757 lines · plain
1lit - LLVM Integrated Tester2============================3 4.. program:: lit5 6SYNOPSIS7--------8 9:program:`lit` [*options*] [*tests*]10 11DESCRIPTION12-----------13 14:program:`lit` is a portable tool for executing LLVM and Clang style test15suites, summarizing their results, and providing indication of failures.16:program:`lit` is designed to be a lightweight testing tool with as simple a17user interface as possible.18 19:program:`lit` should be run with one or more *tests* to run specified on the20command line. Tests can be either individual test files or directories to21search for tests (see :ref:`test-discovery`).22 23Each specified test will be executed (potentially concurrently) and once all24tests have been run :program:`lit` will print summary information on the number25of tests which passed or failed (see :ref:`test-status-results`). The26:program:`lit` program will execute with a non-zero exit code if any tests27fail.28 29By default :program:`lit` will use a succinct progress display and will only30print summary information for test failures. See :ref:`output-options` for31options controlling the :program:`lit` progress display and output.32 33:program:`lit` also includes a number of options for controlling how tests are34executed (specific features may depend on the particular test format). See35:ref:`execution-options` for more information.36 37Finally, :program:`lit` also supports additional options for only running a38subset of the options specified on the command line, see39:ref:`selection-options` for more information.40 41:program:`lit` parses options from the environment variable ``LIT_OPTS`` after42parsing options from the command line. ``LIT_OPTS`` is primarily useful for43supplementing or overriding the command-line options supplied to :program:`lit`44by ``check`` targets defined by a project's build system.45 46:program:`lit` can also read options from response files which are specified as47inputs using the ``@path/to/file.rsp`` syntax. Arguments read from a file must48be one per line and are treated as if they were in the same place as the49original file referencing argument on the command line. A response file can50reference other response files.51 52Users interested in the :program:`lit` architecture or designing a53:program:`lit` testing implementation should see :ref:`lit-infrastructure`.54 55GENERAL OPTIONS56---------------57 58.. option:: -h, --help59 60 Show the :program:`lit` help message and exit.61 62.. option:: --version63 64 Show :program:`lit`'s version number and exit.65 66.. option:: -j N, --workers N67 68 Run ``N`` tests in parallel. By default, this is automatically chosen to69 match the number of detected available CPUs.70 71.. option:: --config-prefix NAME72 73 Search for :file:`{NAME}.cfg` and :file:`{NAME}.site.cfg` when searching for74 test suites, instead of :file:`lit.cfg` and :file:`lit.site.cfg`.75 76.. option:: -D NAME[=VALUE], --param NAME[=VALUE]77 78 Add a user defined parameter ``NAME`` with the given ``VALUE`` (or the empty79 string if not given). The meaning and use of these parameters is test suite80 dependent.81 82.. _output-options:83 84OUTPUT OPTIONS85--------------86 87.. option:: -q, --quiet88 89 Suppress any output except for test failures.90 91.. option:: -s, --succinct92 93 Show less output, for example don't show information on tests that pass.94 Also show a progress bar, unless ``--no-progress-bar`` is specified.95 96.. option:: -v, --verbose97 98 Show more information on test failures, for example the entire test output99 instead of just the test result.100 101 Each command is printed before it is executed. This can be valuable for102 debugging test failures, as the last printed command is the one that failed.103 Moreover, :program:`lit` inserts ``'RUN: at line N'`` after each104 command pipeline in the output to help you locate the source line of105 the failed command.106 107.. option:: -vv, --echo-all-commands108 109 Deprecated alias for -v.110 111.. option:: -a, --show-all112 113 Enable -v, but for all tests not just failed tests.114 115.. option:: -o PATH, --output PATH116 117 Write test results to the provided path.118 119.. option:: --no-progress-bar120 121 Do not use curses based progress bar.122 123.. option:: --show-excluded124 125 Show excluded tests.126 127.. option:: --show-skipped128 129 Show skipped tests.130 131.. option:: --show-unsupported132 133 Show unsupported tests.134 135.. option:: --show-pass136 137 Show passed tests.138 139.. option:: --show-flakypass140 141 Show passed with retry tests.142 143.. option:: --show-xfail144 145 Show expectedly failed tests.146 147.. _execution-options:148 149EXECUTION OPTIONS150-----------------151 152.. option:: --gtest-sharding153 154 Enable sharding for GoogleTest format.155 