)]}'
{
  "commit": "435f49a518c78eec8e2edbbadd912737246cbe20",
  "tree": "106df2617d42ace231e2fa9fcf1e0fd1075874ea",
  "parents": [
    "f56f44001cb5b40089deac094dbb74e5c9f64d81"
  ],
  "author": {
    "name": "Linus Torvalds",
    "email": "torvalds@linux-foundation.org",
    "time": "Fri Oct 29 10:36:49 2010 -0700"
  },
  "committer": {
    "name": "Linus Torvalds",
    "email": "torvalds@linux-foundation.org",
    "time": "Fri Oct 29 10:36:49 2010 -0700"
  },
  "message": "readv/writev: do the same MAX_RW_COUNT truncation that read/write does\n\nWe used to protect against overflow, but rather than return an error, do\nwhat read/write does, namely to limit the total size to MAX_RW_COUNT.\nThis is not only more consistent, but it also means that any broken\nlow-level read/write routine that still keeps counts in \u0027int\u0027 can\u0027t\nbreak.\n\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "52cfeb61da77c14ce7d3838db7517b5b6f88b213",
      "old_mode": 33188,
      "old_path": "fs/compat.c",
      "new_id": "ff66c0d7583d51eb672587a9c685cbc91320b167",
      "new_mode": 33188,
      "new_path": "fs/compat.c"
    },
    {
      "type": "modify",
      "old_id": "9cd9d148105d287b55c9f60af4034f300cc6d1af",
      "old_mode": 33188,
      "old_path": "fs/read_write.c",
      "new_id": "431a0ed610c83a8750592e516ff23ac35ffbe524",
      "new_mode": 33188,
      "new_path": "fs/read_write.c"
    },
    {
      "type": "modify",
      "old_id": "4d07902bc50c0e11dff997ffa16c4aa6e449c214",
      "old_mode": 33188,
      "old_path": "include/linux/fs.h",
      "new_id": "7b7b507ffa1cb269266e74044e8e57ec8e0b5d7b",
      "new_mode": 33188,
      "new_path": "include/linux/fs.h"
    }
  ]
}
