)]}'
{
  "commit": "45575f5a426c01ec68cd33d998267ba2f6278fac",
  "tree": "f34a857f471e5249bda3fca723e4b69ba6ce552c",
  "parents": [
    "a90110c61073eab95d1986322693c2b9a8a6a5f6"
  ],
  "author": {
    "name": "Anton Blanchard",
    "email": "anton@samba.org",
    "time": "Mon Mar 22 17:47:59 2010 +1100"
  },
  "committer": {
    "name": "Linus Torvalds",
    "email": "torvalds@linux-foundation.org",
    "time": "Mon Mar 22 09:57:19 2010 -0700"
  },
  "message": "ppc64 sys_ipc breakage in 2.6.34-rc2\n\nI chased down a fail on ppc64 on 2.6.34-rc2 where an application that\nuses shared memory was getting a SEGV.\n\nCommit baed7fc9b580bd3fb8252ff1d9b36eaf1f86b670 (\"Add generic sys_ipc\nwrapper\") changed the second argument from an unsigned long to an int.\nWhen we call shmget the system call wrappers for sys_ipc will sign\nextend second (ie the size) which truncates it.  It took a while to\ntrack down because the call succeeds and strace shows the untruncated\nsize :)\n\nThe patch below changes second from an int to an unsigned long which\nfixes shmget on ppc64 (and I assume s390, sparc64 and mips64).\n\nSigned-off-by: Anton Blanchard \u003canton@samba.org\u003e\n--\n\nI assume the function prototypes for the other IPC methods would cause us\nto sign or zero extend second where appropriate (avoiding any security\nissues). Come to think of it, the syscall wrappers for each method should do\nthat for us as well.\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "355a3da9ec73dd8bba5f596c1e1d026e53d881ea",
      "old_mode": 33188,
      "old_path": "ipc/syscall.c",
      "new_id": "1d6f53f6b562441bcf5ef99c85067275eecf6c5a",
      "new_mode": 33188,
      "new_path": "ipc/syscall.c"
    }
  ]
}