)]}'
{
  "commit": "982d007a6eec4a0abb404d2355eeec2c041c61ea",
  "tree": "c050631e7fbf5be10edb45ee3be0ded2670b7512",
  "parents": [
    "84d88d5d4efc37dfb8a93a4a58d8a227ee86ffa4"
  ],
  "author": {
    "name": "Linus Torvalds",
    "email": "torvalds@linux-foundation.org",
    "time": "Wed Sep 30 17:57:27 2009 -0700"
  },
  "committer": {
    "name": "Ingo Molnar",
    "email": "mingo@elte.hu",
    "time": "Thu Oct 01 08:01:08 2009 +0200"
  },
  "message": "x86: Optimize cmpxchg64() at build-time some more\n\nTry to avoid the \u0027alternates()\u0027 code when we can statically\ndetermine that cmpxchg8b is fine. We already have that\nCONFIG_x86_CMPXCHG64 (enabled by PAE support), and we could easily\nalso enable it for some of the CPU cases.\n\nNote, this patch only adds CMPXCHG8B for the obvious Intel CPU\u0027s,\nnot for others. (There was something really messy about cmpxchg8b\nand clone CPU\u0027s, so if you enable it on other CPUs later, do it\ncarefully.)\n\nIf we avoid that asm-alternative thing when we can assume the\ninstruction exists, we\u0027ll generate less support crud, and we\u0027ll\navoid the whole issue with that extra \u0027nop\u0027 for padding instruction\nsizes etc.\n\nLKML-Reference: \u003calpine.LFD.2.01.0909301743150.6996@localhost.localdomain\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "527519b8a9f9997241eb5897f316cdbe855bb341",
      "old_mode": 33188,
      "old_path": "arch/x86/Kconfig.cpu",
      "new_id": "f2824fb8c79cad23ad747977e1e32ab4d6b555a1",
      "new_mode": 33188,
      "new_path": "arch/x86/Kconfig.cpu"
    }
  ]
}