Permalink
Commits on Nov 19, 2004
  1. removed comments about the "conflict" between a sanity check and the

    jinmei committed Nov 19, 2004
    standard algorithm of processing a routing header.
Commits on Nov 18, 2004
  1. 2004-11-18 JINMEI, Tatuya <jinmei@isl.rdc.toshiba.co.jp>

    jinmei committed Nov 18, 2004
    	* kame/sys/netinet6/in6.c (in6_if2idlen): introduced a new
    	function, which was previously a local module "if2idlen", and
    	called it when necessary.  With this change we do not store the
    	IFID length in in6_ifextra{}.  It did not always work as intended
    	due to initialization timing issues.
  2. revised the logic of ifid2len:

    jinmei committed Nov 18, 2004
      do not store the length in in6_ifextra{}, since it can sometimes be
      initialized inpropoerly due to timing issues.  instead, make if2idlen
      public and calls it when necessary.  the penalty should be acceptable,
      because (at least so far) we don't need it in a performance-conscious
      path.
      also added the case for IFT_GIF (for configured tunneling) and
      IFT_LOOP (we may skip this case though).
Commits on Nov 17, 2004
  1. removed the previous fix on Nov 12, since it can lead to a buffer ove…

    suz committed Nov 17, 2004
    …rrun.
    
    (RFC2460 section 4.4 needs an update)
  2. wrong paren. Patrick Latifi

    itojun committed Nov 17, 2004
  3. NI_WITHSCOPEID is no more

    itojun committed Nov 17, 2004
Commits on Nov 16, 2004
  1. openbsd: switch to 3.6

    itojun committed Nov 16, 2004
  2. skip ifmcstat for now

    itojun committed Nov 16, 2004
  3. sync w/ 3.6 (netstat is disabled for now)

    itojun committed Nov 16, 2004
  4. sync w/ 3.6. leave pf* for now

    itojun committed Nov 16, 2004
  5. async w/ 3.6. pfsync is commented out for now

    itojun committed Nov 16, 2004
  6. sync w/ 3.6

    itojun committed Nov 16, 2004
Commits on Nov 12, 2004
  1. coped with the difference of the return value of nd6_storeaddr() betw…

    suz committed Nov 12, 2004
    …een KAME and FreeBSD-5.3
  2. a Type-0 Routing-header is skipped if Segments Left is set to zero, r…

    suz committed Nov 12, 2004
    …egardless of the Hdr-Ext-Len value. (RFC2460 section 4.4)
  3. freebsd5.3+kame gets stable enough at least for minimam everyday use

    suz committed Nov 12, 2004
    (advanced features, such as ALTQ, MLDv2 etc, are not checked yet)
  4. removed unnecessary difference from freebsd 5.3 (to prevent kernel cr…

    suz committed Nov 12, 2004
    …ash due to an unexpected lock/unlock introduced during merge work)
  5. remove difference from freebsd53 (to fix kernel crash due to unexpect…

    suz committed Nov 12, 2004
    …ed lock/unlock failure introduced during merge work)
  6. removed compilation warning

    suz committed Nov 12, 2004
  7. catch up with freebsd53

    suz committed Nov 12, 2004
  8. update freebsd5 to freebsd53

    suz committed Nov 12, 2004
Commits on Nov 11, 2004
  1. sync with freebsd53 and makes it compilable

    suz committed Nov 11, 2004
    (their behavior is not confirmed yet...)
  2. sync with freebsd5.3

    suz committed Nov 11, 2004
  3. Initial revision

    suz committed Nov 11, 2004
  4. sync with freebsd 5.3

    suz committed Nov 11, 2004
    (just confirmed compilation, and haven't confirmed the actual behavior yet)
    
    ToDo:
    	make it work properly
    	what to do with KAME PF/ALTQ (right now, I use freebsd5.3's one)
    	check KAME-specific protocol modules
Commits on Nov 8, 2004
  1. Initial revision

    suz committed Nov 8, 2004
  2. Initial revision

    suz committed Nov 8, 2004
  3. fix compilation on !openbsd

    itojun committed Nov 8, 2004
  4. <200411>

    jinmei committed Nov 8, 2004
    2004-11-08  JINMEI, Tatuya  <jinmei@isl.rdc.toshiba.co.jp>
    	* freebsd4/sys/netinet/tcp_input.c (tcp_mssopt): corrected the
    	default MSS computation for TCP/IPv6.
  5. corrected the default MSS computation.

    jinmei committed Nov 8, 2004