1 lit - LLVM Integrated Tester
2 ============================
9 **lit** [*options*] [*tests*]
16 **lit** is a portable tool for executing LLVM and Clang style test suites,
17 summarizing their results, and providing indication of failures. **lit** is
18 designed to be a lightweight testing tool with as simple a user interface as
21 **lit** should be run with one or more *tests* to run specified on the command
22 line. Tests can be either individual test files or directories to search for
23 tests (see "TEST DISCOVERY").
25 Each specified test will be executed (potentially in parallel) and once all
26 tests have been run **lit** will print summary information on the number of tests
27 which passed or failed (see "TEST STATUS RESULTS"). The **lit** program will
28 execute with a non-zero exit code if any tests fail.
30 By default **lit** will use a succinct progress display and will only print
31 summary information for test failures. See "OUTPUT OPTIONS" for options
32 controlling the **lit** progress display and output.
34 **lit** also includes a number of options for controlling how tests are executed
35 (specific features may depend on the particular test format). See "EXECUTION
36 OPTIONS" for more information.
38 Finally, **lit** also supports additional options for only running a subset of
39 the options specified on the command line, see "SELECTION OPTIONS" for
42 Users interested in the **lit** architecture or designing a **lit** testing
43 implementation should see "LIT INFRASTRUCTURE"
53 Show the **lit** help message.
57 **-j** *N*, **--threads**\ =\ *N*
59 Run *N* tests in parallel. By default, this is automatically chosen to match
60 the number of detected available CPUs.
64 **--config-prefix**\ =\ *NAME*
66 Search for *NAME.cfg* and *NAME.site.cfg* when searching for test suites,
67 instead of *lit.cfg* and *lit.site.cfg*.
71 **--param** *NAME*, **--param** *NAME*\ =\ *VALUE*
73 Add a user defined parameter *NAME* with the given *VALUE* (or the empty
74 string if not given). The meaning and use of these parameters is test suite
87 Suppress any output except for test failures.
91 **-s**, **--succinct**
93 Show less output, for example don't show information on tests that pass.
99 Show more information on test failures, for example the entire test output
100 instead of just the test result.
104 **--no-progress-bar**
106 Do not use curses based progress bar.
116 **--path**\ =\ *PATH*
118 Specify an addition *PATH* to use when searching for executables in tests.
124 Run individual tests under valgrind (using the memcheck tool). The
125 *--error-exitcode* argument for valgrind is used so that valgrind failures will
126 cause the program to exit with a non-zero status.
128 When this option is enabled, **lit** will also automatically provide a
129 "valgrind" feature that can be used to conditionally disable (or expect failure
134 **--vg-arg**\ =\ *ARG*
136 When *--vg* is used, specify an additional argument to pass to valgrind itself.
142 When *--vg* is used, enable memory leak checks. When this option is enabled,
143 **lit** will also automatically provide a "vg_leak" feature that can be
144 used to conditionally disable (or expect failure in) certain tests.
151 Track the wall time individual tests take to execute and includes the results in
152 the summary output. This is useful for determining which tests in a test suite
153 take the most time to execute. Note that this option is most useful with *-j
164 **--max-tests**\ =\ *N*
166 Run at most *N* tests and then terminate.
170 **--max-time**\ =\ *N*
172 Spend at most *N* seconds (approximately) running tests and then terminate.
178 Run the tests in a random order.
190 Run **lit** in debug mode, for debugging configuration issues and **lit** itself.
196 List the discovered test suites as part of the standard output.
202 Run Tcl scripts internally (instead of converting to shell scripts).
208 Run each test *N* times. Currently this is primarily useful for timing tests,
209 other results are not collated in any reasonable fashion.
218 **lit** will exit with an exit code of 1 if there are any FAIL or XPASS
219 results. Otherwise, it will exit with the status 0. Other exit codes are used
220 for non-test related failures (for example a user error or an internal program
228 The inputs passed to **lit** can be either individual tests, or entire
229 directories or hierarchies of tests to run. When **lit** starts up, the first
230 thing it does is convert the inputs into a complete list of tests to run as part
233 In the **lit** model, every test must exist inside some *test suite*. **lit**
234 resolves the inputs specified on the command line to test suites by searching
235 upwards from the input path until it finds a *lit.cfg* or *lit.site.cfg*
236 file. These files serve as both a marker of test suites and as configuration
237 files which **lit** loads in order to understand how to find and run the tests
238 inside the test suite.
240 Once **lit** has mapped the inputs into test suites it traverses the list of
241 inputs adding tests for individual files and recursively searching for tests in
244 This behavior makes it easy to specify a subset of tests to run, while still
245 allowing the test suite configuration to control exactly how tests are
246 interpreted. In addition, **lit** always identifies tests by the test suite they
247 are in, and their relative path inside the test suite. For appropriately
248 configured projects, this allows **lit** to provide convenient and flexible
249 support for out-of-tree builds.
256 Each test ultimately produces one of the following six results:
267 The test failed, but that is expected. This is used for test formats which allow
268 specifying that a test does not currently work, but wish to leave it in the test
275 The test succeeded, but it was expected to fail. This is used for tests which
276 were specified as expected to fail, but are now succeeding (generally because
277 the feature they test was broken and has been fixed).
289 The test result could not be determined. For example, this occurs when the test
290 could not be run, the test itself is invalid, or the test was interrupted.
296 The test is not supported in this environment. This is used by test formats
297 which can report unsupported tests.
301 Depending on the test format tests may produce additional information about
302 their status (generally only for failures). See the Output|"OUTPUT OPTIONS"
303 section for more information.
