[odc] Daily src changes for 2006-03-09

ODC auto at squish.net
Fri Mar 10 07:00:38 GMT 2006


OpenBSD src changes summary for 2006-03-09
==========================================

distrib/mac68k                          etc/etc.mac68k/ttys
gnu/egcs                                libexec/login_chpass
libexec/login_krb5-or-pwd               libexec/login_lchpass
libexec/login_passwd                    libexec/login_reject
sbin/dumpfs                             sbin/fsck_ffs
sbin/newfs                              sys/arch/amd64/amd64
sys/arch/macppc/macppc                  sys/arch/mips64/mips64
sys/ddb                                 sys/dev/acpi
sys/dev/isa                             sys/dev/pci
sys/ufs/ffs                             usr.bin/cvs
usr.bin/rcs                             usr.sbin/ospfctl
usr.sbin/ospfd                          

== distrib =========================================================== 01/08 ==

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

mac68k

  ~ ramdisk/install.md                    

  > vt100->vt220; spotted by nick@ (miod@)

== etc =============================================================== 02/08 ==

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

etc.mac68k/ttys

  ~ etc.mac68k/ttys                       

  > vt100->vt220; spotted by nick@ (miod@)

== gnu =============================================================== 03/08 ==

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

egcs

  ~ gcc/c-common.c                        

  > Recognize %z in addition to %Z as a printf format specifier; from espie@
  > who is too shy to commit it. (miod@)

== libexec =========================================================== 04/08 ==

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

login_chpass

  ~ login_chpass.c                        

  > Foil potential timing attacks by using the correct password hash
  > instead of "xx".  In practice this means bcrypt() will be used for
  > non-existent users instead of DES crypt().
  > Adapted from a patch by Peter Philipp.  OK deraadt@ (millert@)

login_krb5-or-pwd

  ~ Makefile                              

  > Foil potential timing attacks by using the correct password hash
  > instead of "xx".  In practice this means bcrypt() will be used for
  > non-existent users instead of DES crypt().
  > Adapted from a patch by Peter Philipp.  OK deraadt@ (millert@)

login_lchpass

  ~ login_lchpass.c                       

  > Foil potential timing attacks by using the correct password hash
  > instead of "xx".  In practice this means bcrypt() will be used for
  > non-existent users instead of DES crypt().
  > Adapted from a patch by Peter Philipp.  OK deraadt@ (millert@)

login_passwd

  ~ Makefile                              ~ common.h
  ~ login_passwd.c                        

  > Foil potential timing attacks by using the correct password hash
  > instead of "xx".  In practice this means bcrypt() will be used for
  > non-existent users instead of DES crypt().
  > Adapted from a patch by Peter Philipp.  OK deraadt@ (millert@)

login_reject

  ~ Makefile                              ~ login_reject.c

  > Foil potential timing attacks by using the correct password hash
  > instead of "xx".  In practice this means bcrypt() will be used for
  > non-existent users instead of DES crypt().
  > Adapted from a patch by Peter Philipp.  OK deraadt@ (millert@)

== sbin ============================================================== 05/08 ==

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

dumpfs

  ~ dumpfs.c                              

  > Remove option -n from newfs as well as all references to fs_postbl()
  > Various testing for a while, okay krw@ (pedro@)

fsck_ffs

  ~ setup.c                               

  > Remove option -n from newfs as well as all references to fs_postbl()
  > Various testing for a while, okay krw@ (pedro@)

newfs

  ~ newfs.8                               

  > Remove -n from synopsis too (pedro@)

  ~ mkfs.c                                ~ newfs.8
  ~ newfs.c                               

  > Remove option -n from newfs as well as all references to fs_postbl()
  > Various testing for a while, okay krw@ (pedro@)

== sys =============================================================== 06/08 ==

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

arch/amd64/amd64

  ~ machdep.c                             

  > Typo fix from 'P-O Yliniemi <peo atsomehostlike bsd-guide.net>'
  > marco@ ok. (weingart@)

arch/macppc/macppc

  ~ opendev.c                             

  > Do not include <lib/libkern/libkern.h> explicitely, get it via
  > <sys/systm.h>. (miod@)

arch/mips64/mips64

  ~ arcbios.c                             

  > Do not include <lib/libkern/libkern.h> explicitely, get it via
  > <sys/systm.h>. (miod@)

ddb

  ~ db_output.c                           

  > Do not include <lib/libkern/libkern.h> explicitely, get it via
  > <sys/systm.h>. (miod@)

dev/acpi

  ~ acpidebug.c                           

  > Create separate function to parse name + scope.  Should shave some bytes of
  > and makes code more readable.
  > ok jordan. (marco@)

