1. 06 Mar, 2014 1 commit
    • Dan Winship's avatar
      platform, devices: add support for vxlan devices · 42df06e5
      Dan Winship authored
      Since vxlan is new-ish, and vxlan IPv6 support in particular has only
      been in the kernel since 3.11, we include our own copy of the vxlan
      netlink constants rather than depending on the installed headers.
  2. 05 Mar, 2014 2 commits
  3. 04 Mar, 2014 1 commit
  4. 27 Feb, 2014 1 commit
    • Dan Winship's avatar
      platform: change sysctl_get/set error logging · 5fe94852
      Dan Winship authored
      Remove the "silent_on_error" flag from nm_platform_sysctl_get(), and
      make both get() and set() log at debug level on ENOENT and error level
      on all other errors, always.
      Also ensure that we don't sometimes write "failed to set 'x' to 'y':
      Success" when a partial write occurs.
  5. 24 Feb, 2014 2 commits
  6. 23 Feb, 2014 1 commit
  7. 18 Feb, 2014 1 commit
  8. 14 Feb, 2014 2 commits
    • Thomas Haller's avatar
      platform: do not check for _exists() before deleting addresses and routes · 2bc90a5f
      Thomas Haller authored
      Before, nm_platform_ip4_address_exists(), et al. look into the cache to see
      whether the address/route already exists and returned an error if it
      Change the semantic of the delete functions, to return success in case of
      "nothing to delete". Also always try to delete the object in the
      kernel. The reason is, that the cache might be out of date and the
      caller really wants to delete it. So, to be sure, we always delete.
      In most cases the object is actually in the cache (because that is
      how the caller came to know that such an object might exist).
      In those cases, the lookup was not useful either, because the object
      was actually cached.
      Signed-off-by: Thomas Haller's avatarThomas Haller <thaller@redhat.com>
    • Thomas Haller's avatar
      platform: add debug logging when adding/deleting routes · e54a3cca
      Thomas Haller authored
      Also, change the logging of nm_platform_ip._address_delete()
      to log what we are about to do, *before* checking for existing
      Signed-off-by: Thomas Haller's avatarThomas Haller <thaller@redhat.com>
  9. 30 Jan, 2014 5 commits
  10. 24 Jan, 2014 6 commits
  11. 22 Jan, 2014 1 commit
  12. 02 Dec, 2013 2 commits
  13. 15 Nov, 2013 1 commit
  14. 13 Nov, 2013 1 commit
  15. 08 Nov, 2013 1 commit
    • Thomas Haller's avatar
      platform: sort slaves after their master devices · 5023af9b
      Thomas Haller authored
      Slaves should get sorted after their masters so that when generating
      connections, the NMManager knows about the masters already.
      The convoluted logic here is to ensure that:
      1) the kernel doesn't pass bad information that causes NM to crash
      or infinite loop
      2) that with complicated parent/child relationships (like a VLAN interface
      with a parent that is also a slave), children always get sorted after
      *all* of their ancestors.  The previous code was only sorting children
      after their immediate parent/master's ifindex, but not actually after
      the parent in the returned list.
  16. 07 Nov, 2013 2 commits
  17. 06 Nov, 2013 1 commit
  18. 25 Sep, 2013 1 commit
    • Thomas Haller's avatar
      core: add nm_platform_ip[46]_*_cmp functions · b1113a0a
      Thomas Haller authored
      New functions to compare two instances of NMPlatformIP4Address, NMPlatformIP6Address,
      NMPlatformIP4Route, NMPlatformIP6Route, respectively.
      These functions return -1, 0 or 1 as result of the comparison. This is similar to
      strcmp with the additional restriction, that only one of these 3 values will be
      Signed-off-by: Thomas Haller's avatarThomas Haller <thaller@redhat.com>
  19. 30 Aug, 2013 1 commit
  20. 29 Aug, 2013 1 commit
  21. 26 Aug, 2013 1 commit
  22. 06 Aug, 2013 1 commit
    • Dan Winship's avatar
      platform: fix initial link-added emission · 1d3ae229
      Dan Winship authored
      nm_platform_query_devices() wasn't including the new "reason" argument
      in its link-added emissions. (This didn't cause any problems since
      NMManager doesn't look at that argument anyway, but it's still
      obviously wrong.)
  23. 03 Aug, 2013 1 commit
  24. 02 Aug, 2013 2 commits
  25. 01 Aug, 2013 1 commit