310 This section describes the **lit** testing architecture for users interested in
311 creating a new **lit** testing implementation, or extending an existing one.
313 **lit** proper is primarily an infrastructure for discovering and running
314 arbitrary tests, and to expose a single convenient interface to these
315 tests. **lit** itself doesn't know how to run tests, rather this logic is
316 defined by *test suites*.
322 As described in "TEST DISCOVERY", tests are always located inside a *test
323 suite*. Test suites serve to define the format of the tests they contain, the
324 logic for finding those tests, and any additional information to run the tests.
326 **lit** identifies test suites as directories containing *lit.cfg* or
327 *lit.site.cfg* files (see also **--config-prefix**). Test suites are initially
328 discovered by recursively searching up the directory hierarchy for all the input
329 files passed on the command line. You can use **--show-suites** to display the
330 discovered test suites at startup.
332 Once a test suite is discovered, its config file is loaded. Config files
333 themselves are Python modules which will be executed. When the config file is
334 executed, two important global variables are predefined:
339 The global **lit** configuration object (a *LitConfig* instance), which defines
340 the builtin test formats, global configuration parameters, and other helper
341 routines for implementing test configurations.
347 This is the config object (a *TestingConfig* instance) for the test suite,
348 which the config file is expected to populate. The following variables are also
349 available on the *config* object, some of which must be set by the config and
350 others are optional or predefined:
352 **name** *[required]* The name of the test suite, for use in reports and
355 **test_format** *[required]* The test format object which will be used to
356 discover and run tests in the test suite. Generally this will be a builtin test
357 format available from the *lit.formats* module.
359 **test_src_root** The filesystem path to the test suite root. For out-of-dir
360 builds this is the directory that will be scanned for tests.
362 **test_exec_root** For out-of-dir builds, the path to the test suite root inside
363 the object directory. This is where tests will be run and temporary output files
366 **environment** A dictionary representing the environment to use when executing
369 **suffixes** For **lit** test formats which scan directories for tests, this
370 variable is a list of suffixes to identify test files. Used by: *ShTest*,
373 **substitutions** For **lit** test formats which substitute variables into a test
374 script, the list of substitutions to perform. Used by: *ShTest*, *TclTest*.
376 **unsupported** Mark an unsupported directory, all tests within it will be
377 reported as unsupported. Used by: *ShTest*, *TclTest*.
379 **parent** The parent configuration, this is the config object for the directory
380 containing the test suite, or None.
382 **root** The root configuration. This is the top-most **lit** configuration in
385 **on_clone** The config is actually cloned for every subdirectory inside a test
386 suite, to allow local configuration on a per-directory basis. The *on_clone*
387 variable can be set to a Python function which will be called whenever a
388 configuration is cloned (for a subdirectory). The function should takes three
389 arguments: (1) the parent configuration, (2) the new configuration (which the
390 *on_clone* function will generally modify), and (3) the test path to the new
391 directory being scanned.
400 Once test suites are located, **lit** recursively traverses the source directory
401 (following *test_src_root*) looking for tests. When **lit** enters a
402 sub-directory, it first checks to see if a nested test suite is defined in that
403 directory. If so, it loads that test suite recursively, otherwise it
404 instantiates a local test config for the directory (see "LOCAL CONFIGURATION
407 Tests are identified by the test suite they are contained within, and the
408 relative path inside that suite. Note that the relative path may not refer to an
409 actual file on disk; some test formats (such as *GoogleTest*) define "virtual
410 tests" which have a path that contains both the path to the actual test file and
411 a subpath to identify the virtual test.
414 LOCAL CONFIGURATION FILES
415 ~~~~~~~~~~~~~~~~~~~~~~~~~
418 When **lit** loads a subdirectory in a test suite, it instantiates a local test
419 configuration by cloning the configuration for the parent direction -- the root
420 of this configuration chain will always be a test suite. Once the test
421 configuration is cloned **lit** checks for a *lit.local.cfg* file in the
422 subdirectory. If present, this file will be loaded and can be used to specialize
423 the configuration for each individual directory. This facility can be used to
424 define subdirectories of optional tests, or to change other configuration
425 parameters -- for example, to change the test format, or the suffixes which
429 TEST RUN OUTPUT FORMAT
430 ~~~~~~~~~~~~~~~~~~~~~~
433 The **lit** output for a test run conforms to the following schema, in both
434 short and verbose modes (although in short mode no PASS lines will be shown).
435 This schema has been chosen to be relatively easy to reliably parse by a machine
436 (for example in buildbot log scraping), and for other tools to generate.
438 Each test result is expected to appear on a line that matches::
440 <result code>: <test name> (<progress info>)
442 where <result-code> is a standard test result such as PASS, FAIL, XFAIL, XPASS,
443 UNRESOLVED, or UNSUPPORTED. The performance result codes of IMPROVED and
444 REGRESSED are also allowed.
446 The <test name> field can consist of an arbitrary string containing no newline.
448 The <progress info> field can be used to report progress information such as
449 (1/300) or can be empty, but even when empty the parentheses are required.
451 Each test result may include additional (multiline) log information in the
454 <log delineator> TEST '(<test name>)' <trailing delineator>
458 where <test name> should be the name of a preceding reported test, <log
459 delineator> is a string of '\*' characters *at least* four characters long (the
460 recommended length is 20), and <trailing delineator> is an arbitrary (unparsed)
463 The following is an example of a test run output which consists of four tests A,
464 B, C, and D, and a log message for the failing test C::
469 \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* TEST 'C' FAILED \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*
470 Test 'C' failed as a result of exit code 1.
471 \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*
479 The **lit** distribution contains several example implementations of test suites
480 in the *ExampleTests* directory.