`fatal' mismatched in current tree

[Zack Weinberg]

On Sun, 18 Apr 1999 11:37:26 -0500, Robert Lipe wrote:
>I've had variations of this for a few days now.   I suspect it's fallout
>from the recent changes in internal error handling.
>
>stage2/xgcc -Bstage2/ -B/usr/local/i686-pc-sco3.2v5.0.5/bin/ -c  -DIN_GCC     
>-W -Wall -O2 -g   -DHAVE_CONFIG_H    -I. -I../../egcs/gcc -I../../egcs/gcc/con
>fig -I../../egcs/gcc/../include ../../egcs/gcc/gen-protos.c
>../../egcs/gcc/gen-protos.c: In function `fatal':
>../../egcs/gcc/gen-protos.c:199: argument `s' doesn't match prototype
>../../egcs/gcc/system.h:387: prototype declaration
>
>In gen-protos.c, it's
>	void
>	fatal (s)
>	     char *s;
>	{
>
>In system.h, it's
>	extern void fatal PVPROTO((const char *, ...)) ATTRIBUTE_PRINTF_1 ATTRI
>BUTE_NORETURN;
>
>I'm unsure which is on the way in and which is on the way out, so no patch...

gen-protos should be changed to use const char *.  Is there any other
place where this happens for you?

zw