Date: Mon, 28 May 2001 21:43:58 +1000 (EST) From: Bruce Evans <bde@zeta.org.au> To: Sheldon Hearn <sheldonh@uunet.co.za> Cc: Dima Dorfman <dd@FreeBSD.org>, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/usr.bin/jot Makefile jot.c Message-ID: <Pine.BSF.4.21.0105282121480.1612-100000@besplex.bde.org> In-Reply-To: <50418.991042601@axl.fw.uunet.co.za>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 28 May 2001, Sheldon Hearn wrote: > On Sun, 27 May 2001 17:55:52 MST, Dima Dorfman wrote: > > > Modified files: > > usr.bin/jot Makefile jot.c > > Log: > > Silence warnings and compile with WARNS=2 on i386 and alpha. > > Since when do you have to add a prototype for main()?! Since -Wmissing-prototypes was added to CFLAGS (by WARNS=2 or BDECFLAGS) to detect the potential error of not declaring extern functions in the right place (which is never in *.c). main() is a special case, however. It can't be declared in a header file in C because in the hosted (non-freestanding) case there are several valid but inconsistent prototypes for it: int main(void); /* required to work by ISO C90 */ int main(int argc, char **argv); /* same */ int main(int argc, char **argv, char **envp); /* optional POSIX */ int main(anything); /* optional implementation-defined */ gcc shouldn't warn about main() not being prototyped before it is defined even with -Wmissing-prototypes, but it currently warns about it if the definition is old-style. gcc -ffreestanding also seems to be broken. I think main() should not be special then, but it is. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0105282121480.1612-100000>