1% Testing the JDK 2 3## Using "make test" (the run-test framework) 4 5This new way of running tests is developer-centric. It assumes that you have 6built a JDK locally and want to test it. Running common test targets is simple, 7and more complex ad-hoc combination of tests is possible. The user interface is 8forgiving, and clearly report errors it cannot resolve. 9 10The main target `test` uses the jdk-image as the tested product. There is 11also an alternate target `exploded-test` that uses the exploded image 12instead. Not all tests will run successfully on the exploded image, but using 13this target can greatly improve rebuild times for certain workflows. 14 15Previously, `make test` was used to invoke an old system for running tests, and 16`make run-test` was used for the new test framework. For backward compatibility 17with scripts and muscle memory, `run-test` (and variants like 18`exploded-run-test` or `run-test-tier1`) are kept as aliases. 19 20Some example command-lines: 21 22 $ make test-tier1 23 $ make test-jdk_lang JTREG="JOBS=8" 24 $ make test TEST=jdk_lang 25 $ make test-only TEST="gtest:LogTagSet gtest:LogTagSetDescriptions" GTEST="REPEAT=-1" 26 $ make test TEST="hotspot:hotspot_gc" JTREG="JOBS=1;TIMEOUT_FACTOR=8;JAVA_OPTIONS=-XshowSettings -Xlog:gc+ref=debug" 27 $ make test TEST="jtreg:test/hotspot:hotspot_gc test/hotspot/jtreg/native_sanity/JniVersion.java" 28 $ make test TEST="micro:java.lang.reflect" MICRO="FORK=1;WARMUP_ITER=2" 29 $ make exploded-test TEST=tier2 30 31### Configuration 32 33To be able to run JTReg tests, `configure` needs to know where to find the 34JTReg test framework. If it is not picked up automatically by configure, use 35the `--with-jtreg=<path to jtreg home>` option to point to the JTReg framework. 36Note that this option should point to the JTReg home, i.e. the top directory, 37containing `lib/jtreg.jar` etc. (An alternative is to set the `JT_HOME` 38environment variable to point to the JTReg home before running `configure`.) 39 40To be able to run microbenchmarks, `configure` needs to know where to find the 41JMH dependency. Use `--with-jmh=<path to JMH jars>` to point to a directory 42containing the core JMH and transitive dependencies. The recommended 43dependencies can be retrieved by running `sh make/devkit/createJMHBundle.sh`, 44after which `--with-jmh=build/jmh/jars` should work. 45 46## Test selection 47 48All functionality is available using the `test` make target. In this use case, 49the test or tests to be executed is controlled using the `TEST` variable. To 50speed up subsequent test runs with no source code changes, `test-only` can be 51used instead, which do not depend on the source and test image build. 52 53For some common top-level tests, direct make targets have been generated. This 54includes all JTReg test groups, the hotspot gtest, and custom tests (if 55present). This means that `make test-tier1` is equivalent to `make test 56TEST="tier1"`, but the latter is more tab-completion friendly. For more complex 57test runs, the `test TEST="x"` solution needs to be used. 58 59The test specifications given in `TEST` is parsed into fully qualified test 60descriptors, which clearly and unambigously show which tests will be run. As an 61example, `:tier1` will expand to `jtreg:$(TOPDIR)/test/hotspot/jtreg:tier1 62jtreg:$(TOPDIR)/test/jdk:tier1 jtreg:$(TOPDIR)/test/langtools:tier1 63jtreg:$(TOPDIR)/test/nashorn:tier1 jtreg:$(TOPDIR)/test/jaxp:tier1`. You can 64always submit a list of fully qualified test descriptors in the `TEST` variable 65if you want to shortcut the parser. 66 67### Common Test Groups 68 69Ideally, all tests are run for every change but this may not be practical due to the limited 70testing resources, the scope of the change, etc. 71 72The source tree currently defines a few common test groups in the relevant `TEST.groups` 73files. There are test groups that cover a specific component, for example `hotspot_gc`. 