)]}'
{
  "commit": "e26ff6c44e7546b9f761e00bb7cf76a3fa8bfc5d",
  "tree": "0b517db8c2a89950a86ae2b0700b02bc2afd3190",
  "parents": [
    "5a7f5aa29b33e9cd01a0062bcf8d2aa4d76852f9"
  ],
  "author": {
    "name": "Alan Modra",
    "email": "amodra@gmail.com",
    "time": "Fri Jul 26 08:49:48 2024 +0930"
  },
  "committer": {
    "name": "Alan Modra",
    "email": "amodra@gmail.com",
    "time": "Fri Jul 26 10:34:54 2024 +0930"
  },
  "message": "ARM print_insn_mve assertion\n\nThis corrects objdump -d -m armv8.1-m.main output for a testcase found\nby oss-fuzz, .inst 0xee2fee79, which hits an assertion.\n\nObviously the switch case constants should be binary, not hex.\nCorrecting that is enough to cure this assertion, but I don\u0027t see any\npoint in singling out the invalid case 0b10.  In fact, it is just plain\nwrong to print \"undefined instruction: size equals zero    undefined\ninstruction: size equals two\".\n\nI also don\u0027t see the need for defensive programming here as is done\nelsewhere in checking that \"value\" is in range before indexing\nmve_vec_sizename.  There is exactly one MVE_VSHLL_T2 entry in\nmve_opcodes.  It is easy to verify that \"value\" is only two bits.\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "480e4c21655971980fabe1e2b51fb12f049003ce",
      "old_mode": 33188,
      "old_path": "opcodes/arm-dis.c",
      "new_id": "d1d7ca309934d31daf47bc529d2c1bb8313022f5",
      "new_mode": 33188,
      "new_path": "opcodes/arm-dis.c"
    }
  ]
}