156.. option:: --no-gtest-sharding157 158 Disable sharding for GoogleTest format.159 160.. option:: --path PATH161 162 Specify an additional ``PATH`` to use when searching for executables in tests.163 164.. option:: --vg165 166 Run individual tests under valgrind (using the memcheck tool). The167 ``--error-exitcode`` argument for valgrind is used so that valgrind failures168 will cause the program to exit with a non-zero status.169 170 When this option is enabled, :program:`lit` will also automatically provide a171 "``valgrind``" feature that can be used to conditionally disable (or expect172 failure in) certain tests.173 174.. option:: --vg-leak175 176 When :option:`--vg` is used, enable memory leak checks. When this option is177 enabled, :program:`lit` will also automatically provide a "``vg_leak``"178 feature that can be used to conditionally disable (or expect failure in)179 certain tests.180 181.. option:: --vg-arg ARG182 183 When :option:`--vg` is used, specify an additional argument to pass to184 :program:`valgrind` itself.185 186.. option:: --no-execute187 188 Don't execute any tests (assume that they pass).189 190.. option:: --xunit-xml-output XUNIT_XML_OUTPUT191 192 Write XUnit-compatible XML test reports to the specified file.193 194.. option:: --report-failures-only195 196 Only include failures (see :ref:`test-status-results`) in the report.197 198.. option:: --resultdb-output RESULTDB_OUTPUT199 200 Write LuCI ResultDB compatible JSON to the specified file.201 202.. option:: --time-trace-output TIME_TRACE_OUTPUT203 204 Write Chrome tracing compatible JSON to the specified file205 206.. option:: --timeout MAXINDIVIDUALTESTTIME207 208 Maximum time to spend running a single test (in seconds). 0 means no time209 limit. [Default: 0]210 211.. option:: --timeout N212 213 Spend at most ``N`` seconds (approximately) running each individual test.214 ``0`` means no time limit, and ``0`` is the default. Note that this is not an215 alias for :option:`--max-time`; the two are different kinds of maximums.216 217.. option:: --max-failures MAX_FAILURES218 219 Stop execution after the given number of failures.220 221.. option:: --max-retries-per-test N222 223 Retry running failed tests at most ``N`` times.224 Out of the following options to rerun failed tests the225 :option:`--max-retries-per-test` is the only one that doesn't226 require a change in the test scripts or the test config:227 228 * :option:`--max-retries-per-test` lit option229 * ``config.test_retry_attempts`` test suite option230 * ``ALLOW_RETRIES:`` annotation in test script231 232 Any option in the list above overrules its predecessor.233 234.. option:: --allow-empty-runs235 236 Do not fail the run if all tests are filtered out.237 238.. option:: --per-test-coverage239 240 Emit the necessary test coverage data, divided per test case (involves241 setting a unique value to LLVM_PROFILE_FILE for each RUN). The coverage242 data files will be emitted in the directory specified by ``config.test_exec_root``.243 244.. option:: --ignore-fail245 246 Exit with status zero even if some tests fail.247 248.. option:: --skip-test-time-recording249 250 Do not track elapsed wall time for each test.251 252.. option:: --time-tests253 254 Track the wall time individual tests take to execute and includes the results255 in the summary output. This is useful for determining which tests in a test256 suite take the most time to execute.257 258.. _selection-options:259 260SELECTION OPTIONS261-----------------262 263By default, `lit` will run failing tests first, then run tests in descending264execution time order to optimize concurrency. The execution order can be265changed using the :option:`--order` option.266 267The timing data is stored in the `test_exec_root` in a file named268`.lit_test_times.txt`. If this file does not exist, then `lit` checks the269`test_source_root` for the file to optionally accelerate clean builds.270 271.. option:: --max-tests N272 273 Run at most ``N`` tests and then terminate.274 275.. option:: --max-time N276 277 Spend at most ``N`` seconds (approximately) running tests and then terminate.278 Note that this is not an alias for :option:`--timeout`; the two are279 different kinds of maximums.280 281.. option:: --order {lexical,random,smart}282 283 Define the order in which tests are run. The supported values are:284 285 - lexical - tests will be run in lexical order according to the test file286 path. This option is useful when predictable test order is desired.287 288 - random - tests will be run in random order.289 290 - smart - tests that failed previously will be run first, then the remaining291 tests, all in descending execution time order. This is the default as it292 optimizes concurrency.293 294.. option:: --shuffle295 296 Run the tests in a random order, not failing/slowest