74It is a good idea to look into `TEST.groups` files to get a sense what tests are relevant 75to a particular JDK component. 76 77Component-specific tests may miss some unintended consequences of a change, so other 78tests should also be run. Again, it might be impractical to run all tests, and therefore 79_tiered_ test groups exist. Tiered test groups are not component-specific, but rather cover 80the significant parts of the entire JDK. 81 82Multiple tiers allow balancing test coverage and testing costs. Lower test tiers are supposed to 83contain the simpler, quicker and more stable tests. Higher tiers are supposed to contain 84progressively more thorough, slower, and sometimes less stable tests, or the tests that require 85special configuration. 86 87Contributors are expected to run the tests for the areas that are changed, and the first N tiers 88they can afford to run, but at least tier1. 89 90A brief description of the tiered test groups: 91 92- `tier1`: This is the lowest test tier. Multiple developers run these tests every day. 93Because of the widespread use, the tests in `tier1` are carefully selected and optimized to run 94fast, and to run in the most stable manner. The test failures in `tier1` are usually followed up 95on quickly, either with fixes, or adding relevant tests to problem list. GitHub Actions workflows, 96if enabled, run `tier1` tests. 97 98- `tier2`: This test group covers even more ground. These contain, among other things, 99tests that either run for too long to be at `tier1`, or may require special configuration, 100or tests that are less stable, or cover the broader range of non-core JVM and JDK features/components 101(for example, XML). 102 103- `tier3`: This test group includes more stressful tests, the tests for corner cases 104not covered by previous tiers, plus the tests that require GUIs. As such, this suite 105should either be run with low concurrency (`TEST_JOBS=1`), or without headful tests 106(`JTREG_KEYWORDS=\!headful`), or both. 107 108- `tier4`: This test group includes every other test not covered by previous tiers. It includes, 109for example, `vmTestbase` suites for Hotspot, which run for many hours even on large 110machines. It also runs GUI tests, so the same `TEST_JOBS` and `JTREG_KEYWORDS` caveats 111apply. 112 113### JTReg 114 115JTReg tests can be selected either by picking a JTReg test group, or a selection 116of files or directories containing JTReg tests. 117 118JTReg test groups can be specified either without a test root, e.g. `:tier1` 119(or `tier1`, the initial colon is optional), or with, e.g. `hotspot:tier1`, 120`test/jdk:jdk_util` or `$(TOPDIR)/test/hotspot/jtreg:hotspot_all`. The test 121root can be specified either as an absolute path, or a path relative to the 122JDK top directory, or the `test` directory. For simplicity, the hotspot 123JTReg test root, which really is `hotspot/jtreg` can be abbreviated as 124just `hotspot`. 125 126When specified without a test root, all matching groups from all test roots 127will be added. Otherwise, only the group from the specified test root will be 128added. 129 130Individual JTReg tests or directories containing JTReg tests can also be 131specified, like `test/hotspot/jtreg/native_sanity/JniVersion.java` or 132`hotspot/jtreg/native_sanity`. Just like for test root selection, you can 133either specify an absolute path (which can even point to JTReg tests outside 134the source tree), or a path relative to either the JDK top directory or the 135`test` directory. `hotspot` can be used as an alias for `hotspot/jtreg` here as 136well. 137 138As long as the test groups or test paths can be uniquely resolved, you do not 139need to enter the `jtreg:` prefix. If this is not possible, or if you want to 140use a fully qualified test descriptor, add `jtreg:`, e.g. 141`jtreg:test/hotspot/jtreg/native_sanity`. 142 143### Gtest 144 145Since the Hotspot Gtest suite is so quick, the default is to run all tests. 