)]}'
{
  "commit": "3ebf44902f77537b5784eb5059c2b78d8b5a920a",
  "tree": "77290d029e6652ed09bb23c05b8b203bc569cd69",
  "parents": [
    "a0a24741cac414aba5918e9939afafa70c37f952"
  ],
  "author": {
    "name": "Steven Whitehouse",
    "email": "swhiteho@redhat.com",
    "time": "Tue Jul 10 12:28:27 2007 +0100"
  },
  "committer": {
    "name": "Steven Whitehouse",
    "email": "swhiteho@redhat.com",
    "time": "Tue Jul 10 12:28:27 2007 +0100"
  },
  "message": "[GFS2] Accept old format NFS filehandles\n\nOn Tue, 2007-07-10 at 10:06 +0100, Christoph Hellwig wrote:\n\u003e \u003e -#define GFS2_LARGE_FH_SIZE 10\n\u003e \u003e -\n\u003e \u003e -struct gfs2_fh_obj {\n\u003e \u003e -   struct gfs2_inum_host this;\n\u003e \u003e -   u32 imode;\n\u003e \u003e -};\n\u003e \u003e +#define GFS2_LARGE_FH_SIZE 8\n\u003e\n\u003e Because gfs2_decode_fh only accepts file handles with GFS2_LARGE_FH_SIZE\n\u003e or GFS2_LARGE_FH_SIZE you don\u0027t accept filehandles sent out by and older\n\u003e gfs version anymore.  Stale filehandles because of a new kernel version\n\u003e are a big no-no, so please add back code to handle the old filehandles\n\u003e on the decode side.\n\u003e\n\nThis should fix that problem I think since its only relating to end of\nthe fh we can just ignore that field in order to accept the older\nformat.\n\nSigned-off-by: Steven Whitehouse \u003cswhiteho@redhat.com\u003e\nCc: Christoph Hellwig \u003chch@infradead.org\u003e\nCc: Wendy Cheng \u003cwcheng@redhat.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "e317db2a5548c35ff97d77097f1fe54a0fd90331",
      "old_mode": 33188,
      "old_path": "fs/gfs2/ops_export.c",
      "new_id": "99ea5659bc2c58a5a41f482043ef25db8b6284ea",
      "new_mode": 33188,
      "new_path": "fs/gfs2/ops_export.c"
    }
  ]
}