first. Deprecated,297 use :option:`--order` instead.298 299.. option:: -i, --incremental300 301 Run failed tests first (DEPRECATED: use ``--order=smart``).302 303.. option:: --filter REGEXP304 305 Run only those tests whose name matches the regular expression specified in306 ``REGEXP``. The environment variable ``LIT_FILTER`` can be also used in place307 of this option, which is especially useful in environments where the call308 to ``lit`` is issued indirectly.309 310.. option:: --filter-out REGEXP311 312 Filter out those tests whose name matches the regular expression specified in313 ``REGEXP``. The environment variable ``LIT_FILTER_OUT`` can be also used in314 place of this option, which is especially useful in environments where the315 call to ``lit`` is issued indirectly.316 317.. option:: --xfail LIST318 319 Treat those tests whose name is in the semicolon separated list ``LIST`` as320 ``XFAIL``. This can be helpful when one does not want to modify the test321 suite. The environment variable ``LIT_XFAIL`` can be also used in place of322 this option, which is especially useful in environments where the call to323 ``lit`` is issued indirectly.324 325 A test name can specified as a file name relative to the test suite directory.326 For example:327 328 .. code-block:: none329 330 LIT_XFAIL="affinity/kmp-hw-subset.c;offloading/memory_manager.cpp"331 332 In this case, all of the following tests are treated as ``XFAIL``:333 334 .. code-block:: none335 336 libomp :: affinity/kmp-hw-subset.c337 libomptarget :: nvptx64-nvidia-cuda :: offloading/memory_manager.cpp338 libomptarget :: x86_64-pc-linux-gnu :: offloading/memory_manager.cpp339 340 Alternatively, a test name can be specified as the full test name341 reported in LIT output. For example, we can adjust the previous342 example not to treat the ``nvptx64-nvidia-cuda`` version of343 ``offloading/memory_manager.cpp`` as XFAIL:344 345 .. code-block:: none346 347 LIT_XFAIL="affinity/kmp-hw-subset.c;libomptarget :: x86_64-pc-linux-gnu :: offloading/memory_manager.cpp"348 349.. option:: --xfail-not LIST350 351 Do not treat the specified tests as ``XFAIL``. The environment variable352 ``LIT_XFAIL_NOT`` can also be used in place of this option. The syntax is the353 same as for :option:`--xfail` and ``LIT_XFAIL``. :option:`--xfail-not` and354 ``LIT_XFAIL_NOT`` always override all other ``XFAIL`` specifications,355 including an :option:`--xfail` appearing later on the command line. The356 primary purpose is to suppress an ``XPASS`` result without modifying a test357 case that uses the ``XFAIL`` directive.358 359.. option:: --exclude-xfail360 361 ``XFAIL`` tests won't be run, unless they are listed in the ``--xfail-not``362 (or ``LIT_XFAIL_NOT``) lists.363 364.. option:: --num-shards M365 366 Divide the set of selected tests into ``M`` equal-sized subsets or367 "shards", and run only one of them. Must be used with the368 ``--run-shard=N`` option, which selects the shard to run. The environment369 variable ``LIT_NUM_SHARDS`` can also be used in place of this370 option. These two options provide a coarse mechanism for partitioning large371 testsuites, for parallel execution on separate machines (say in a large372 testing farm).373 374.. option:: --run-shard N375 376 Select which shard to run, assuming the ``--num-shards=M`` option was377 provided. The two options must be used together, and the value of ``N``378 must be in the range ``1..M``. The environment variable379 ``LIT_RUN_SHARD`` can also be used in place of this option.380 381ADDITIONAL OPTIONS382------------------383 384.. option:: --debug385 386 Run :program:`lit` in debug mode, for debugging configuration issues and387 :program:`lit` itself.388 389.. option:: --show-suites390 391 List the discovered test suites and exit.392 393.. option:: --show-tests394 395 List all of the discovered tests and exit.396 397.. option:: --show-used-features398 399 Show all features used in the test suite (in ``XFAIL``, ``UNSUPPORTED`` and400 ``REQUIRES``) and exit.401 402.. option:: --update-tests403 404 Pass failing tests to functions in the ``lit_config.test_updaters`` list to405 check whether any of them know how to update the test to make it pass.406 407EXIT STATUS408-----------409 410:program:`lit` will exit with an exit code of 1 if there are any failures411(see :ref:`test-status-results`) and :option:`--ignore-fail` has not been412passed. Otherwise, it will exit with the status 0. Other exit codes are used413for non-test related failures (for example a user error or an internal program414error).415 416.. _test-discovery:417 418TEST DISCOVERY419--------------420 421The inputs passed to :program:`lit` can be either individual tests, or entire422directories or hierarchies of tests to run. When :program:`lit` starts up, the423first thing it does is convert