[odc] Daily src changes for 2008-05-16

[ODC]

OpenBSD src changes summary for 2008-05-16
==========================================

distrib/notes                           distrib/sets
distrib/socppc                          lib/libc
sbin/pfctl                              sys/arch/i386/conf
sys/arch/socppc/include                 sys/arch/socppc/stand
sys/arch/socppc/stand/boot              sys/arch/socppc/stand/mkboot
sys/dev/acpi                            sys/dev/pci
sys/dev/usb                             sys/kern
sys/sys                                 usr.bin/ssh
usr.sbin/ntpd                           usr.sbin/pkg_add
usr.sbin/relayd                         

== distrib =========================================================== 01/06 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/distrib

notes

  ~ socppc/contents                       ~ socppc/prep

  > Fix installation instructions. (kettenis@)

sets

  ~ lists/base/md.socppc                  

  > sync (kettenis@)

  ~ lists/man/mi                          

  > sync (kettenis@)

socppc

  ~ ramdisk/Makefile                      

  > Create bsd.bin, which can be loaded directly using the U-Boot firmware.
  > (kettenis@)

== lib =============================================================== 02/06 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/lib

libc

  ~ stdio/vfprintf.c                      

  > C99 says that for each va_copy() there must be a matching va_end().
  > Replace the non-portable hackery in __find_arguments() with a union.
  > From FreeBSD. (millert@)

  ~ stdio/vfprintf.c                      

  > Fix ssize_t vs. size_t mismatch (millert@)

== sbin ============================================================== 03/06 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/sbin

pfctl

  ~ pfctl.8                               

  > There is not really a network address translation device.
  > Remove it from the name section.
  > OK mcbride, henning (mpf@)

  ~ pfctl.8                               

  > no need to quote the argument to .Nd, now that it's nice and short; (jmc@)

== sys =============================================================== 04/06 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/sys

arch/i386/conf

  ~ RAMDISK                               ~ RAMDISKB
  ~ RAMDISKC                              

  > reenable PCIVERBOSE now that acpi isnt pushing the media over. (dlg@)

arch/socppc/include

  + kcore.h                               

  > Another missing header file; pointed out by checkflist (and miod@).
  > (kettenis@)

arch/socppc/stand

  ~ Makefile                              ~ Makefile.inc

  > There's no point in having an ELF bootloader; turn it into a U-Boot image.
  > (kettenis@)

arch/socppc/stand/boot

  ~ Makefile                              

  > There's no point in having an ELF bootloader; turn it into a U-Boot image.
  > (kettenis@)

arch/socppc/stand/mkboot

  ~ mkboot.8                              

  > some macro fixage; (jmc@)

dev/acpi

  ~ acpi.c                                ~ acpiac.c
  ~ acpiasus.c                            ~ acpibat.c
  ~ acpibtn.c                             ~ acpidebug.c
  ~ acpidock.c                            ~ acpiec.c
  ~ acpiprt.c                             ~ acpithinkpad.c
  ~ acpitz.c                              ~ amltypes.h
  ~ dsdt.c                                ~ dsdt.h

  > back out jordans new parser.
  > it didnt have enough oks from the right people, it did break machines
  > that were previously working, and it bloated the RAMDISKS and install
  > media beyond the constraints of the media.
  > requested by deraadt@ (dlg@)

dev/pci

  ~ pcidevs                               

  > add Intel GME965 device IDs (reyk@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > sync (reyk@)

  ~ agp_i810.c                            

  > add Intel GME965
  > ok kettenis@ (reyk@)

dev/usb

  ~ uvideo.c                              

  > Fix kernel crash when device gets detached. (mglocker@)

  ~ uvideo.c                              

  > Free sample buffer on detach.  Noted by miod@ (mglocker@)

kern

  ~ subr_pool.c                           

  > unsigned -> u_int and warnmess -> warnmsg
  > for pool_sethardlimit.
  > prodded by and ok tedu@ (thib@)

  ~ vfs_init.c                            

  > remove some old debug cruft, make some comments nicer
  > and shuffle declerations around.
  > OK blambert@ (thib@)

  ~ vfs_init.c                            

  > merge vfs_opv_init into vfs_op_init and remove the former,
  > as they where called consecutively in vfs_init. (thib@)

sys

  ~ pool.h                                

  > unsigned -> u_int and warnmess -> warnmsg
  > for pool_sethardlimit.
  > prodded by and ok tedu@ (thib@)

  ~ systm.h                               

  > merge vfs_opv_init into vfs_op_init and remove the former,
  > as they where called consecutively in vfs_init. (thib@)

== usr.bin =========================================================== 05/06 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin

ssh

  + PROTOCOL                              

  > document our protocol extensions and deviations; ok markus@ (djm@)

== usr.sbin ========================================================== 06/06 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin

ntpd

  ~ ntpd.8                                ~ ntpd.c
  ~ ntpd.h                                

  > Allow ntpd to step the time at startup, now matter how large or small the
  > offset. The 180s limit doesn't help very much these days.
  > ok otto@ mbalmer@ (ckuethe@)

pkg_add

  ~ pkg_mklocatedb                        ~ pkg_mklocatedb.1

  > add a -d repository option in order not to have to specify all pkgnames.
  > as noticed by naddy and ajacoutot, we now have enough packages that this
  > stresses shell limits if we specify everything on the command line. ;-)
  > (espie@)

  ~ OpenBSD/PackageRepository.pm          

  > tweak ftp call, remove extraneous -o option.
  > okay sthen@ (espie@)

  ~ pkg_mklocatedb                        ~ pkg_mklocatedb.1

  > - sort usage()
  > - use .Bk/.Ek to avoid ugly SYNOPSIS split (jmc@)

relayd

  ~ pfe_filter.c                          

  > Clear source nodes on table changes in sticky mode. This has the
  > disadvantage of removing valid src nodes, but the advantage of not sending
  > out traffic to hosts that are not up anymore.
  > From a diff by <mkoc at prime.pl>, ok reyk@ (pyr@)

===============================================================================