)]}'
{
  "commit": "deb47060b5812c6c41ecc790cb28898fcbc45c93",
  "tree": "633edbe482938ad02d306180b054811aa29f5269",
  "parents": [
    "49a621b2d3a01a2f998de9014ef14c1e61a9d5b0"
  ],
  "author": {
    "name": "Simon Marchi",
    "email": "simon.marchi@efficios.com",
    "time": "Thu Feb 05 15:42:50 2026 -0500"
  },
  "committer": {
    "name": "Simon Marchi",
    "email": "simon.marchi@polymtl.ca",
    "time": "Fri Feb 06 09:55:27 2026 -0500"
  },
  "message": "gdb/testsuite: make gdb.dwarf2/dw2-empty-inline-ranges.exp work when ASLR can\u0027t be disabled\n\nA colleague that goes by the name \"Luis Machado\" reported a failure when\nrunning gdb.dwarf2/dw2-empty-inline-ranges.exp.  After investigation, we\nfound that the conditions for the test to fail are:\n\n - running in an environment where GDB can\u0027t disable address space\n   randomization (such as in a container where that capability is\n   removed)\n\n - a toolchain generating position-independent executables\n\nThe test does a first run to grab the addresses of a few labels defined\nin the source file.  It then crafts the DWARF using these addresses.\nAnd then it does a second run for the actual test stuff.\n\nWhen the executable is PIE and ASLR is active, then the addresses in\nboth runs don\u0027t agree, which the test doesn\u0027t expect.\n\nIt is possible to emulate the failure by inserting some:\n\n    gdb_test_no_output \"set disable-randomization off\"\n\nafter both \"prepare_for_testing\" calls.  The (first) failure then\nmanifests as:\n\n    FAIL: gdb.dwarf2/dw2-empty-inline-ranges.exp: dwarf_version\u003d4: empty_loc\u003dstart: entry_pc_type\u003dempty: stopped at entry-pc\n\nThis test compares the expected stop PC \"entry_pc\" with the actual stop\nPC \"pc\".  In my example run, they were indeed different:\n\n    pc       \u003d 0x5603ec67a159\n    entry_pc \u003d 0x55baba6a9159\n\nThe simplest way to fix this, which this patch implements, is to use\n\"nopie\" when building the binaries.  I don\u0027t think this affects the\neffectiveness of the test.\n\nAlso, in the first run, it is longer necessary to run the inferior\nbefore grabbing the addresses, they are going to be the same with a\nnon-PIE executable.  So remove that.\n\nChange-Id: I3e9b1c056c30a49e01e1a1d3bf28f7e0e409f3c9\nApproved-By: Andrew Burgess \u003caburgess@redhat.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "ef75f642ac7dc512cd33d4ea574d71ba76316aa1",
      "old_mode": 33188,
      "old_path": "gdb/testsuite/gdb.dwarf2/dw2-empty-inline-ranges.exp",
      "new_id": "a3c88d950f8148f85fc4ec96e87737dba5b16040",
      "new_mode": 33188,
      "new_path": "gdb/testsuite/gdb.dwarf2/dw2-empty-inline-ranges.exp"
    }
  ]
}