1lit - LLVM Integrated Tester 2============================ 3 4 5SYNOPSIS 6-------- 7 8 9**lit** [*options*] [*tests*] 10 11 12DESCRIPTION 13----------- 14 15 16**lit** is a portable tool for executing LLVM and Clang style test suites, 17summarizing their results, and providing indication of failures. **lit** is 18designed to be a lightweight testing tool with as simple a user interface as 19possible. 20 21**lit** should be run with one or more *tests* to run specified on the command 22line. Tests can be either individual test files or directories to search for 23tests (see "TEST DISCOVERY"). 24 25Each specified test will be executed (potentially in parallel) and once all 26tests have been run **lit** will print summary information on the number of tests 27which passed or failed (see "TEST STATUS RESULTS"). The **lit** program will 28execute with a non-zero exit code if any tests fail. 29 30By default **lit** will use a succinct progress display and will only print 31summary information for test failures. See "OUTPUT OPTIONS" for options 32controlling the **lit** progress display and output. 33 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 36OPTIONS" for more information. 37 38Finally, **lit** also supports additional options for only running a subset of 39the options specified on the command line, see "SELECTION OPTIONS" for 40more information. 41 42Users interested in the **lit** architecture or designing a **lit** testing 43implementation should see "LIT INFRASTRUCTURE" 44 45 46GENERAL OPTIONS 47--------------- 48 49 50 51**-h**, **--help** 52 53 Show the **lit** help message. 54 55 56 57**-j** *N*, **--threads**\ =\ *N* 58 59 Run *N* tests in parallel. By default, this is automatically chosen to match 60 the number of detected available CPUs. 61 62 63 64**--config-prefix**\ =\ *NAME* 65 66 Search for *NAME.cfg* and *NAME.site.cfg* when searching for test suites, 67 instead of *lit.cfg* and *lit.site.cfg*. 68 69 70 71**--param** *NAME*, **--param** *NAME*\ =\ *VALUE* 72 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 75 dependent. 76 77 78 79 80OUTPUT OPTIONS 81-------------- 82 83 84 85**-q**, **--quiet** 86 87 Suppress any output except for test failures. 88 89 90 91**-s**, **--succinct** 92 93 Show less output, for example don't show information on tests that pass. 94 95 96 97**-v**, **--verbose** 98 99 Show more information on test failures, for example the entire test output 100 instead of just the test result. 101 102 103 104**--no-progress-bar** 105 106 Do not use curses based progress bar. 107 108 109 110 111EXECUTION OPTIONS 112----------------- 113 114 115 116**--path**\ =\ *PATH* 117 118 Specify an addition *PATH* to use when searching for executables in tests. 119 120 121 122**--vg** 123 124 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. 127 128 129 130**--vg-arg**\ =\ *ARG* 131 132 When *--vg* is used, specify an additional argument to pass to valgrind itself. 133 134 135 136**--time-tests** 137 138 Track the wall time individual tests take to execute and includes the results in 139 the summary output. This is useful for determining which tests in a test suite 140 take the most time to execute. Note that this option is most useful with *-j 141 1*. 142 143 144 145 146SELECTION OPTIONS 147----------------- 148 149 150 151**--max-tests**\ =\ *N* 152 153 Run at most *N* tests and then terminate. 154 155 156 157**--max-time**\ =\ *N* 158 159 Spend at most *N* seconds (approximately) running tests and then terminate. 160 161 162 163**--shuffle** 164 165 Run the tests in a random order. 166 167 168 169 170ADDITIONAL OPTIONS 171------------------ 172 173 174 175**--debug** 176 177 Run **lit** in debug mode, for debugging configuration issues and **lit** itself. 178 179 180 181**--show-suites** 182 183 List the discovered test suites as part of the standard output. 184 185 186 187**--no-tcl-as-sh** 188 189 Run Tcl scripts internally (instead of converting to shell scripts). 190 191 192 193**--repeat**\ =\ *N* 194 195 Run each test *N* times. Currently this is primarily useful for timing tests, 196 other results are not collated in any reasonable fashion. 197 198 199 200 201EXIT STATUS 202----------- 203 204 205**lit** will exit with an exit code of 1 if there are any FAIL or XPASS 206results. Otherwise, it will exit with the status 0. Other exit codes are used 207for non-test related failures (for example a user error or an internal program 208error). 209 210 211TEST DISCOVERY 212-------------- 213 214 215The inputs passed to **lit** can be either individual tests, or entire 216directories or hierarchies of tests to run. When **lit** starts up, the first 217thing it does is convert the inputs into a complete list of tests to run as part 218of *test discovery*. 