In article <20060710163535.E36DE1A48E0@mail.eecs.harvard.edu>,
David Holland <dholland+netbsd@eecs.harvard.edu> wrote:
>A whole bunch of files are failing to compile with warnings like this:
>
>../../../../kern/tty.c: In function 'ttyoutput':
>../../../../kern/tty.c:740: warning: pointer targets in passing argument
>1 of 'b_to_q' differ in signedness
>
>This is today (20060710)'s tree and a system built last Friday
>(20060707).
>
>I've filed PRs with patches for dev/ic/aic7xxx_osm.c and
>dev/scsipi/cd.c, but I think I'm going to stop for now (as I have
>other things I need to get done today...)
>
>These are the files that are failing on me (but I'm not building
>GENERIC so it's probably not a complete list):
>
>   fs/cd9660/cd9660_lookup.c
>   fs/cd9660/cd9660_node.c
>   fs/cd9660/cd9660_rrip.c
>   fs/cd9660/cd9660_vfsops.c
>   fs/cd9660/cd9660_vnops.c
>   kern/core_elf32.c
>   opencrypto/cryptodev.c
>   opencrypto/cryptosoft.c
>   dev/dkwedge/dk.c
>   ufs/ffs/ffs_vfsops.c
>   dev/usb/hid.c
>   netinet/if_arp.c
>   ufs/mfs/mfs_vfsops.c
>   dev/scsipi/scsiconf.c
>   dev/scsipi/scsipi_verbose.c
>   dev/scsipi/sd.c
>   kern/tty.c
>   dev/usb/uhidev.c
>   arch/i386/pci/viapcib.c
>   arch/i386/i386/vm86.c
>   dev/vnd.c
>   dev/wscons/wsemul_vt100_subr.c
>
>gcc -v gives
>
>   Using built-in specs.
>   Target: i386--netbsdelf
>   Configured with: /usr/src/tools/gcc/../../gnu/dist/gcc4/configure
>--enable-long-long --disable-multilib --enable-threads --disable-symvers
>--build=i386-unknown-netbsdelf3.99.17 --host=i386--netbsdelf
>--target=i386--netbsdelf
>   Thread model: posix
>   gcc version 4.1.2 20060628 prerelease (NetBSD nb1 20060628)
>
>and there's nothing pertinent in /etc/mk.conf.
>
>It seems surprising that nobody else would be running into this, but
>as far as I can tell I'm not doing anything unusual or obviously
>stupid, and the machine was freshly installed on Friday...

Well, you are not doing something stupid... You are just running with
stale .mk files in /usr/share/mk... Eventually we should fix all of
these, but for now we have just disabled the errors... So keep sending
PR's with fixes :-)

christos