)]}'
{
  "commit": "84a0bd5b2830722cf80ff6ad33ef98101a947e14",
  "tree": "942f596d5a6b48bde181ed2ab063d697a0f2e73d",
  "parents": [
    "a9fa20a7af1f152d2d89c44c274a310ac654e3ad"
  ],
  "author": {
    "name": "Jason Wessel",
    "email": "jason.wessel@windriver.com",
    "time": "Thu Aug 05 09:22:19 2010 -0500"
  },
  "committer": {
    "name": "Jason Wessel",
    "email": "jason.wessel@windriver.com",
    "time": "Thu Aug 05 09:22:19 2010 -0500"
  },
  "message": "gdbstub: Optimize kgdb\u0027s \"thread:\" response for the gdb serial protocol\n\nThe gdb debugger understands how to parse short versions of the thread\nreference string as long as the bytes are paired in sets of two\ncharacters.  The kgdb implementation was always sending 8 leading\nzeros which could be omitted, and further optimized in the case of\nnon-negative thread numbers.  The negative numbers are used to\nreference a specific cpu in the case of kgdb.\n\nAn example of the previous i386 stop packet looks like:\n    T05thread:00000000000003bb;\n\nNew stop packet response:\n    T05thread:03bb;\n\nThe previous ThreadInfo response looks like:\n    m00000000fffffffe,0000000000000001,0000000000000002,0000000000000003,0000000000000004,0000000000000005,0000000000000006,0000000000000007,000000000000000c,0000000000000088,000000000000008a,000000000000008b,000000000000008c,000000000000008d,000000000000008e,00000000000000d4,00000000000000d5,00000000000000dd\n\nNew ThreadInfo response:\n    mfffffffe,01,02,03,04,05,06,07,0c,88,8a,8b,8c,8d,8e,d4,d5,dd\n\nA few bytes saved means better response time when using kgdb over a\nserial line.\n\nSigned-off-by: Jason Wessel \u003cjason.wessel@windriver.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "3517fd7196351495ae4ed075afe6ad9602482080",
      "old_mode": 33188,
      "old_path": "kernel/debug/gdbstub.c",
      "new_id": "e117cfd75887c2423fb8b56dec7923ed8c7d711e",
      "new_mode": 33188,
      "new_path": "kernel/debug/gdbstub.c"
    }
  ]
}