1 ============================== 2 LLVM test-suite Makefile Guide 3 ============================== 4 5 .. contents:: 6 :local: 7 8 Overview 9 ======== 10 11 This document describes the features of the Makefile-based LLVM 12 test-suite. This way of interacting with the test-suite is deprecated in 13 favor of running the test-suite using LNT, but may continue to prove 14 useful for some users. See the Testing Guide's :ref:`test-suite Quickstart 15 <test-suite-quickstart>` section for more information. 16 17 Test suite Structure 18 ==================== 19 20 The ``test-suite`` module contains a number of programs that can be 21 compiled with LLVM and executed. These programs are compiled using the 22 native compiler and various LLVM backends. The output from the program 23 compiled with the native compiler is assumed correct; the results from 24 the other programs are compared to the native program output and pass if 25 they match. 26 27 When executing tests, it is usually a good idea to start out with a 28 subset of the available tests or programs. This makes test run times 29 smaller at first and later on this is useful to investigate individual 30 test failures. To run some test only on a subset of programs, simply 31 change directory to the programs you want tested and run ``gmake`` 32 there. Alternatively, you can run a different test using the ``TEST`` 33 variable to change what tests or run on the selected programs (see below 34 for more info). 35 36 In addition for testing correctness, the ``test-suite`` directory also 37 performs timing tests of various LLVM optimizations. It also records 38 compilation times for the compilers and the JIT. This information can be 39 used to compare the effectiveness of LLVM's optimizations and code 40 generation. 41 42 ``test-suite`` tests are divided into three types of tests: MultiSource, 43 SingleSource, and External. 44 45 - ``test-suite/SingleSource`` 46 47 The SingleSource directory contains test programs that are only a 48 single source file in size. These are usually small benchmark 49 programs or small programs that calculate a particular value. Several 50 such programs are grouped together in each directory. 51 52 - ``test-suite/MultiSource`` 53 54 The MultiSource directory contains subdirectories which contain 55 entire programs with multiple source files. Large benchmarks and 56 whole applications go here. 57 58 - ``test-suite/External`` 59 60 The External directory contains Makefiles for building code that is 61 external to (i.e., not distributed with) LLVM. The most prominent 62 members of this directory are the SPEC 95 and SPEC 2000 benchmark 63 suites. The ``External`` directory does not contain these actual 64 tests, but only the Makefiles that know how to properly compile these 65 programs from somewhere else. The presence and location of these 66 external programs is configured by the test-suite ``configure`` 67 script. 68 69 Each tree is then subdivided into several categories, including 70 applications, benchmarks, regression tests, code that is strange 71 grammatically, etc. These organizations should be relatively self 72 explanatory. 73 74 Some tests are known to fail. Some are bugs that we have not fixed yet; 75 others are features that we haven't added yet (or may never add). In the 76 regression tests, the result for such tests will be XFAIL (eXpected 77 FAILure). In this way, you can tell the difference between an expected 78 and unexpected failure. 79 80 The tests in the test suite have no such feature at this time. If the 81 test passes, only warnings and other miscellaneous output will be 82 generated. If a test fails, a large <program> FAILED message will be 83 displayed. This will help you separate benign warnings from actual test 84 failures. 85 86 Running the test suite 87 ====================== 88 89 First, all tests are executed within the LLVM object directory tree. 90 They *are not* executed inside of the LLVM source tree. This is because 91 the test suite creates temporary files during execution. 92 93 To run the test suite, you need to use the following steps: 94 95 #. ``cd`` into the ``llvm/projects`` directory in your source tree. 96 #. Check out the ``test-suite`` module with: 97 98 .. code-block:: bash 99 100 % svn co http://llvm.org/svn/llvm-project/test-suite/trunk test-suite 101 102 This will get the test suite into ``llvm/projects/test-suite``. 103 104 #. Configure and build ``llvm``. 105 106 #. Configure and build ``llvm-gcc``. 107 108 #. Install ``llvm-gcc`` somewhere. 109 110 #. *Re-configure* ``llvm`` from the top level of each build tree (LLVM 111 object directory tree) in which you want to run the test suite, just 112 as you do before building LLVM. 113 114 During the *re-configuration*, you must either: (1) have ``llvm-gcc`` 115 you just built in your path, or (2) specify the directory where your 116 just-built ``llvm-gcc`` is installed using 117 ``--with-llvmgccdir=$LLVM_GCC_DIR``. 118 119 You must also tell the configure machinery that the test suite is 120 available so it can be configured for your build tree: 121 122 .. code-block:: bash 123 124 % cd $LLVM_OBJ_ROOT ; $LLVM_SRC_ROOT/configure [--with-llvmgccdir=$LLVM_GCC_DIR] 125 126 [Remember that ``$LLVM_GCC_DIR`` is the directory where you 127 *installed* llvm-gcc, not its src or obj directory.] 128 129 #. You can now run the test suite from your build tree as follows: 130 131 .. code-block:: bash 132 133 % cd $LLVM_OBJ_ROOT/projects/test-suite 134 % make 135 136 Note that the second and third steps only need to be done once. After 137 you have the suite checked out and configured, you don't need to do it 138 again (unless the test code or configure script changes). 