)]}'
{
  "commit": "2e87c0d40a387c5b1f1afd3ce61ecdc7dad0e3e8",
  "tree": "3bf6c4ccefca7193eaa1474f4151033aa3935cfc",
  "parents": [
    "cdbfce275777f2fd142e3a3c73469807a4c40207"
  ],
  "author": {
    "name": "Nico Huber",
    "email": "nico.h@gmx.de",
    "time": "Sat Apr 18 00:46:39 2020 +0200"
  },
  "committer": {
    "name": "Matt DeVillier",
    "email": "matt.devillier@gmail.com",
    "time": "Fri May 08 21:40:53 2020 +0000"
  },
  "message": "gma: Map dummy PTEs for buggy VT-d\n\nExtend Setup_Default_GTT() to add 128 dummy page table entries\nafter the framebuffer. Apparently the IOMMU may report spurious\ntype 6 errors if nothing is mapped in this range. Also check for\ncorrect alignment of the start of the framebuffer and that the\nadditonal PTEs fit into the GTT.\n\nWithout such dummy entries, an OS driver might report failed\nreads at random addresses. It may or may not look like this:\n\n    DMAR: [DMA Read] Request device [00:02.0] PASID ffffffff fault addr\\\n    5669eb8000 [fault reason 06] PTE Read access is not set\n\nChange-Id: I1b97c6b44c1ffb37d119541c1c97ffe21d244da8\nSigned-off-by: Nico Huber \u003cnico.h@gmx.de\u003e\nReviewed-on: https://review.coreboot.org/c/libgfxinit/+/40499\nReviewed-by: Michael Niewöhner\nReviewed-by: Paul Menzel \u003cpaulepanter@users.sourceforge.net\u003e\nReviewed-by: Angel Pons \u003cth3fanbus@gmail.com\u003e\nReviewed-by: Matt DeVillier \u003cmatt.devillier@gmail.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "be623e431ca5d93592519802f6843845fc0b138d",
      "old_mode": 33188,
      "old_path": "common/hw-gfx-gma.adb",
      "new_id": "9157bbe38621954d0b96a79e964d4afffd6e70e2",
      "new_mode": 33188,
      "new_path": "common/hw-gfx-gma.adb"
    }
  ]
}