)]}'
{
  "commit": "421c4a00d9e2bdfffe9817600c4b4abb39464a3f",
  "tree": "29bfb096dd5f9023efa791a640aa586fc5790201",
  "parents": [
    "73dcc5e15bad199afc9154af04c1ac4e7b486620"
  ],
  "author": {
    "name": "Tom Tromey",
    "email": "tromey@adacore.com",
    "time": "Wed Aug 13 10:16:46 2025 -0600"
  },
  "committer": {
    "name": "Tom Tromey",
    "email": "tromey@adacore.com",
    "time": "Wed Aug 13 14:36:04 2025 -0600"
  },
  "message": "Refine range check in create_addrmap_from_gdb_index\n\nPR symtab/33247 points out that this check in\ncreate_addrmap_from_gdb_index:\n\n      if (lo \u003e hi)\n\t{\n\t  complaint (_(\".gdb_index address table has invalid range (%s - %s)\"),\n\t\t     hex_string (lo), hex_string (hi));\n\n... should probably use \"\u003e\u003d\" instead.  Reading a bit further the\nreason seems obvious:\n\n      mutable_map.set_empty (lo, hi - 1, index-\u003eunits[cu_index]);\n\nHere if lo\u003d\u003dhi, then this will insert a \"reversed\" range into the\naddrmap.\n\nApparently some LLVM tool can erroneously create a .gdb_index like\nthis.\n\nNo test because it seems like more trouble to write than it\u0027s worth.\n\nBug: https://sourceware.org/bugzilla/show_bug.cgi?id\u003d33247\nApproved-By: Simon Marchi \u003csimon.marchi@efficios.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "76317fe3191a46f629fc4d7ba253208dea10879c",
      "old_mode": 33188,
      "old_path": "gdb/dwarf2/read-gdb-index.c",
      "new_id": "7dfba7324513818a210d93f1c8510ab3148ce76b",
      "new_mode": 33188,
      "new_path": "gdb/dwarf2/read-gdb-index.c"
    }
  ]
}