This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: ANSIfy cp/parser.c
Current discussion focusses on using C89 syntax without
conditionalization. I'd like to redirect things a bit. An entirely
serious question: Are we prepared to drop support for hosts and hosted
targets which do not provide a C89 compliant standard library? This
would, for instance, permit the total removal of fix-header (not to be
confused with fixincludes); considerable trimming down of system.h and
configure.in; the removal of quite a bit of libiberty; and so on.
I don't personally see that much of an advantage to C89 syntax, but
being able to make more assumptions about the library baseline would
be a big improvement IMO.
It's my belief that most of these hosts and targets were nailed by the
deprecation list I posted for 3.3 a few weeks back, but I am not 100%
sure. And I got pushback on a few of them, notably SunOS 4.
zw