)]}'
{
  "commit": "3ef0f720e47e895b613b0305eb0a483e3ec11f23",
  "tree": "e696a950d76c90199661515e1068fc00102a15bf",
  "parents": [
    "3b73a223661ed137c5d3d2635f954382e94f5a43"
  ],
  "author": {
    "name": "Miklos Szeredi",
    "email": "mszeredi@suse.cz",
    "time": "Wed May 14 16:05:37 2008 -0700"
  },
  "committer": {
    "name": "Linus Torvalds",
    "email": "torvalds@linux-foundation.org",
    "time": "Wed May 14 19:11:13 2008 -0700"
  },
  "message": "mm: fix infinite loop in filemap_fault\n\nfilemap_fault will go into an infinite loop if -\u003ereadpage() fails\nasynchronously.\n\nAFAICS the bug was introduced by this commit, which removed the wait after the\nfinal readpage:\n\n   commit d00806b183152af6d24f46f0c33f14162ca1262a\n   Author: Nick Piggin \u003cnpiggin@suse.de\u003e\n   Date:   Thu Jul 19 01:46:57 2007 -0700\n\n       mm: fix fault vs invalidate race for linear mappings\n\nFix by reintroducing the wait_on_page_locked() after -\u003ereadpage() to make sure\nthe page is up-to-date before jumping back to the beginning of the function.\n\nI\u0027ve noticed this while testing nfs exporting on fuse.  The patch\nfixes it.\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@suse.cz\u003e\nCc: Nick Piggin \u003cnpiggin@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "2dead9adf8b76806e66ac1b0169510773be58e14",
      "old_mode": 33188,
      "old_path": "mm/filemap.c",
      "new_id": "1e6a7d34874fd31f3e9549c7540a4ce28c2402c6",
      "new_mode": 33188,
      "new_path": "mm/filemap.c"
    }
  ]
}