know this is > > not valid C code but I want to compile it anyway". By using a compiler > > option. This is not draconian, and you sound quite silly. > > If we are not forcing code change, why bother with making it an error > at all? The only reason for doing so that was provided was that this > _is_ a way of forcing people to change their programs.
It stops people writing these errors in the first place. Not all code compiled with GCC is old, crufty K&R code that must be preserved for eternity. People are still using C to write new programs, and they are still making avoidable mistakes. The default for new code using new -std modes should be safer and less error prone.
