[odc] Daily src changes for 2004-04-12

ODC auto at squish.net
Tue Apr 13 08:00:38 BST 2004


OpenBSD src changes summary for 2004-04-12
==========================================

distrib/alpha                           distrib/miniroot
distrib/mvme68k                         lib/libc
regress/sys                             sbin/dhclient
share/man                               sys/arch/hp300/dev
sys/arch/m68k/m68k                      sys/arch/mvme88k/dev
sys/arch/mvme88k/include                sys/arch/sparc/dev
sys/dev/i2o                             sys/dev/ic
sys/dev/pci                             sys/kern
sys/netinet                             sys/nfs
sys/sys                                 usr.sbin/dhcrelay
usr.sbin/mopd                           usr.sbin/openssl

== distrib =========================================================== 01/07 ==

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

alpha

  ~ common/disktab.shadow                 

  > removeable -> removable;
  > ok deraadt@ beck@ (jmc@)

miniroot

  ~ install.sub                           

  > Refactor network initialization code into explicit v4 routines to
  > prepare for v6 initialization. Minor verbiage changes, and not asking
  > for v4 default route if no v4 addresses were configured are the only
  > user visible changes.
  > ok deraadt at .rw@)

mvme68k

  ~ ramdisk/disktab.shadow                

  > removeable -> removable;
  > ok deraadt@ beck@ (jmc@)

== lib =============================================================== 02/07 ==

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

libc

  ~ stdlib/malloc.c                       

  > Clean up malloc_active state when aborting.
  > This allows for safe abort handling, without tripping into
  > false recursivity problems.
  > Ok tedu@, deraadt@ (tdeval@)

== regress =========================================================== 03/07 ==

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

sys

  ~ kern/Makefile                         

  > don't check rfork.	the RFMEM changes had two effects.  some of the
  > regress conditions (rfmem-stack) are no longer true, and other tests
  > have to be rewritten to handle shared stacks.  spotted by david@ (tedu@)

== sbin ============================================================== 04/07 ==

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

dhclient

  ~ dhclient-script                       

  > simplify, from Han Boetes & pb@, theo ok (henning@)

  ~ dhclient-script                       

  > spacing (deraadt@)

== share ============================================================= 05/07 ==

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

man

  ~ man4/gdt.4                            

  > update models (deraadt@)

== sys =============================================================== 06/07 ==

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

arch/hp300/dev

  ~ ctreg.h                               ~ hdreg.h

  > removeable -> removable;
  > ok deraadt@ beck@ (jmc@)

arch/m68k/m68k

  ~ process_machdep.c                     

  > Fix process_frame macro.
  > ok miod@ (kettenis@)

arch/mvme88k/dev

  ~ bugio.c                               

  > Minor tweaks to appease gcc 3. (miod@)

arch/mvme88k/include

  ~ m88410.h                              

  > Minor tweaks to appease gcc 3. (miod@)

arch/sparc/dev

  ~ xdreg.h                               ~ xyreg.h

  > removeable -> removable;
  > ok deraadt@ beck@ (jmc@)

dev/i2o

  ~ i2o.h                                 ~ ioprbs.c

  > removeable -> removable;
  > ok deraadt@ beck@ (jmc@)

dev/ic

  ~ gdt_common.c                          

  > This has been a pain in the ass for about 8 months which made new ICP
  > cards lock up under load: skip two initializations that we were
  > previously told to use by ICP themselves (ie. from their linux
  > driver).  now we are not supposed to use those APIs, because the new
  > firmwares for their new cards become unrealiable.  and that is all we
  > have been able to find out.  sigh.
  > For some cards, you must also be sure to get a new boot firmware from
  > ICP because they made another mistake: Any bootblock that tried to
  > switch back and forth between real and protected mode ended up with
  > corrupted registers.
  > from Achim_Leubner at adaptec.comia niklas, tested on a few cards
  > by a few people. (deraadt@)

  ~ mpt_openbsd.c                         

  > Fix DMA length on complex SGL.  Apparantly this fixes the outstanding
  > 1.03.xx mpt firmware issue.  From NetBSD pointed out by deraadt at .k
  > deraadt@, krw@ (marco@)

  ~ aic7xxx.h                             ~ aic7xxxvar.h

  > removeable -> removable;
  > ok deraadt@ beck@ (jmc@)

dev/pci

  ~ if_bge.c                              

  > Enable the memory arbiter before turning off the PXE restart.  This
  > prevents NMI's from happening when resetting the chip on some
  > hardware.
  > From: FreeBSD
  > ok deraadt@ (brad@)

kern

  ~ uipc_syscalls.c                       

  > make sockargs take void *, combine a len check.
  > from pedro martelletto, ok markus@ (tedu@)

netinet

  ~ tcp_input.c                           

  > add and use dropafterack_ratelim; ok deraadt@ (markus@)

  ~ tcp_usrreq.c                          

  > remove duplicate if.  from pedro martelletto, ok markus@ (tedu@)

nfs

  ~ nfs_vfsops.c                          

  > make sockargs take void *, combine a len check.
  > from pedro martelletto, ok markus@ (tedu@)

sys

  ~ socketvar.h                           

  > make sockargs take void *, combine a len check.
  > from pedro martelletto, ok markus@ (tedu@)

== usr.sbin ========================================================== 07/07 ==

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

dhcrelay

  + Makefile                              + bpf.c
  + dhcp.h                                + dhcpd.h
  + dhcrelay.8                            + dhcrelay.c
  + dispatch.c                            + errwarn.c
  + packet.c                              + tree.h

  > initial cut at split out dhcrelay, whacked from usr.sbin/dhcp/ and
  > sbin/dhclient (henning@)

  ~ dhcrelay.c                            

  > less pathetic usage() (henning@)

  ~ Makefile                              

  > oups (henning@)

  ~ dhcrelay.c                            

  > kill pidfile gunk (henning@)

  ~ dhcrelay.c                            

  > the rocket science department suggests: getopt (henning@)

  ~ dhcrelay.c                            

  > spacing (deraadt@)

  ~ dhcrelay.c                            

  > se daemon(3) (henning@)

  ~ dhcrelay.c                            

  > one big sweep of KNF, ansi and such (henning@)

  ~ dhcrelay.c                            

  > KNF (henning@)

  ~ dhcrelay.c                            

  > remove those overly helpfull comments like explaining that time(3) gets the
  > current time (henning@)

  ~ dhcrelay.c                            

  > relaying back bootp answers should work again (henning@)

mopd

  ~ mopd/mopd.c                           ~ mopd/process.c
  ~ mopd/process.h                        

  > ~1000 lines worth of ansi, KNF, various cleaning, replacing a homebrown
  > daemon() equivalent and such things, theo ok (henning@)

openssl

  ~ openssl.1                             

  > sync w/ 0.9.7d import (one typo); (jmc@)

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



More information about the odc mailing list