)]}'
{
  "commit": "b8fa4100350432504df438014e2e5e9c1bbb6325",
  "tree": "0fc416746eeca68bbba228958636dd919fc3785f",
  "parents": [
    "d0023f820e003857248d14f2213ac3930283f16c",
    "ee684b6f2830047d19877e5547989740f18b1a5d"
  ],
  "author": {
    "name": "David S. Miller",
    "email": "davem@davemloft.net",
    "time": "Wed Feb 13 13:22:24 2013 -0500"
  },
  "committer": {
    "name": "David S. Miller",
    "email": "davem@davemloft.net",
    "time": "Wed Feb 13 13:22:24 2013 -0500"
  },
  "message": "Merge branch \u0027tcp_tsoffset\u0027\n\nAndrey Vagin says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nIf a TCP socket will get live-migrated from one box to another the\ntimestamps (which are typically ON) will get screwed up -- the new\nkernel will generate TS values that has nothing to do with what they\nwere on dump. The solution is to yet again fix the kernel and put a\n\"timestamp offset\" on a socket.\n\nA socket offset is added in places where externally visible tcp\ntimestamp option is parsed/initialized.\n\nConnections in the SYN_RECV state are not supported, global\ntcp_time_stamp is used for them, because repair mode doesn\u0027t support\nthis state. In a future it can be implemented by the similar way as for\nTIME_WAIT sockets.\n\nFor time-wait sockets offset is inhereted by a proper tcp_sock.\n\nA per-socket offset can be set only for sockets in repair mode.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n",
  "tree_diff": []
}