Sometime in Apr 2006, Alan Barrett wrote:
> On Sun, 09 Apr 2006, Iain Hibbert wrote:
> > > Perhaps your copy of arlex.c is older than your copy of arlex.l, in
> > > which case deleting both and doing another cvs update might help.
> >
> > Hm, I wonder if this is the cause of my troubles.. should "cvs update"
> > preserve the server timestamps? It doesnt seem to be doing so (cvs
> > checkout does) and since arlex.c arrives first it has the older
> > timestamp..
>
> No, cvs update doesn't preserve timestamps.
>
> Please try the appended patch, which should stop the reachover make
> from trying to run yacc or lex.

btw, I just noticed that I still have this patch applied on my local
sources and that without it, I still get the same failure on
binutils/ar/arlex.c when building from a readonly source tree. Should it
be applied?

> --apb (Alan Barrett)
>
> Index: gnu/usr.bin/binutils/common/Makefile.prog
> ===================================================================
> --- gnu/usr.bin/binutils/common/Makefile.prog	27 Nov 2003 15:02:21 -0000	1.7
> +++ gnu/usr.bin/binutils/common/Makefile.prog	9 Apr 2006 10:29:49 -0000
> @@ -49,8 +49,9 @@
>  DPADD+=		${IBERTYOBJ}/libiberty.a
>  .endif
>
> -.y.c .l.c:
> -
>  .PATH: ${DIST}/binutils ${DIST}/binutils/doc
>
>  .include <bsd.prog.mk>
> +
> +# Override the .y.c and .y.l rules *after* <bsd.prog.mk>.
> +.y.c .l.c:


iain