219 220In the **lit** model, every test must exist inside some *test suite*. **lit** 221resolves the inputs specified on the command line to test suites by searching 222upwards from the input path until it finds a *lit.cfg* or *lit.site.cfg* 223file. These files serve as both a marker of test suites and as configuration 224files which **lit** loads in order to understand how to find and run the tests 225inside the test suite. 226 227Once **lit** has mapped the inputs into test suites it traverses the list of 228inputs adding tests for individual files and recursively searching for tests in 229directories. 230 231This behavior makes it easy to specify a subset of tests to run, while still 232allowing the test suite configuration to control exactly how tests are 233interpreted. In addition, **lit** always identifies tests by the test suite they 234are in, and their relative path inside the test suite. For appropriately 235configured projects, this allows **lit** to provide convenient and flexible 236support for out-of-tree builds. 237 238 239TEST STATUS RESULTS 240------------------- 241 242 243Each test ultimately produces one of the following six results: 244 245 246**PASS** 247 248 The test succeeded. 249 250 251 252**XFAIL** 253 254 The test failed, but that is expected. This is used for test formats which allow 255 specifying that a test does not currently work, but wish to leave it in the test 256 suite. 257 258 259 260**XPASS** 261 262 The test succeeded, but it was expected to fail. This is used for tests which 263 were specified as expected to fail, but are now succeeding (generally because 264 the feature they test was broken and has been fixed). 265 266 267 268**FAIL** 269 270 The test failed. 271 272 273 274**UNRESOLVED** 275 276 The test result could not be determined. For example, this occurs when the test 277 could not be run, the test itself is invalid, or the test was interrupted. 278 279 280 281**UNSUPPORTED** 282 283 The test is not supported in this environment. This is used by test formats 284 which can report unsupported tests. 285 286 287 288Depending on the test format tests may produce additional information about 289their status (generally only for failures). See the Output|"OUTPUT OPTIONS" 290section for more information. 291 292 293LIT INFRASTRUCTURE 294------------------ 295 296 297This section describes the **lit** testing architecture for users interested in 298creating a new **lit** testing implementation, or extending an existing one. 299 300**lit** proper is primarily an infrastructure for discovering and running 301arbitrary tests, and to expose a single convenient interface to these 302tests. **lit** itself doesn't know how to run tests, rather this logic is 303defined by *test suites*. 304 305TEST SUITES 306~~~~~~~~~~~ 307 308 309As described in "TEST DISCOVERY", tests are always located inside a *test 310suite*. Test suites serve to define the format of the tests they contain, the 311logic for finding those tests, and any additional information to run the tests. 312 313**lit** identifies test suites as directories containing *lit.cfg* or 314*lit.site.cfg* files (see also **--config-prefix**). Test suites are initially 315discovered by recursively searching up the directory hierarchy for all the input 316files passed on the command line. You can use **--show-suites** to display the 317discovered test suites at startup. 318 319Once a test suite is discovered, its config file is loaded. Config files 320themselves are Python modules which will be executed. When the config file is 321executed, two important global variables are predefined: 322 323 324**lit** 325 326 The global **lit** configuration object (a *LitConfig* instance), which defines 327 the builtin test formats, global configuration parameters, and other helper 328 routines for implementing test configurations. 329 330 331 332**config** 333 334 This is the config object (a *TestingConfig* instance) for the test suite, 335 which the config file is expected to populate. The following variables are also 336 available on the *config* object, some of which must be set by the config and 337 others are optional or predefined: 338 339 **name** *[required]* The name of the test suite, for use in reports and 340 diagnostics. 341 342 **test_format** *[required]* The test format object which will be used to 343 discover and run tests in the test suite. Generally this will be a builtin test 344 format available from the *lit.formats* module. 345 346 **test_src_root** The filesystem path to the test suite root. For out-of-dir 347 builds this is the directory that will be scanned for tests. 348 349 **test_exec_root** For out-of-dir builds, the path to the test suite root inside 350 the object directory. This is where tests will be run and temporary output files 351 placed. 352 353 **environment** A dictionary representing the environment to use when executing 354 tests in the suite. 