Skip to content
Commits on Jun 17, 2004
  1. need $(LIBS) for eaytest

    itojun committed
Commits on Jun 16, 2004
  1. no EAYDEBUG needs for plog() in crypto_openssl.c

    sakane committed
  2. same change as the previous one for other OSes.

    jinmei committed
  3. use sizeof(q) to initialize the variable q

    jinmei committed
    (mostly a cosmetic change)
  4. insufficient paren in macro def. Patrick Latifi

    itojun committed
  5. 2004-06-16 JINMEI, Tatuya <jinmei@isl.rdc.toshiba.co.jp>

    jinmei committed
    	* kame/sys/netinet6/in6.c (in6_update_ifa): do not attempt to join
    	multicast groups or to perform DAD when an existing address is
    	modified.  This change will avoid unnecessary memory
    	consumption in this case.
  6. terminated in6_update_ifa if !hostIsNew before joining multicast groups

    jinmei committed
    and/or performing DAD.  this change will fix the problem that the previous
    change attempted to solve without a bad side-effect.
Commits on Jun 15, 2004
  1. cleaned up whitespaces

    suz committed
  2. back out the previous change; duplication check was actually done

    jinmei committed
    in if_addmulti().
  3. more wording

    jinmei committed
  4. wording

    jinmei committed
  5. fixed MLDv2 Query Interval calculation

    suz committed
  6. use a dedicated group-search function

    suz committed
  7. fixed compilation error

    suz committed
  8. - use a dedicated function for group-search, instead of embedding it …

    suz committed
    …within a routine
    
    - refines code (lowered indentation level)
  9. tell people we are removing #if

    itojun committed
  10. remove freebsd[23] support

    itojun committed
  11. byebye bsdi

    itojun committed
  12. clean some #ifdef

    itojun committed
  13. cleanup some #if

    itojun committed
  14. cleanup unneeded #ifdef

    itojun committed
Commits on Jun 14, 2004
  1. check before joining multicast group.

    itojun committed
  2. sync with mld6.c Rev 1.101 (0 is not NULL)

    suz committed
  3. cleanup some #if

    itojun committed
  4. MLDv1-compat-mode related fixes

    suz committed
  5. corrected a failure return code.

    jinmei committed
  6. Fixed my misunderstanding on MLDv1-compat-mode:

    suz committed
    RFC3810 just says MLDv1-compat-mode is used to perceive an MLDv1
    report/done as an equivalent MLDv2 message, and does not say anything
    regarding MLDv2 message handling (except for some few exceptions).
    
    Current pim6sd ignores any MLDv2 source list in MLDv1-compat-mode.
    Actually this ignorance itself is not so problematic in terms of multicast
    routing.  But it's better to keep maintaining source-list even in
    MLDv1-compat-mode, considering a routing convergence time when shifting
    from MLDv1-compat-mode to MLDv2-mode.
    
    So in this patch, pim6sd receives every MLDv2 messages as it is even in
    MLDv1-compat-mode, except for some exceptions stated in RFC3810.
  7. cleanup unneeded #if

    itojun committed
  8. use u_int8_t instead of caddr_t

    itojun committed
Something went wrong with that request. Please try again.