146This is specified by just `gtest`, or as a fully qualified test descriptor 147`gtest:all`. 148 149If you want, you can single out an individual test or a group of tests, for 150instance `gtest:LogDecorations` or `gtest:LogDecorations.level_test_vm`. This 151can be particularly useful if you want to run a shaky test repeatedly. 152 153For Gtest, there is a separate test suite for each JVM variant. The JVM variant 154is defined by adding `/<variant>` to the test descriptor, e.g. 155`gtest:Log/client`. If you specify no variant, gtest will run once for each JVM 156variant present (e.g. server, client). So if you only have the server JVM 157present, then `gtest:all` will be equivalent to `gtest:all/server`. 158 159### Microbenchmarks 160 161Which microbenchmarks to run is selected using a regular expression 162following the `micro:` test descriptor, e.g., `micro:java.lang.reflect`. This 163delegates the test selection to JMH, meaning package name, class name and even 164benchmark method names can be used to select tests. 165 166Using special characters like `|` in the regular expression is possible, but 167needs to be escaped multiple times: `micro:ArrayCopy\\\\\|reflect`. 168 169### Special tests 170 171A handful of odd tests that are not covered by any other testing framework are 172accessible using the `special:` test descriptor. Currently, this includes 173`failure-handler` and `make`. 174 175 * Failure handler testing is run using `special:failure-handler` or just 176 `failure-handler` as test descriptor. 177 178 * Tests for the build system, including both makefiles and related 179 functionality, is run using `special:make` or just `make` as test 180 descriptor. This is equivalent to `special:make:all`. 181 182 A specific make test can be run by supplying it as argument, e.g. 183 `special:make:idea`. As a special syntax, this can also be expressed as 184 `make-idea`, which allows for command lines as `make test-make-idea`. 185 186## Test results and summary 187 188At the end of the test run, a summary of all tests run will be presented. This 189will have a consistent look, regardless of what test suites were used. This is 190a sample summary: 191 192 ============================== 193 Test summary 194 ============================== 195 TEST TOTAL PASS FAIL ERROR 196 >> jtreg:jdk/test:tier1 1867 1865 2 0 << 197 jtreg:langtools/test:tier1 4711 4711 0 0 198 jtreg:nashorn/test:tier1 133 133 0 0 199 ============================== 200 TEST FAILURE 201 202Tests where the number of TOTAL tests does not equal the number of PASSed tests 203will be considered a test failure. These are marked with the `>> ... <<` marker 204for easy identification. 205 206The classification of non-passed tests differs a bit between test suites. In 207the summary, ERROR is used as a catch-all for tests that neither passed nor are 208classified as failed by the framework. This might indicate test framework 209error, timeout or other problems. 210 211In case of test failures, `make test` will exit with a non-zero exit value. 212 213All tests have their result stored in `build/$BUILD/test-results/$TEST_ID`, 214where TEST_ID is a path-safe conversion from the fully qualified test 215descriptor, e.g. for `jtreg:jdk/test:tier1` the TEST_ID is 216`jtreg_jdk_test_tier1`. This path is also printed in the log at the end of the 217test run. 218 219Additional work data is stored in `build/$BUILD/test-support/$TEST_ID`. For 220some frameworks, this directory might contain information that is useful in 221determining the cause of a failed test. 222 223## Test suite control 224 225It is possible to control various aspects of the test suites using make control 226variables. 