Permalink
Commits on May 14, 2005
Commits on May 13, 2005
Commits on May 12, 2005
  1. sync with 5.4-RELEASE (it includes a fix for the FreeBSD Security Adv…

    …isory FreeBSD-SA-05:06-08)
    suz committed May 12, 2005
  2. sync with freebsd54

    suz committed May 12, 2005
  3. commented out res_init.c on freebsd5, because some function in this f…

    …ile conflicts with the same on in freebsd54's libc
    suz committed May 12, 2005
  4. fixed a compilation failure on openbsd

    suz committed May 12, 2005
  5. fixed a compilation failure

    suz committed May 12, 2005
  6. sync with freebsd 5.4

    suz committed May 12, 2005
  7. Initial revision

    suz committed May 12, 2005
  8. sync with freebsd 5.4

    suz committed May 12, 2005
  9. sync with freebsd-5.4

    suz committed May 12, 2005
  10. Initial revision

    suz committed May 12, 2005
Commits on May 9, 2005
  1. make sure that prefix go away when interface vanishes. from netbsd

    XXX is there any possibility for dangling link (i.e. refcnt > 1) to "pr"?
    itojun committed May 9, 2005
Commits on May 7, 2005
  1. renamd a sysctl "net.inet.igmp.igmp_version" to "net.inet.igmp.versio…

    …n" for brevity
    suz committed May 7, 2005
Commits on May 4, 2005
  1. typo in comments

    jinmei committed May 4, 2005
  2. make it compile

    itojun committed May 4, 2005
Commits on May 3, 2005
  1. 2005-05-03 JINMEI, Tatuya <jinmei@isl.rdc.toshiba.co.jp>

    	* kame/kame/dhcp6/dhcp6c_ia.c (ia_timo): kept the latest server's
    	DUID even in the REBIND state in case of sending a Release message
    	in that state.
    jinmei committed May 3, 2005
  2. keep latest server's DUID even in the REBIND state in case of sending

    a Release message in this state.
    jinmei committed May 3, 2005
  3. <200505>

    2005-05-03  JINMEI, Tatuya  <jinmei@isl.rdc.toshiba.co.jp>
    	* kame/kame/dhcp6/cfparse.y: corrected "address parameters" so
    	that we don't have to specify the meaningless (but mandatory)
    	prefix length in the context of IA_NA.
    
    2005-05-03  JINMEI, Tatuya  <jinmei@isl.rdc.toshiba.co.jp>
    	* kame/kame/dhcp6/config.c (configure_commit): made sure that
    	interface parameters are initialized with the default values even
    	if the interface is not explicitly configured in the configuration
    	file.  This particularly made sure that the preference option is
    	not included unless explicitly specified.  A memory leakage for
    	the client when specifying a script file was also fixed.
    jinmei committed May 3, 2005
  4. corrected "address parameters" so that we don't have to specify the

    meaningless (but mandatory) prefix length in the context of IA_NA.
    (this also makes the implementation synchronized with the man pages)
    jinmei committed May 3, 2005
  5. (style) combined two short lines

    jinmei committed May 3, 2005
  6. corrected the syntax of a remove argument

    jinmei committed May 3, 2005
  7. made sure initialize interface parameters even if the interface is not

    explicitly configured in the configuration file.
    in particular, made sure that preference option is not included unless
    explicitly specified.
    it also fixed a minor memory leakage.
    jinmei committed May 3, 2005
Commits on May 2, 2005
  1. fixed a compilation error

    suz committed May 2, 2005
  2. removed a freebsd4-specific include file

    suz committed May 2, 2005
  3. it's no longer built in KAME SNAP

    suz committed May 2, 2005
  4. prefix command is obsolete

    suz committed May 2, 2005
  5. removed no-used files

    suz committed May 2, 2005
  6. sync with freebsd to reduce the diff

    suz committed May 2, 2005
Commits on May 1, 2005
  1. cnd starts up prior to other daemons

    t-momose committed May 1, 2005
Commits on Apr 28, 2005
  1. 2005-04-29 JINMEI, Tatuya <jinmei@isl.rdc.toshiba.co.jp>

    	* kame/kame/dhcp6/dhcp6c.c (client6_mainloop): corrected error
    	handling for select(2) in order to avoid unexpected blocking when
    	receiving a signal.
    jinmei committed Apr 28, 2005
  2. client6_mainloop(): made sure to continue the loop when select(2) failed

    (likely due to a signal).  Otherwise, the process would misunderstand that
    it received a packet and would block in client6_recv().
    jinmei committed Apr 28, 2005