)]}'
{
  "commit": "3ec033fab4ac2e6289b5928f949e054ff89e8ae8",
  "tree": "90636fe5cccaf5d54fc38970c5cd0bcbfb81369e",
  "parents": [
    "fb908c299d5497a28d7e319c21d366989cfa7743"
  ],
  "author": {
    "name": "Kevin Buettner",
    "email": "kevinb@redhat.com",
    "time": "Thu Sep 21 17:58:05 2023 -0700"
  },
  "committer": {
    "name": "Kevin Buettner",
    "email": "kevinb@redhat.com",
    "time": "Thu Sep 28 18:46:04 2023 -0700"
  },
  "message": "gdb/testsuite: Add relative versus absolute LD_LIBRARY_PATH test\n\nAt one time, circa 2006, there was a bug, which was presumably fixed\nwithout adding a test case:\n\n    If you provided some relative path to the shared library, such as\n    with\n\n\texport LD_LIBRARY_PATH\u003d.\n\n    then gdb would fail to match the shared library name during the\n    TLS lookup.\n\nI think there may have been a bit more to it than is provided by that\nexplanation, since the test also takes care to split the debug info\ninto a separate file.\n\nIn any case, this commit is based on one of Red Hat\u0027s really old\nlocal patches.  I\u0027ve attempted to update it and remove a fair amount\nof cruft, hopefully without losing any critical elements from the\ntest.\n\nTesting on Fedora 38 (correctly) shows 1 unsupported test for\nnative-gdbserver and 5 PASSes for the native target as well as\nnative-extended-gdbserver.\n\nIn his review of v1 of this patch, Lancelot SIX observed that\n\u0027thread_local\u0027 could be used in place of \u0027__thread\u0027 in the C source\nfiles.  But it only became available via the standard in C11, so I\nused additional_flags\u003d-std\u003dc11 for compiling both the shared object\nand the main program.\n\nAlso, while testing with CC_FOR_TARGET\u003dclang, I found that\n\u0027additional_flags\u003d-Wl,-soname\u003d${binsharedbase}\u0027 caused clang\nto complain that this linker flag was unused when compiling\nthe source file, so I moved this linker option to \u0027ldflags\u003d\u0027.\n\nMy testing for this v2 patch shows the same results as with v1,\nbut I\u0027ve done additional testing with CC_FOR_TARGET\u003dclang as\nwell.  The results are the same as when gcc is used.\n\nCo-Authored-by: Jan Kratochvil \u003cjan@jankratochvil.net\u003e\nReviewed-By: Lancelot Six \u003clancelot.six@amd.com\u003e\n",
  "tree_diff": [
    {
      "type": "add",
      "old_id": "0000000000000000000000000000000000000000",
      "old_mode": 0,
      "old_path": "/dev/null",
      "new_id": "0c0903a0453f268ad8fe35c8b4524edb13cd7557",
      "new_mode": 33188,
      "new_path": "gdb/testsuite/gdb.threads/tls-sepdebug-main.c"
    },
    {
      "type": "add",
      "old_id": "0000000000000000000000000000000000000000",
      "old_mode": 0,
      "old_path": "/dev/null",
      "new_id": "c051eb1041d1dbe7e486a12411cf9d8f5cdfc833",
      "new_mode": 33188,
      "new_path": "gdb/testsuite/gdb.threads/tls-sepdebug-shared.c"
    },
    {
      "type": "add",
      "old_id": "0000000000000000000000000000000000000000",
      "old_mode": 0,
      "old_path": "/dev/null",
      "new_id": "ad67bfd37def3eb817e3d24d4dcee2c543c6f651",
      "new_mode": 33188,
      "new_path": "gdb/testsuite/gdb.threads/tls-sepdebug.exp"
    }
  ]
}