the inputs into a complete list of tests to run424as part of *test discovery*.425 426In the :program:`lit` model, every test must exist inside some *test suite*.427:program:`lit` resolves the inputs specified on the command line to test suites428by searching upwards from the input path until it finds a :file:`lit.cfg` or429:file:`lit.site.cfg` file. These files serve as both a marker of test suites430and as configuration files which :program:`lit` loads in order to understand431how to find and run the tests inside the test suite.432 433Once :program:`lit` has mapped the inputs into test suites it traverses the434list of inputs adding tests for individual files and recursively searching for435tests in directories.436 437This behavior makes it easy to specify a subset of tests to run, while still438allowing the test suite configuration to control exactly how tests are439interpreted. In addition, :program:`lit` always identifies tests by the test440suite they are in, and their relative path inside the test suite. For441appropriately configured projects, this allows :program:`lit` to provide442convenient and flexible support for out-of-tree builds.443 444.. _test-status-results:445 446TEST STATUS RESULTS447-------------------448 449Each test ultimately produces one of the following eight results:450 451**PASS**452 453 The test succeeded.454 455**FLAKYPASS**456 457 The test succeeded after being re-run more than once. This only applies to458 tests containing an ``ALLOW_RETRIES:`` annotation.459 460**XFAIL**461 462 The test failed, but that is expected. This is used for test formats which allow463 specifying that a test does not currently work, but wish to leave it in the test464 suite.465 466**XPASS**467 468 The test succeeded, but it was expected to fail. This is used for tests which469 were specified as expected to fail, but are now succeeding (generally because470 the feature they test was broken and has been fixed).471 472**FAIL**473 474 The test failed.475 476**UNRESOLVED**477 478 The test result could not be determined. For example, this occurs when the test479 could not be run, the test itself is invalid, or the test was interrupted.480 481**UNSUPPORTED**482 483 The test is not supported in this environment. This is used by test formats484 which can report unsupported tests.485 486**TIMEOUT**487 488 The test was run, but it timed out before it was able to complete.489 490Unresolved (**UNRESOLVED**), timed out (**TIMEOUT**), failed (**FAIL**) and491unexpectedly passed (**XPASS**) tests are considered failures.492 493Depending on the test format tests may produce additional information about494their status (generally only for failures). See the :ref:`output-options`495section for more information.496 497.. _lit-infrastructure:498 499LIT INFRASTRUCTURE500------------------501 502This section describes the :program:`lit` testing architecture for users interested in503creating a new :program:`lit` testing implementation, or extending an existing one.504 505:program:`lit` proper is primarily an infrastructure for discovering and running506arbitrary tests, and to expose a single convenient interface to these507tests. :program:`lit` itself doesn't know how to run tests, rather this logic is508defined by *test suites*.509 510TEST SUITES511~~~~~~~~~~~512 513As described in :ref:`test-discovery`, tests are always located inside a *test514suite*. Test suites serve to define the format of the tests they contain, the515logic for finding those tests, and any additional information to run the tests.516 517:program:`lit` identifies test suites as directories containing ``lit.cfg`` or518``lit.site.cfg`` files (see also :option:`--config-prefix`). Test suites are519initially discovered by recursively searching up the directory hierarchy for520all the input files passed on the command line. You can use521:option:`--show-suites` to display the discovered test suites at startup.522 523Once a test suite is discovered, its config file is loaded. Config files524themselves are Python modules which will be executed. When the config file is525executed, two important global variables are predefined:526 527**lit_config**528 529 The global **lit** configuration object (a *LitConfig* instance), which defines530 the builtin test formats, global configuration parameters, and other helper531 routines for implementing test configurations.532 533**config**534 535 This is the config object (a *TestingConfig* instance) for the test suite,536 which the config file is expected to populate. The following variables are also537 available on the *config* object, some of which must be set by the config and538 others are optional or predefined:539 540 **name** *[required]* The name of the test suite, for use in reports and541 diagnostics.542 543 **test_format** *[required]* The