C++: tighten up reinterpret_cast<>

Jason Merrill jason@cygnus.com
Tue Jan 19 17:59:00 GMT 1999


>>>>> Alexandre Oliva <oliva@dcc.unicamp.br> writes:

 > On Jan 18, 1999, Jason Merrill <jason@cygnus.com> wrote:
 >> Applied, except for the pedwarn->error change; that's no longer useful,
 >> since g++ now defaults to pedwarns being errors.

 > But what if the users compiles with -fpermissive, shouldn't this be an 
 > error anyway?

No, that's the point of -fpermissive.

Jason



More information about the Gcc-patches mailing list