Correct check for buffer/no buffer

fix testcase

  Fix bug regarding node that missed 2 LCP's (that was not included in next LCP after SR)

  (autotest tests)

  yet another test prg bug

  add list of pages being unmaped, so we can wait for it to be empty before dropping extents

  (introduced by only updating extent pages after pageout)

  (in test framework)

    handle partially complete LCP better in SR

    - only update extent pages *after* flush of real page has been done
    - sync both create/drop of table into undolog (for disk tables)

  (problem introduced by stew's timeout handling)

  fix SR1 and SR2

  add even more tests

  (recommit for correct merge order)

  Dont add insert+delete tuple(s) to lcp-keep-list

into  perch.ndb.mysql.com:/home/jonas/src/51-telco-gca

  dont assert if 2 LCP's are being run during a node recovery

  fix also undo files

  continue on unknown result (no contact)

  make free 64 bit (as it represents free bytes, not free extents as originally designed)

into  perch.ndb.mysql.com:/home/jonas/src/51-telco-gca

  Not very clever fix for DIH incorrect REDO handling
  - Dont report GCP_SAVE_CONF until first LCP has been complete during NR

into  perch.ndb.mysql.com:/home/jonas/src/51-telco-gca

  In TC init node status for already started nodes during node restart
  (not present in 5.1)

into  perch.ndb.mysql.com:/home/jonas/src/51-telco-gca

    Add better handling of GCP Stop
    Only kill "offending" node

into  clam.ndb.mysql.com:/space/pekka/ndb/version/my51-bug29176

into  perch.ndb.mysql.com:/home/jonas/src/51-telco-gca

into  perch.ndb.mysql.com:/home/jonas/src/51-telco-gca

  Fix 3 problems in REDO validation that could(will) lead to "Error while reading REDO log"

- implement read ahead during disk data scan

- add config param to have better behavior with large tables

  Improve buddy high order allocation
  Make removeCommonArea O(1) instead of O(N)
  Add limit to left/right search