227 228These variables use a keyword=value approach to allow multiple values to be 229set. So, for instance, `JTREG="JOBS=1;TIMEOUT_FACTOR=8"` will set the JTReg 230concurrency level to 1 and the timeout factor to 8. This is equivalent to 231setting `JTREG_JOBS=1 JTREG_TIMEOUT_FACTOR=8`, but using the keyword format 232means that the `JTREG` variable is parsed and verified for correctness, so 233`JTREG="TMIEOUT_FACTOR=8"` would give an error, while `JTREG_TMIEOUT_FACTOR=8` 234would just pass unnoticed. 235 236To separate multiple keyword=value pairs, use `;` (semicolon). Since the shell 237normally eats `;`, the recommended usage is to write the assignment inside 238qoutes, e.g. `JTREG="...;..."`. This will also make sure spaces are preserved, 239as in `JTREG="JAVA_OPTIONS=-XshowSettings -Xlog:gc+ref=debug"`. 240 241(Other ways are possible, e.g. using backslash: `JTREG=JOBS=1\;TIMEOUT_FACTOR=8`. 242Also, as a special technique, the string `%20` will be replaced with space for 243certain options, e.g. `JTREG=JAVA_OPTIONS=-XshowSettings%20-Xlog:gc+ref=debug`. 244This can be useful if you have layers of scripts and have trouble getting 245proper quoting of command line arguments through.) 246 247As far as possible, the names of the keywords have been standardized between 248test suites. 249 250### General keywords (TEST_OPTS) 251 252Some keywords are valid across different test suites. If you want to run tests 253from multiple test suites, or just don't want to care which test suite specific 254control variable to use, then you can use the general TEST_OPTS control 255variable. 256 257There are also some keywords that applies globally to the test runner system, 258not to any specific test suites. These are also available as TEST_OPTS keywords. 259 260#### JOBS 261 262Currently only applies to JTReg. 263 264#### TIMEOUT_FACTOR 265 266Currently only applies to JTReg. 267 268#### JAVA_OPTIONS 269 270Applies to JTReg, GTest and Micro. 271 272#### VM_OPTIONS 273 274Applies to JTReg, GTest and Micro. 275 276#### AOT_MODULES 277 278Applies to JTReg and GTest. 279 280#### JCOV 281 282This keywords applies globally to the test runner system. If set to `true`, it 283enables JCov coverage reporting for all tests run. To be useful, the JDK under 284test must be run with a JDK built with JCov instrumentation (`configure 285--with-jcov=<path to directory containing lib/jcov.jar>`, `make jcov-image`). 286 287The simplest way to run tests with JCov coverage report is to use the special 288target `jcov-test` instead of `test`, e.g. `make jcov-test TEST=jdk_lang`. This 289will make sure the JCov image is built, and that JCov reporting is enabled. 290 291The JCov report is stored in `build/$BUILD/test-results/jcov-output/report`. 292 293Please note that running with JCov reporting can be very memory intensive. 294 295#### JCOV_DIFF_CHANGESET 296 297While collecting code coverage with JCov, it is also possible to find coverage 298for only recently changed code. JCOV_DIFF_CHANGESET specifies a source 299revision. A textual report will be generated showing coverage of the diff 300between the specified revision and the repository tip. 301 302The report is stored in 303`build/$BUILD/test-results/jcov-output/diff_coverage_report` file. 304 305### JTReg keywords 306 307#### JOBS 308 309The test concurrency (`-concurrency`). 310 311Defaults to TEST_JOBS (if set by `--with-test-jobs=`), otherwise it defaults to 312JOBS, except for Hotspot, where the default is *number of CPU cores/2*, 313but never more than *memory size in GB/2*. 314 315#### TIMEOUT_FACTOR 316 317The timeout factor (`-timeoutFactor`). 318 319Defaults to 4. 320 321#### FAILURE_HANDLER_TIMEOUT 322 323Sets the argument `-timeoutHandlerTimeout` for JTReg. The default value is 0. 324This is only valid if the failure handler is built. 325 326#### TEST_MODE 327 328The test mode (`agentvm` or `othervm`). 329 330Defaults to `agentvm`. 331 332#### ASSERT 333 334Enable asserts (`-ea -esa`, or none). 335 336Set to `true` or `false`. If true, adds `-ea -esa`. Defaults to true, except 337for hotspot. 338 339#### VERBOSE 340 341The verbosity level (`-verbose`). 342 343Defaults to `fail,error,summary`. 