test format object which will be used to544 discover and run tests in the test suite. Generally this will be a builtin test545 format available from the *lit.formats* module.546 547 **test_source_root** The filesystem path to the test suite root. For out-of-dir548 builds this is the directory that will be scanned for tests.549 550 **test_exec_root** For out-of-dir builds, the path to the test suite root inside551 the object directory. This is where tests will be run and temporary output files552 placed.553 554 **environment** A dictionary representing the environment to use when executing555 tests in the suite.556 557 **standalone_tests** When true, mark a directory with tests expected to be run558 standalone. Test discovery is disabled for that directory. *lit.suffixes* and559 *lit.excludes* must be empty when this variable is true.560 561 **suffixes** For **lit** test formats which scan directories for tests, this562 variable is a list of suffixes to identify test files. Used by: *ShTest*.563 564 **substitutions** For **lit** test formats which substitute variables into a test565 script, the list of substitutions to perform. Used by: *ShTest*.566 567 **unsupported** Mark an unsupported directory, all tests within it will be568 reported as unsupported. Used by: *ShTest*.569 570 **parent** The parent configuration, this is the config object for the directory571 containing the test suite, or None.572 573 **root** The root configuration. This is the top-most :program:`lit` configuration in574 the project.575 576 **pipefail** Normally a test using a shell pipe fails if any of the commands577 on the pipe fail. If this is not desired, setting this variable to false578 makes the test fail only if the last command in the pipe fails.579 580 **available_features** A set of features that can be used in `XFAIL`,581 `REQUIRES`, and `UNSUPPORTED` directives.582 583TEST DISCOVERY584~~~~~~~~~~~~~~585 586Once test suites are located, :program:`lit` recursively traverses the source587directory (following *test_source_root*) looking for tests. When :program:`lit`588enters a sub-directory, it first checks to see if a nested test suite is589defined in that directory. If so, it loads that test suite recursively,590otherwise it instantiates a local test config for the directory (see591:ref:`local-configuration-files`).592 593Tests are identified by the test suite they are contained within, and the594relative path inside that suite. Note that the relative path may not refer to595an actual file on disk; some test formats (such as *GoogleTest*) define596"virtual tests" which have a path that contains both the path to the actual597test file and a subpath to identify the virtual test.598 599.. _local-configuration-files:600 601LOCAL CONFIGURATION FILES602~~~~~~~~~~~~~~~~~~~~~~~~~603 604When :program:`lit` loads a subdirectory in a test suite, it instantiates a605local test configuration by cloning the configuration for the parent directory606--- the root of this configuration chain will always be a test suite. Once the607test configuration is cloned :program:`lit` checks for a *lit.local.cfg* file608in the subdirectory. If present, this file will be loaded and can be used to609specialize the configuration for each individual directory. This facility can610be used to define subdirectories of optional tests, or to change other611configuration parameters --- for example, to change the test format, or the612suffixes which identify test files.613 614SUBSTITUTIONS615~~~~~~~~~~~~~616 617:program:`lit` allows patterns to be substituted inside RUN commands. It also618provides the following base set of substitutions, which are defined in619TestRunner.py:620 621 ======================= ==============622 Macro Substitution623 ======================= ==============624 %s source path (path to the file currently being run)625 %S source dir (directory of the file currently being run)626 %p same as %S627 %{pathsep} path separator628 %{fs-src-root} root component of file system paths pointing to the LLVM checkout629 %{fs-tmp-root} root component of file system paths pointing to the test's temporary directory630 %{fs-sep} file system path separator631 %t a path unique to the test (which may be used to make files or directories)632 %basename_t The last path component of %t but without the ``.tmp`` extension (deprecated, use ``%{t:stem}`` instead)633 %% %634 %/s %s but ``\`` is replaced by ``/``635 %/S %S but ``\`` is replaced by ``/``636 %/p %p but ``\`` is replaced by ``/``637 %/t %t but ``\`` is replaced by ``/``638 %{s:basename} The last path component of %s639 %{t:stem} The last path component of %t but without the ``.tmp`` extension (alias for %basename_t)640 %{s:real} %s after expanding all symbolic links and substitute drives641 %{S:real} %S after expanding all symbolic links