  ~ acpidebug.c                           

  > Pretty. (marco@)

  ~ acpidebug.c                           

  > One more KNF nit. (marco@)

  ~ acpicpu.c                             

  > Fixed free before malloc. bad marco
  > ok marco@ (jordan@)

  ~ dsdt.c                                

  > Removed unused prototypes
  > Fixed aml_notify to use known node
  > Added new method/node evaluator
  > Renamed some functions for easier naming
  > ok marco@ (jordan@)

  ~ dsdt.c                                

  > Remove bad XXX
  > ok jordan (marco@)

  ~ acpidebug.c                           ~ dsdt.h
  ~ dsdt.c                                

  > Used common aml_opname for debugger
  > Removed redundant code
  > ok marco@ (jordan@)

dev/isa

  ~ spkr.c                                

  > KNF, deregister, no functional change. (miod@)

  ~ spkr.c                                

  > Unimplemented ioctl commands should return ENOTTY, not EINVAL. (miod@)

  ~ mcd.c                                 

  > Zero-fill the fake softc during probe, and put a plausible dv_xname, so
  > that
  > probe failures don't display garbage in their messages. (miod@)

  ~ mcd.c                                 

  > Do not include <lib/libkern/libkern.h> explicitely, get it via
  > <sys/systm.h>. (miod@)

dev/pci

  ~ amdpm.c                               

  > unbreak the tree. i dunno how this compiled on my box at home.
  > found by marco@ (dlg@)

ufs/ffs

  ~ ffs_vfsops.c                          

  > Better handling of ancient file systems that still have 4.2 BSD
  > rotational layout tables. Okay krw@ (pedro@)

  ~ ffs_vfsops.c                          

  > - Slightly more informative error message (requested by krw@)
  > - Bring back '&& !ronly' into the conditional, so these ancient file
  > systems can be mounted again read-only (requested by millert@) (pedro@)

  ~ fs.h                                  

  > Remove option -n from newfs as well as all references to fs_postbl()
  > Various testing for a while, okay krw@ (pedro@)

== usr.bin =========================================================== 07/08 ==

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

cvs

  ~ Makefile                              ~ rcs.c
  ~ rcs.h                                 + rcstime.c

  > move joris' timezone support into rcstime.c; more work to be done there.
  > OK joris at .sa@)

  ~ rcstime.c                             

  > trailing spaces; (joris@)

rcs

  ~ Makefile                              

  > should have been part of previous commit (add rcstime.c here too). (xsa@)

== usr.sbin ========================================================== 08/08 ==

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

ospfctl

  ~ ospfctl.c                             

  > Start using *_name function that are now shared between ospfd and ospfctl.
  > OK norby@ (claudio@)

  ~ ospfctl.c                             

  > Track interface uptime.
  > ok claudio@ (norby@)

ospfd

  ~ ospfe.c                               

  > Log link state changes.
  > ok claudio@ (norby@)

  ~ interface.c                           ~ ospfd.h

  > if_event_names is only used by interface.c so move it there. (claudio@)

  ~ in_cksum.c                            ~ ospfd.h

  > in_cksum() should return a u_int16_t and not int. Switch a variable to
  > unsigned. Found by lint. (claudio@)

  ~ ospfe.h                               ~ packet.c

  > Switch send_packet from char * to void * and from int to size_t. The
  > function
  > is now similar to sendto et al. (claudio@)

  ~ lsreq.c                               ~ database.c

  > Mark event callbacks with ARGSUSED to silence lint. (claudio@)

  ~ interface.c                           

  > KNF, remove some unneded type casts and add some ARGSUSED. This makes lint
  > almost happy. (claudio@)

  ~ hello.c                               

  > Correctly warn about unsupported interface types as it is done in other
  > places. (claudio@)

  ~ interface.c                           ~ log.c
  ~ neighbor.c                            ~ ospfd.h
  ~ ospfe.h                               

  > Move *_name functions to log.c and use these for both ospfd and ospfctl.
  > Stuff that is only used in ospfd debug messages is simplified.
  > OK norby@ (claudio@)

  ~ ospf.h                                

  > It seems that 60 secs is a reasonable amount of time to reach state FULL.
  > (norby@)

  ~ rde.c                                 

  > More spring cleaning with lint. Change a few types and add some comments to
  > help lint. (claudio@)

  ~ log.c                                 

  > Add some NOTREACHED to please lint. (claudio@)

  ~ interface.c                           ~ ospfd.h

  > Track interface uptime.
  > ok claudio@ (norby@)

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



More information about the odc mailing list