344 345#### RETAIN 346 347What test data to retain (`-retain`). 348 349Defaults to `fail,error`. 350 351#### MAX_MEM 352 353Limit memory consumption (`-Xmx` and `-vmoption:-Xmx`, or none). 354 355Limit memory consumption for JTReg test framework and VM under test. Set to 0 356to disable the limits. 357 358Defaults to 512m, except for hotspot, where it defaults to 0 (no limit). 359 360#### MAX_OUTPUT 361 362Set the property `javatest.maxOutputSize` for the launcher, to change the 363default JTReg log limit. 364 365#### KEYWORDS 366 367JTReg keywords sent to JTReg using `-k`. Please be careful in making sure that 368spaces and special characters (like `!`) are properly quoted. To avoid some 369issues, the special value `%20` can be used instead of space. 370 371#### EXTRA_PROBLEM_LISTS 372 373Use additional problem lists file or files, in addition to the default 374ProblemList.txt located at the JTReg test roots. 375 376If multiple file names are specified, they should be separated by space (or, to 377help avoid quoting issues, the special value `%20`). 378 379The file names should be either absolute, or relative to the JTReg test root of 380the tests to be run. 381 382#### RUN_PROBLEM_LISTS 383 384Use the problem lists to select tests instead of excluding them. 385 386Set to `true` or `false`. 387If `true`, JTReg will use `-match:` option, otherwise `-exclude:` will be used. 388Default is `false`. 389 390#### OPTIONS 391 392Additional options to the JTReg test framework. 393 394Use `JTREG="OPTIONS=--help all"` to see all available JTReg options. 395 396#### JAVA_OPTIONS 397 398Additional Java options for running test classes (sent to JTReg as 399`-javaoption`). 400 401#### VM_OPTIONS 402 403Additional Java options to be used when compiling and running classes (sent to 404JTReg as `-vmoption`). 405 406This option is only needed in special circumstances. To pass Java options to 407your test classes, use `JAVA_OPTIONS`. 408 409#### LAUNCHER_OPTIONS 410 411Additional Java options that are sent to the java launcher that starts the 412JTReg harness. 413 414#### AOT_MODULES 415 416Generate AOT modules before testing for the specified module, or set of 417modules. If multiple modules are specified, they should be separated by space 418(or, to help avoid quoting issues, the special value `%20`). 419 420#### RETRY_COUNT 421 422Retry failed tests up to a set number of times. Defaults to 0. 423 424### Gtest keywords 425 426#### REPEAT 427 428The number of times to repeat the tests (`--gtest_repeat`). 429 430Default is 1. Set to -1 to repeat indefinitely. This can be especially useful 431combined with `OPTIONS=--gtest_break_on_failure` to reproduce an intermittent 432problem. 433 434#### OPTIONS 435 436Additional options to the Gtest test framework. 437 438Use `GTEST="OPTIONS=--help"` to see all available Gtest options. 439 440#### AOT_MODULES 441 442Generate AOT modules before testing for the specified module, or set of 443modules. If multiple modules are specified, they should be separated by space 444(or, to help avoid quoting issues, the special value `%20`). 445 446### Microbenchmark keywords 447 448#### FORK 449 450Override the number of benchmark forks to spawn. Same as specifying `-f <num>`. 451 452#### ITER 453 454Number of measurement iterations per fork. Same as specifying `-i <num>`. 455 456#### TIME 457 458Amount of time to spend in each measurement iteration, in seconds. Same as 459specifying `-r <num>` 460 461#### WARMUP_ITER 462 463Number of warmup iterations to run before the measurement phase in each fork. 464Same as specifying `-wi <num>`. 465 466#### WARMUP_TIME 467 468Amount of time to spend in each warmup iteration. Same as specifying `-w <num>`. 469 470#### RESULTS_FORMAT 471 472Specify to have the test run save a log of the values. Accepts the same values 473as `-rff`, i.e., `text`, `csv`, `scsv`, `json`, or `latex`. 