[odc] Daily src changes for 2006-03-21

[ODC]

OpenBSD src changes summary for 2006-03-21
==========================================

bin/ksh                                 bin/rm
etc/security                            lib/libusbhid
regress/sbin                            regress/usr.bin
sbin/ifconfig                           sbin/pfctl
share/man                               sys/conf
sys/crypto                              sys/dev/acpi
sys/dev/usb                             sys/sys
usr.bin/rcs                             usr.sbin/ospfd
usr.sbin/pkg_add                        

== bin =============================================================== 01/09 ==

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

ksh

  ~ c_sh.c                                

  > Partially revert last commit; jaredy@ showed the NULL deref can't
  > happen and tedu@ confirmed this is a known bug in Coverity's analysis.
  > (otto@)

  ~ syn.c                                 

  > Revert for the moment, according to Mike Belopuhov this produces
  > errors in rc and netstart. (otto@)

rm

  ~ rm.c                                  

  > Call fts_close() when done; from NetBSD's Coverity scan; ok millert@
  > jaredy@ (otto@)

== etc =============================================================== 02/09 ==

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

security

  ~ security                              

  > Fix for PR 5043: shell startup scripts might contain binary characters but
  > grep should assume ASCII text, fixes umask detection
  > ok millert@ jaredy@ (david@)

== lib =============================================================== 03/09 ==

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

libusbhid

  ~ usage.c                               

  > - plug memleak when handling reallocation failure in hid_start()
  > - avoid NULL dereference while reclaiming memory during error
  > handling in hid_start()
  > ok ray dhill (jaredy@)

== regress =========================================================== 04/09 ==

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

sbin

  ~ ipsecctl/ike1.ok                      ~ ipsecctl/ike2.ok
  ~ ipsecctl/ike3.ok                      ~ ipsecctl/ike4.ok
  ~ ipsecctl/ike5.ok                      ~ ipsecctl/ike6.ok
  ~ ipsecctl/ike8.ok                      

  > adopt to recent changes.  noticed by david@ (hshoexer@)

usr.bin

  ~ rcs/Makefile                          + rcs/rcs-aflag.out
  + rcs/rcs-eflag.out                     

  > add tests for `rcs -a' and `rcs -e', just not enabled yet.
  > "Looks good" ray at .sa@)

== sbin ============================================================== 05/09 ==

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

ifconfig

  ~ ifconfig.c                            

  > Correctly check for the end of the cmds table. There is no need to check
  > for
  > c_func2. Until now ifconfig accepted something like ifconfig tun0 1.2.3.4
  > 1.2.3.5 foobar without error. Additionally change the error message to
  > a more comprehensible message. OK markus@, henning@ (claudio@)

pfctl

  ~ pfctl_parser.c                        

  > instead of sizeof(array) / sizeof(element) computation, use the existing
  > end-of-array NULL marker, shuts up source analysis tool, from deraadt@
  > (dhartmei@)

== share ============================================================= 06/09 ==

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

man

  ~ man9/style.9                          

  > Encourage using strtonum(3).
  > Sync variable declarations with variables used in getopt(3) loop.
  > OK otto@ (ray@)

== sys =============================================================== 07/09 ==

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

conf

  ~ files                                 

  > Implementation of the Michael MIC as defined in IEEE 802.11i for TKIP.
  > The MIC generates a weak 64bit digest protected by an additional key.
  > Obviously, this digest alg is required for future IEEE 802.11i/WPA
  > support.
  > test vectors passed on alpha amd64 mvme68k mvme88k sgi sparc sparc64 vax
  > i386
  > ok djm@ (reyk@)

crypto

  + michael.c                             + michael.h

  > Implementation of the Michael MIC as defined in IEEE 802.11i for TKIP.
  > The MIC generates a weak 64bit digest protected by an additional key.
  > Obviously, this digest alg is required for future IEEE 802.11i/WPA
  > support.
  > test vectors passed on alpha amd64 mvme68k mvme88k sgi sparc sparc64 vax
  > i386
  > ok djm@ (reyk@)

dev/acpi

  ~ dsdt.c                                ~ acpidebug.c

  > Disassembler now pretty prints AML code (like acpidump)
  > ok marco@ (jordan@)

dev/usb

  ~ if_ral.c                              ~ if_ralreg.h
  ~ if_ralvar.h                           

  > Make ural(4) read and show real MAC/BBP rev instead of hardcoded RT2570
  > "Go ahead" damien@ via jsg@ (pedro@)

sys

  ~ select.h                              

  > x << long generates a lint warning, so silence it by casting __NFDBITS
  > to unsigned; this is a conseqeunce of changing size_t to u_long.
  > ok deraadt@ millert@ cloder@ (otto@)

== usr.bin =========================================================== 08/09 ==

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

rcs

  ~ rcsprog.c                             

  > Allow creation of an RCS file if RCS_CREATE flag set.
  > OK xsa@ (ray@)

  ~ rcsprog.c                             

  > Instead of duplicating rcs_suffixes and running strsep on it again,
  > just use the previous suffixes.  Since all slashes were replaced
  > with NULs, reading `suffixes' will give the first extension.
  > OK xsa@ (ray@)

  ~ rcsprog.c                             

  > add support for `rcs -e[logins]'; "Looks good" ray at .sa@)

== usr.sbin ========================================================== 09/09 ==

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

ospfd

  ~ rde_spf.c                             

  > Invalidating the full RIB and then only recalculating part of it is not
  > that smart. Kill the global rt_invalidate() instead clear only that part of
  > the RIB that is actually recalculated. OK norby@ (claudio@)

  ~ control.c                             

  > Kill debug message that floods the log if ospfctl is killed while dumping
  > large tables. (claudio@)

pkg_add

  ~ OpenBSD/Ustar.pm                      

  > fix endless loop in case of error in archive header. (espie@)

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