)]}'
{
  "commit": "cf97d5ebfe7129d5d9e7afa212eb0cadff215c3a",
  "tree": "9ba92504a8e08123bfa24009c694a70712320ddb",
  "parents": [
    "f7c08eeb6c00c5a99e0994619643ef5f57bf2ad4"
  ],
  "author": {
    "name": "Nathaniel Wesley Filardo",
    "email": "wes.filardo@scisemi.com",
    "time": "Mon Nov 11 16:21:33 2024 +0000"
  },
  "committer": {
    "name": "Nathaniel Wesley Filardo",
    "email": "VP331RHQ115POU58JFRLKB7OPA0L18E3@cmx.ietfng.org",
    "time": "Mon Nov 11 17:57:11 2024 +0000"
  },
  "message": "switcher: enable IRQs while in fault handlers\n\nFault handlers should not run with IRQs disabled, as that would allow\nany compartment with a fault handler to wedge the machine (and while in\nprinciple still auditable, that\u0027s a big step back from the structured\nIRQ disposition programming otherwise present in the RTOS).  This\nimplies that invariants established and protected under IRQ deferral no\nlonger hold on the fault path; the attendant loss of atomicity is just\nsomething the compartment will have to deal with.\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "1ea2bbf9cff9d69b1f4aca544db05fc93b51acbe",
      "old_mode": 33188,
      "old_path": "sdk/core/switcher/entry.S",
      "new_id": "c113d266dff1b68726e37a2b05eb328f1c51f3dd",
      "new_mode": 33188,
      "new_path": "sdk/core/switcher/entry.S"
    }
  ]
}