474 475#### VM_OPTIONS 476 477Additional VM arguments to provide to forked off VMs. Same as `-jvmArgs <args>` 478 479#### OPTIONS 480 481Additional arguments to send to JMH. 482 483## Notes for Specific Tests 484 485### Docker Tests 486 487Docker tests with default parameters may fail on systems with glibc versions 488not compatible with the one used in the default docker image (e.g., Oracle 489Linux 7.6 for x86). For example, they pass on Ubuntu 16.04 but fail on Ubuntu 49018.04 if run like this on x86: 491 492``` 493$ make test TEST="jtreg:test/hotspot/jtreg/containers/docker" 494``` 495 496To run these tests correctly, additional parameters for the correct docker 497image are required on Ubuntu 18.04 by using `JAVA_OPTIONS`. 498 499``` 500$ make test TEST="jtreg:test/hotspot/jtreg/containers/docker" \ 501 JTREG="JAVA_OPTIONS=-Djdk.test.docker.image.name=ubuntu 502 -Djdk.test.docker.image.version=latest" 503``` 504 505### Non-US locale 506 507If your locale is non-US, some tests are likely to fail. To work around this 508you can set the locale to US. On Unix platforms simply setting `LANG="en_US"` 509in the environment before running tests should work. On Windows, setting 510`JTREG="VM_OPTIONS=-Duser.language=en -Duser.country=US"` helps for most, but 511not all test cases. 512 513For example: 514 515``` 516$ export LANG="en_US" && make test TEST=... 517$ make test JTREG="VM_OPTIONS=-Duser.language=en -Duser.country=US" TEST=... 518``` 519 520### PKCS11 Tests 521 522It is highly recommended to use the latest NSS version when running PKCS11 523tests. Improper NSS version may lead to unexpected failures which are hard to 524diagnose. For example, sun/security/pkcs11/Secmod/AddTrustedCert.java may fail 525on Ubuntu 18.04 with the default NSS version in the system. To run these tests 526correctly, the system property `test.nss.lib.paths` is required on Ubuntu 18.04 527to specify the alternative NSS lib directories. 528 529For example: 530 531``` 532$ make test TEST="jtreg:sun/security/pkcs11/Secmod/AddTrustedCert.java" \ 533 JTREG="JAVA_OPTIONS=-Dtest.nss.lib.paths=/path/to/your/latest/NSS-libs" 534``` 535 536For more notes about the PKCS11 tests, please refer to 537test/jdk/sun/security/pkcs11/README. 538 539### Client UI Tests 540 541Some Client UI tests use key sequences which may be reserved by the operating 542system. Usually that causes the test failure. So it is highly recommended to 543disable system key shortcuts prior testing. The steps to access and disable 544system key shortcuts for various platforms are provided below. 545 546#### MacOS 547 548Choose Apple menu; System Preferences, click Keyboard, then click Shortcuts; 549select or deselect desired shortcut. 550 551For example, 552test/jdk/javax/swing/TooltipManager/JMenuItemToolTipKeyBindingsTest/JMenuItemToolTipKeyBindingsTest.java 553fails on MacOS because it uses `CTRL + F1` key sequence to show or hide tooltip 554message but the key combination is reserved by the operating system. To run the 555test correctly the default global key shortcut should be disabled using the 556steps described above, and then deselect "Turn keyboard access on or off" 557option which is responsible for `CTRL + F1` combination. 558 559#### Linux 560 561Open the Activities overview and start typing Settings; Choose Settings, click 562Devices, then click Keyboard; set or override desired shortcut. 563 564#### Windows 565 566Type `gpedit` in the Search and then click Edit group policy; navigate to User 567Configuration -> Administrative Templates -> Windows Components -> File 568Explorer; in the right-side pane look for "Turn off Windows key hotkeys" and 569double click on it; enable or disable hotkeys. 570 571Note: restart is required to make the settings take effect. 572 573--- 574# Override some definitions in the global css file that are not optimal for 575# this document. 576header-includes: 577 - '<style type="text/css">pre, code, tt { color: #1d6ae5; }</style>' 578--- 579