[comp.sys.att] strange compiler bug

wsinpdb@lso.win.tue.nl (Paul de Bra) (04/10/90)

When compiling metafont on a 386, using the AT&T (development system 4.1.6)
c-compiler, I get bogus errors like this:

"mf2.c", line 1811: illegal character: 043 (octal)
"mf2.c", line 1811: cannot recover from earlier errors: goodbye!

"mf4.c", line 1362: illegal character: 043 (octal)
"mf4.c", line 1362: cannot recover from earlier errors: goodbye!

The commands for compiling each file look like:
cc -O -I.. -c mf2.c

I compiled the stuff with gcc without getting any errors.
Also, when I say

cc -P -I.. mf2.c; cc -O -I.. -c mf2.i

I get no errors, so just by splitting the compilation in 2 supposedly
equivalent parts the errors go away.

Has anyone seen this before?
(Please don't tell me I should use gcc instead. I want to try the X11 stuff
and the gnu utils, using gas and gnu-ld, don't like the shared libraries.)

Paul.
(debra@research.att.com)

scjones@sdrc.UUCP (Larry Jones) (04/12/90)

In article <1064@tuewsd.lso.win.tue.nl>, wsinpdb@lso.win.tue.nl (Paul de Bra) writes:
> When compiling metafont on a 386, using the AT&T (development system 4.1.6)
> c-compiler, I get bogus errors like this:
> 
> "mf2.c", line 1811: illegal character: 043 (octal)
> "mf2.c", line 1811: cannot recover from earlier errors: goodbye!
> 
> I compiled the stuff with gcc without getting any errors.
> Also, when I say
> 
> cc -P -I.. mf2.c; cc -O -I.. -c mf2.i
> 
> I get no errors, so just by splitting the compilation in 2 supposedly
> equivalent parts the errors go away.

Well, 043 is '#', so it would seem that you have some
preprocessor directives that aren't getting interpreted by the
preprocessor but are instead making their way into the compiler
which is choking on them.  The place that I've seen this most
often is when the '#' isn't in column 1.  Since the problem goes
away when you run the file through the preprocessor twice (the
net result of your double cc), I would guess that there is a
comment before the '#' which gets removed by the first cpp so
that the '#' is in column 1 for the second pass.
----
Larry Jones                         UUCP: uunet!sdrc!scjones
SDRC                                      scjones@SDRC.UU.NET
2000 Eastman Dr.                    BIX:  ltl
Milford, OH  45150-2789             AT&T: (513) 576-2070
"You know how Einstein got bad grades as a kid?  Well MINE are even WORSE!"
-Calvin

gwyn@smoke.BRL.MIL (Doug Gwyn) (04/13/90)

In article <1064@tuewsd.lso.win.tue.nl> wsinpdb@lso.win.tue.nl (Paul de Bra) writes:
>When compiling metafont on a 386, using the AT&T (development system 4.1.6)
>c-compiler, I get bogus errors like this:
>"mf2.c", line 1811: illegal character: 043 (octal)
>"mf2.c", line 1811: cannot recover from earlier errors: goodbye!

Since 043 is a '#' character, I suspect the source code is attempting
to use the ANSI-C stringizing or pasting features, which generally
aren't supported by older compilers.  Look on line 1811 of mf2.c and
see how it's trying to use the '#' character..