139 140 Configuring External Tests 141 -------------------------- 142 143 In order to run the External tests in the ``test-suite`` module, you 144 must specify *--with-externals*. This must be done during the 145 *re-configuration* step (see above), and the ``llvm`` re-configuration 146 must recognize the previously-built ``llvm-gcc``. If any of these is 147 missing or neglected, the External tests won't work. 148 149 * *--with-externals* 150 151 * *--with-externals=<directory>* 152 153 This tells LLVM where to find any external tests. They are expected to 154 be in specifically named subdirectories of <``directory``>. If 155 ``directory`` is left unspecified, ``configure`` uses the default value 156 ``/home/vadve/shared/benchmarks/speccpu2000/benchspec``. Subdirectory 157 names known to LLVM include: 158 159 * spec95 160 161 * speccpu2000 162 163 * speccpu2006 164 165 * povray31 166 167 Others are added from time to time, and can be determined from 168 ``configure``. 169 170 Running different tests 171 ----------------------- 172 173 In addition to the regular "whole program" tests, the ``test-suite`` 174 module also provides a mechanism for compiling the programs in different 175 ways. If the variable TEST is defined on the ``gmake`` command line, the 176 test system will include a Makefile named 177 ``TEST.<value of TEST variable>.Makefile``. This Makefile can modify 178 build rules to yield different results. 179 180 For example, the LLVM nightly tester uses ``TEST.nightly.Makefile`` to 181 create the nightly test reports. To run the nightly tests, run 182 ``gmake TEST=nightly``. 183 184 There are several TEST Makefiles available in the tree. Some of them are 185 designed for internal LLVM research and will not work outside of the 186 LLVM research group. They may still be valuable, however, as a guide to 187 writing your own TEST Makefile for any optimization or analysis passes 188 that you develop with LLVM. 189 190 Generating test output 191 ---------------------- 192 193 There are a number of ways to run the tests and generate output. The 194 most simple one is simply running ``gmake`` with no arguments. This will 195 compile and run all programs in the tree using a number of different 196 methods and compare results. Any failures are reported in the output, 197 but are likely drowned in the other output. Passes are not reported 198 explicitly. 199 200 Somewhat better is running ``gmake TEST=sometest test``, which runs the 201 specified test and usually adds per-program summaries to the output 202 (depending on which sometest you use). For example, the ``nightly`` test 203 explicitly outputs TEST-PASS or TEST-FAIL for every test after each 204 program. Though these lines are still drowned in the output, it's easy 205 to grep the output logs in the Output directories. 206 207 Even better are the ``report`` and ``report.format`` targets (where 208 ``format`` is one of ``html``, ``csv``, ``text`` or ``graphs``). The 209 exact contents of the report are dependent on which ``TEST`` you are 210 running, but the text results are always shown at the end of the run and 211 the results are always stored in the ``report.<type>.format`` file (when 212 running with ``TEST=<type>``). The ``report`` also generate a file 213 called ``report.<type>.raw.out`` containing the output of the entire 214 test run. 215 216 Writing custom tests for the test suite 217 --------------------------------------- 218 219 Assuming you can run the test suite, (e.g. 220 "``gmake TEST=nightly report``" should work), it is really easy to run 221 optimizations or code generator components against every program in the 222 tree, collecting statistics or running custom checks for correctness. At 223 base, this is how the nightly tester works, it's just one example of a 224 general framework. 225 226 Lets say that you have an LLVM optimization pass, and you want to see 227 how many times it triggers. First thing you should do is add an LLVM 228 `statistic <ProgrammersManual.html#Statistic>`_ to your pass, which will 229 tally counts of things you care about. 230 231 Following this, you can set up a test and a report that collects these 232 and formats them for easy viewing. This consists of two files, a 233 "``test-suite/TEST.XXX.Makefile``" fragment (where XXX is the name of 234 your test) and a "``test-suite/TEST.XXX.report``" file that indicates 235 how to format the output into a table. There are many example reports of 236 various levels of sophistication included with the test suite, and the 237 framework is very general. 238 239 If you are interested in testing an optimization pass, check out the 240 "libcalls" test as an example. It can be run like this: 241 242 .. code-block:: bash 243 244 % cd llvm/projects/test-suite/MultiSource/Benchmarks # or some other level 245 % make TEST=libcalls report 246 247 This will do a bunch of stuff, then eventually print a table like this: 248 249 :: 250 251 Name | total | #exit | 252 ... 253 FreeBench/analyzer/analyzer | 51 | 6 | 254 FreeBench/fourinarow/fourinarow | 1 | 1 | 255 FreeBench/neural/neural | 19 | 9 | 256 FreeBench/pifft/pifft | 5 | 3 | 257 MallocBench/cfrac/cfrac | 1 | * | 258 MallocBench/espresso/espresso | 52 | 12 | 259 MallocBench/gs/gs | 4 | * | 260 Prolangs-C/TimberWolfMC/timberwolfmc | 302 | * | 261 Prolangs-C/agrep/agrep | 33 | 12 | 262 Prolangs-C/allroots/allroots | * | * | 263 Prolangs-C/assembler/assembler | 47 | * | 264 Prolangs-C/bison/mybison | 74 | * | 265 ... 266 267 This basically is grepping the -stats output and displaying it in a 268 table. You can also use the "TEST=libcalls report.html" target to get 269 the table in HTML form, similarly for report.csv and report.tex. 270 271 The source for this is in ``test-suite/TEST.libcalls.*``. The format is 272 pretty simple: the Makefile indicates how to run the test (in this case, 273 "``opt -simplify-libcalls -stats``"), and the report contains one line 274 for each column of the output. The first value is the header for the 275 column and the second is the regex to grep the output of the command 276 for. There are lots of example reports that can do fancy stuff. 277
