)]}'
{
  "commit": "e06b8b98da071f7dd78fb7822991694288047df0",
  "tree": "f948b3f6ee9303bcce76e6500f0aba335b8b1d0e",
  "parents": [
    "587c90616a5b44e6ccfac38e64d4fecee51d588c"
  ],
  "author": {
    "name": "Sam Ravnborg",
    "email": "sam@ravnborg.org",
    "time": "Wed Feb 13 22:43:28 2008 +0100"
  },
  "committer": {
    "name": "Sam Ravnborg",
    "email": "sam@ravnborg.org",
    "time": "Thu Feb 14 23:33:21 2008 +0100"
  },
  "message": "kbuild: allow -fstack-protector to take effect\n\nArjan van de Ven \u003carjan@infradead.org\u003e wrote:\n\u003d\u003d\u003d\nI just read the excellent LWN writeup of the vmsplice\nsecurity thing, and that got me wondering why this attack\nwasn\u0027t stopped by the CONFIG_CC_STACKPROTECTOR option...\nbecause it plain should have been...\n\nSome analysis later.. it turns out that the following line\nin the top level Makefile, added by you in October 2007,\nentirely disables CONFIG_CC_STACKPROTECTOR ;(\nWith this line removed the exploit will be nicely stopped.\n\nCFLAGS          +\u003d $(call cc-option, -fno-stack-protector)\n\nNow I realize that certain distros have patched gcc to\ncompensate for their lack of distro wide CFLAGS, and it\u0027s\ngreat to work around that... but would there be a way to NOT\ndisable this for CONFIG_CC_STACKPROTECTOR please?\nIt would have made this exploit not possible for those kernels\nthat enable this feature (and that includes distros like Fedora)\n\u003d\u003d\u003d\n\nMove the assignment to KBUILD_CFLAGS up before including\nthe arch specific Makefile so arch makefiles may override\nthe setting.\n\nSigned-off-by: Sam Ravnborg \u003csam@ravnborg.org\u003e\nCc: Arjan van de Ven \u003carjan@infradead.org\u003e\nCc: stable@kernel.org\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "c162370c7367383005f11e654501db4b1005e450",
      "old_mode": 33188,
      "old_path": "Makefile",
      "new_id": "d73865188372362facec7bc1372917332667a49f",
      "new_mode": 33188,
      "new_path": "Makefile"
    }
  ]
}