)]}'
{
  "commit": "cafdd8ba08935d9b161bb781851dc4c0e6f70427",
  "tree": "ee6a5305902cfd6abf818e398c538694db24dcb7",
  "parents": [
    "c33880aaddbbab1ccf36f4457ed1090621f2e39a"
  ],
  "author": {
    "name": "William Lee Irwin III",
    "email": "wli@holomorphy.com",
    "time": "Tue May 24 19:31:09 2005 -0700"
  },
  "committer": {
    "name": "Linus Torvalds",
    "email": "torvalds@ppc970.osdl.org",
    "time": "Tue May 24 20:08:13 2005 -0700"
  },
  "message": "[PATCH] try_to_unmap_cluster() passes out-of-bounds pte to pte_unmap()\n\ntry_to_unmap_cluster() does:\n        for (pte \u003d pte_offset_map(pmd, address);\n                        address \u003c end; pte++, address +\u003d PAGE_SIZE) {\n\t\t...\n\t}\n\n\tpte_unmap(pte);\n\nIt may take a little staring to notice, but pte can actually fall off the\nend of the pte page in this iteration, which makes life difficult for\nkmap_atomic() and the users not expecting it to BUG().  Of course, we\u0027re\nsomewhat lucky in that arithmetic elsewhere in the function guarantees that\nat least one iteration is made, lest this force larger rearrangements to be\nmade.  This issue and patch also apply to non-mm mainline and with trivial\nadjustments, at least two related kernels.\n\nDiscovered during internal testing at Oracle.\n\nSigned-off-by: William Irwin \u003cwli@holomorphy.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "a6203b4e12786f4bdbf9ad8aefbb0ae287c8a91a",
      "old_mode": 33188,
      "old_path": "mm/rmap.c",
      "new_id": "9827409eb7c7ea4026e27b34fdaceacd909cabc3",
      "new_mode": 33188,
      "new_path": "mm/rmap.c"
    }
  ]
}