355 356 **suffixes** For **lit** test formats which scan directories for tests, this 357 variable is a list of suffixes to identify test files. Used by: *ShTest*, 358 *TclTest*. 359 360 **substitutions** For **lit** test formats which substitute variables into a test 361 script, the list of substitutions to perform. Used by: *ShTest*, *TclTest*. 362 363 **unsupported** Mark an unsupported directory, all tests within it will be 364 reported as unsupported. Used by: *ShTest*, *TclTest*. 365 366 **parent** The parent configuration, this is the config object for the directory 367 containing the test suite, or None. 368 369 **root** The root configuration. This is the top-most **lit** configuration in 370 the project. 371 372 **on_clone** The config is actually cloned for every subdirectory inside a test 373 suite, to allow local configuration on a per-directory basis. The *on_clone* 374 variable can be set to a Python function which will be called whenever a 375 configuration is cloned (for a subdirectory). The function should takes three 376 arguments: (1) the parent configuration, (2) the new configuration (which the 377 *on_clone* function will generally modify), and (3) the test path to the new 378 directory being scanned. 379 380 381 382 383TEST DISCOVERY 384~~~~~~~~~~~~~~ 385 386 387Once test suites are located, **lit** recursively traverses the source directory 388(following *test_src_root*) looking for tests. When **lit** enters a 389sub-directory, it first checks to see if a nested test suite is defined in that 390directory. If so, it loads that test suite recursively, otherwise it 391instantiates a local test config for the directory (see "LOCAL CONFIGURATION 392FILES"). 393 394Tests are identified by the test suite they are contained within, and the 395relative path inside that suite. Note that the relative path may not refer to an 396actual file on disk; some test formats (such as *GoogleTest*) define "virtual 397tests" which have a path that contains both the path to the actual test file and 398a subpath to identify the virtual test. 399 400 401LOCAL CONFIGURATION FILES 402~~~~~~~~~~~~~~~~~~~~~~~~~ 403 404 405When **lit** loads a subdirectory in a test suite, it instantiates a local test 406configuration by cloning the configuration for the parent direction -- the root 407of this configuration chain will always be a test suite. Once the test 408configuration is cloned **lit** checks for a *lit.local.cfg* file in the 409subdirectory. If present, this file will be loaded and can be used to specialize 410the configuration for each individual directory. This facility can be used to 411define subdirectories of optional tests, or to change other configuration 412parameters -- for example, to change the test format, or the suffixes which 413identify test files. 414 415 416TEST RUN OUTPUT FORMAT 417~~~~~~~~~~~~~~~~~~~~~~ 418 419 420The b<lit> output for a test run conforms to the following schema, in both short 421and verbose modes (although in short mode no PASS lines will be shown). This 422schema has been chosen to be relatively easy to reliably parse by a machine (for 423example in buildbot log scraping), and for other tools to generate. 424 425Each test result is expected to appear on a line that matches: 426 427<result code>: <test name> (<progress info>) 428 429where <result-code> is a standard test result such as PASS, FAIL, XFAIL, XPASS, 430UNRESOLVED, or UNSUPPORTED. The performance result codes of IMPROVED and 431REGRESSED are also allowed. 432 433The <test name> field can consist of an arbitrary string containing no newline. 434 435The <progress info> field can be used to report progress information such as 436(1/300) or can be empty, but even when empty the parentheses are required. 437 438Each test result may include additional (multiline) log information in the 439following format. 440 441<log delineator> TEST '(<test name>)' <trailing delineator> 442... log message ... 443<log delineator> 444 445where <test name> should be the name of a preceding reported test, <log 446delineator> is a string of '\*' characters *at least* four characters long (the 447recommended length is 20), and <trailing delineator> is an arbitrary (unparsed) 448string. 449 450The following is an example of a test run output which consists of four tests A, 451B, C, and D, and a log message for the failing test C:: 452 453 PASS: A (1 of 4) 454 PASS: B (2 of 4) 455 FAIL: C (3 of 4) 456 \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* TEST 'C' FAILED \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* 457 Test 'C' failed as a result of exit code 1. 458 \*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\*\* 459 PASS: D (4 of 4) 460 461 462LIT EXAMPLE TESTS 463~~~~~~~~~~~~~~~~~ 464 465 466The **lit** distribution contains several example implementations of test suites 467in the *ExampleTests* directory. 468 469 470SEE ALSO 471-------- 472 473 474valgrind(1) 475