)]}'
{
  "commit": "c2bbb7f8a9618920546ff40ec59488ebb9b81da9",
  "tree": "39c03e0d6a13028e3271f87a13e39e876c82e982",
  "parents": [
    "503b3d0df3fb2054d4496107bf54b2a0cdc30d53"
  ],
  "author": {
    "name": "Al Viro",
    "email": "viro@zeniv.linux.org.uk",
    "time": "Sun Dec 23 17:00:51 2018 -0500"
  },
  "committer": {
    "name": "Al Viro",
    "email": "viro@zeniv.linux.org.uk",
    "time": "Sun Dec 23 19:50:03 2018 -0500"
  },
  "message": "convert alloc_super() to struct fs_context\n\nIMO it\u0027s premature by several cycles - David might hope to get\nrid of sget_userns() and friends, but it will take a while and\nuntil that\u0027s done...  Let\u0027s just say that on-stack instances\nof struct fs_context, skipping the normal constructors and not\nbeing fed to destructor are Not Nice(tm) in general.  Yes, in\nthis particular case we are safe, but it really doesn\u0027t buy\nus anything worthwhile.\n\n*IF* all callers of alloc_super() end up with handy fs_context -\nsure, this will be the obvious change.  Until then... I\u0027m taking\nthat to the very end of the series lest it gets lost, but it\u0027s\nnot going to be in any pull requests this cycle.  And probably\nthe next cycle or two as well.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "aef58beb0edb69cd3dd14d951f9b76e69d12e577",
      "old_mode": 33188,
      "old_path": "fs/super.c",
      "new_id": "254291b911c9584c1c2461f0f2d3d18e6f989f8f",
      "new_mode": 33188,
      "new_path": "fs/super.c"
    }
  ]
}