Hmmm.  Just looking at this again, it appears that even though I have
USE_NEW_TOOLCHAIN set, and that it in fact built a toolchain copy of
lint, it didn't use the toolchain copy;  instead it used the previously
installed (1.5T) version!

Forcing it to use the toolchain version with 

	make LINT=/usr/obj/tools/obj.i386/bin/lint DESTDIR=/ build

seems to do the trick!

From looking at the rest of the build log, it appears that we do not use
the toolchain for anything, unless I'm doing something wrong.  Here's my
/etc/mk.conf file:

	OBJMACHINE=true
	MKOBJDIRS=yes
	MKTOOLS=yes
	__TOOLARCH!=uname -m
	TOOLDIR=/usr/obj/tools/obj.${__TOOLARCH}
	USE_NEW_TOOLCHAIN=yes

	HAVE_MOTIF=YES
	USA_RESIDENT=YES
	USE_RSAREF2=YES
	ACCEPTABLE_LICENSES=shareware no-commercial-use

On Thursday, 16 Aug 2001 17:16:55 -0700 (PDT), Paul Goyette wrote:

> Got the following error both with my original (vintage 1.5T) toolchain
> as well as with the USE_NEW_TOOLCHAIN stuff:
> 
> lint -chapbxzF -w -X 272 -d //usr/include -D_LIBC -DNLS -DYP -DHESIOD

  ^^^^ wrong copy of lint will get used here.

> -DLIBC_SCCS -DSYSLIBC_SCCS -D_REENTRANT -I/usr/src/lib/libc/include
> -DINET6 -D__DBINTERFACE_PRIVATE
> -I/usr/src/lib/libc/../../libexec/ld.elf_so -I/usr/src/lib/libc/dlfcn
> -DWITH_RUNE -DRUNEMOD_MAJOR=3
> -D_PATH_LOCALEMODULE=\"/usr/lib/runemodule\" -DRESOLVSORT -I.
> -DPOSIX_MISTAKE -DPORTMAP -DFLOATING_POINT -i
> /usr/src/lib/libc/net/__cmsg_alignbytes.c
> /usr/include/uvm/uvm_page.h(135): warning: nonportable bit-field type
> [34]
> *** Error code 1
> 
> Stop.
> 
> Sources updated via cvs approx 11:00 AM PDT (UTC-7).

----------------------------------------------------------------------
|   Paul Goyette   | PGP DSS Key fingerprint: |  E-mail addresses:   |
| Network Engineer | BCD7 5301 9513 58A6 0DBC |  paul@whooppee.com   |
|  & World Cruiser | 91EB ADB1 A280 3B79 9221 | pgoyette@juniper.net |
----------------------------------------------------------------------