and substitute drives642 %{p:real} %p after expanding all symbolic links and substitute drives643 %{t:real} %t after expanding all symbolic links and substitute drives644 %{/s:real} %/s after expanding all symbolic links and substitute drives645 %{/S:real} %/S after expanding all symbolic links and substitute drives646 %{/p:real} %/p after expanding all symbolic links and substitute drives647 %{/t:real} %/t after expanding all symbolic links and substitute drives648 %{/s:regex_replacement} %/s but escaped for use in the replacement of a ``s@@@`` command in sed649 %{/S:regex_replacement} %/S but escaped for use in the replacement of a ``s@@@`` command in sed650 %{/p:regex_replacement} %/p but escaped for use in the replacement of a ``s@@@`` command in sed651 %{/t:regex_replacement} %/t but escaped for use in the replacement of a ``s@@@`` command in sed652 %:s On Windows, %/s but a ``:`` is removed if its the second character.653 Otherwise, %s but with a single leading ``/`` removed.654 %:S On Windows, %/S but a ``:`` is removed if its the second character.655 Otherwise, %S but with a single leading ``/`` removed.656 %:p On Windows, %/p but a ``:`` is removed if its the second character.657 Otherwise, %p but with a single leading ``/`` removed.658 %:t On Windows, %/t but a ``:`` is removed if its the second character.659 Otherwise, %t but with a single leading ``/`` removed.660 %{readfile:<filename>} Reads the file specified.661 ======================= ==============662 663Other substitutions are provided that are variations on this base set and664further substitution patterns can be defined by each test module. See the665modules :ref:`local-configuration-files`.666 667More detailed information on substitutions can be found in the668:doc:`../TestingGuide`.669 670TEST RUN OUTPUT FORMAT671~~~~~~~~~~~~~~~~~~~~~~672 673The :program:`lit` output for a test run conforms to the following schema, in674both short and verbose modes (although in short mode no PASS lines will be675shown). This schema has been chosen to be relatively easy to reliably parse by676a machine (for example in buildbot log scraping), and for other tools to677generate.678 679Each test result is expected to appear on a line that matches:680 681.. code-block:: none682 683 <result code>: <test name> (<progress info>)684 685where ``<result-code>`` is a standard test result such as PASS, FAIL, XFAIL,686XPASS, UNRESOLVED, or UNSUPPORTED. The performance result codes of IMPROVED and687REGRESSED are also allowed.688 689The ``<test name>`` field can consist of an arbitrary string containing no690newline.691 692The ``<progress info>`` field can be used to report progress information such693as (1/300) or can be empty, but even when empty the parentheses are required.694 695Should a test be allowed retries (see ``ALLOW_RETRIES:`` annotation) and it696needed more than one attempt to succeed, then ``<progress info>`` is extended697by this information:698 699.. code-block:: none700 701 , <num_attempts_made> of <max_allowed_attempts> attempts702 703Each test result may include additional (multiline) log information in the704following format:705 706.. code-block:: none707 708 <log delineator> TEST '(<test name>)' <trailing delineator>709 ... log message ...710 <log delineator>711 712where ``<test name>`` should be the name of a preceding reported test, ``<log713delineator>`` is a string of "*" characters *at least* four characters long714(the recommended length is 20), and ``<trailing delineator>`` is an arbitrary715(unparsed) string.716 717The following is an example of a test run output which consists of four tests A,718B, C, and D, and a log message for the failing test C:719 720.. code-block:: none721 722 PASS: A (1 of 4)723 PASS: B (2 of 4)724 FAIL: C (3 of 4)725 ******************** TEST 'C' FAILED ********************726 Test 'C' failed as a result of exit code 1.727 ********************728 PASS: D (4 of 4)729 730DEFAULT FEATURES731~~~~~~~~~~~~~~~~~732 733For convenience :program:`lit` automatically adds **available_features** for734some common use cases.735 736:program:`lit` adds a feature based on the operating system being built on, for737example: `system-darwin`, `system-linux`, etc. :program:`lit` also738automatically adds a feature based on the current architecture, for example739`target-x86_64`, `target-aarch64`, etc.740 741When building with sanitizers enabled, :program:`lit` automatically adds the742short name of the sanitizer, for example: `asan`, `tsan`, etc.743 744To see the full list of features that can be added, see745*llvm/utils/lit/lit/llvm/config.py*.746 747LIT EXAMPLE TESTS748~~~~~~~~~~~~~~~~~749 750The :program:`lit` distribution contains several example implementations of751test suites in the *ExampleTests* directory.752 753SEE ALSO754--------755 756valgrind(1)757