1 .. SPDX-License-Identifier: GPL-2.0 1 .. SPDX-License-Identifier: GPL-2.0
2 2
>> 3 ===========
>> 4 Using KUnit
>> 5 ===========
>> 6
>> 7 The purpose of this document is to describe what KUnit is, how it works, how it
>> 8 is intended to be used, and all the concepts and terminology that are needed to
>> 9 understand it. This guide assumes a working knowledge of the Linux kernel and
>> 10 some basic knowledge of testing.
>> 11
>> 12 For a high level introduction to KUnit, including setting up KUnit for your
>> 13 project, see :doc:`start`.
>> 14
>> 15 Organization of this document
>> 16 =============================
>> 17
>> 18 This document is organized into two main sections: Testing and Common Patterns.
>> 19 The first covers what unit tests are and how to use KUnit to write them. The
>> 20 second covers common testing patterns, e.g. how to isolate code and make it
>> 21 possible to unit test code that was otherwise un-unit-testable.
>> 22
>> 23 Testing
>> 24 =======
>> 25
>> 26 What is KUnit?
>> 27 --------------
>> 28
>> 29 "K" is short for "kernel" so "KUnit" is the "(Linux) Kernel Unit Testing
>> 30 Framework." KUnit is intended first and foremost for writing unit tests; it is
>> 31 general enough that it can be used to write integration tests; however, this is
>> 32 a secondary goal. KUnit has no ambition of being the only testing framework for
>> 33 the kernel; for example, it does not intend to be an end-to-end testing
>> 34 framework.
>> 35
>> 36 What is Unit Testing?
>> 37 ---------------------
>> 38
>> 39 A `unit test <https://martinfowler.com/bliki/UnitTest.html>`_ is a test that
>> 40 tests code at the smallest possible scope, a *unit* of code. In the C
>> 41 programming language that's a function.
>> 42
>> 43 Unit tests should be written for all the publicly exposed functions in a
>> 44 compilation unit; so that is all the functions that are exported in either a
>> 45 *class* (defined below) or all functions which are **not** static.
>> 46
3 Writing Tests 47 Writing Tests
4 ============= !! 48 -------------
5 49
6 Test Cases 50 Test Cases
7 ---------- !! 51 ~~~~~~~~~~
8 52
9 The fundamental unit in KUnit is the test case 53 The fundamental unit in KUnit is the test case. A test case is a function with
10 the signature ``void (*)(struct kunit *test)`` !! 54 the signature ``void (*)(struct kunit *test)``. It calls a function to be tested
11 and then sets *expectations* for what should h 55 and then sets *expectations* for what should happen. For example:
12 56
13 .. code-block:: c 57 .. code-block:: c
14 58
15 void example_test_success(struct kunit 59 void example_test_success(struct kunit *test)
16 { 60 {
17 } 61 }
18 62
19 void example_test_failure(struct kunit 63 void example_test_failure(struct kunit *test)
20 { 64 {
21 KUNIT_FAIL(test, "This test ne 65 KUNIT_FAIL(test, "This test never passes.");
22 } 66 }
23 67
24 In the above example, ``example_test_success`` !! 68 In the above example ``example_test_success`` always passes because it does
25 nothing; no expectations are set, and therefor !! 69 nothing; no expectations are set, so all expectations pass. On the other hand
26 other hand ``example_test_failure`` always fai !! 70 ``example_test_failure`` always fails because it calls ``KUNIT_FAIL``, which is
27 which is a special expectation that logs a mes !! 71 a special expectation that logs a message and causes the test case to fail.
28 fail. <<
29 72
30 Expectations 73 Expectations
31 ~~~~~~~~~~~~ 74 ~~~~~~~~~~~~
32 An *expectation* specifies that we expect a pi !! 75 An *expectation* is a way to specify that you expect a piece of code to do
33 test. An expectation is called like a function !! 76 something in a test. An expectation is called like a function. A test is made
34 expectations about the behavior of a piece of !! 77 by setting expectations about the behavior of a piece of code under test; when
35 expectations fail, the test case fails and inf !! 78 one or more of the expectations fail, the test case fails and information about
36 logged. For example: !! 79 the failure is logged. For example:
37 80
38 .. code-block:: c 81 .. code-block:: c
39 82
40 void add_test_basic(struct kunit *test 83 void add_test_basic(struct kunit *test)
41 { 84 {
42 KUNIT_EXPECT_EQ(test, 1, add(1 85 KUNIT_EXPECT_EQ(test, 1, add(1, 0));
43 KUNIT_EXPECT_EQ(test, 2, add(1 86 KUNIT_EXPECT_EQ(test, 2, add(1, 1));
44 } 87 }
45 88
46 In the above example, ``add_test_basic`` makes !! 89 In the above example ``add_test_basic`` makes a number of assertions about the
47 behavior of a function called ``add``. The fir !! 90 behavior of a function called ``add``; the first parameter is always of type
48 ``struct kunit *``, which contains information !! 91 ``struct kunit *``, which contains information about the current test context;
49 The second parameter, in this case, is what th !! 92 the second parameter, in this case, is what the value is expected to be; the
50 last value is what the value actually is. If ` 93 last value is what the value actually is. If ``add`` passes all of these
51 expectations, the test case, ``add_test_basic` 94 expectations, the test case, ``add_test_basic`` will pass; if any one of these
52 expectations fails, the test case will fail. 95 expectations fails, the test case will fail.
53 96
54 A test case *fails* when any expectation is vi !! 97 It is important to understand that a test case *fails* when any expectation is
55 continue to run, and try other expectations un !! 98 violated; however, the test will continue running, potentially trying other
56 otherwise terminated. This is as opposed to *a !! 99 expectations until the test case ends or is otherwise terminated. This is as
57 later. !! 100 opposed to *assertions* which are discussed later.
58 101
59 To learn about more KUnit expectations, see Do !! 102 To learn about more expectations supported by KUnit, see :doc:`api/test`.
60 103
61 .. note:: 104 .. note::
62 A single test case should be short, easy to !! 105 A single test case should be pretty short, pretty easy to understand,
63 single behavior. !! 106 focused on a single behavior.
64 107
65 For example, if we want to rigorously test the !! 108 For example, if we wanted to properly test the add function above, we would
66 additional tests cases which would test each p !! 109 create additional tests cases which would each test a different property that an
67 should have as shown below: !! 110 add function should have like this:
68 111
69 .. code-block:: c 112 .. code-block:: c
70 113
71 void add_test_basic(struct kunit *test 114 void add_test_basic(struct kunit *test)
72 { 115 {
73 KUNIT_EXPECT_EQ(test, 1, add(1 116 KUNIT_EXPECT_EQ(test, 1, add(1, 0));
74 KUNIT_EXPECT_EQ(test, 2, add(1 117 KUNIT_EXPECT_EQ(test, 2, add(1, 1));
75 } 118 }
76 119
77 void add_test_negative(struct kunit *t 120 void add_test_negative(struct kunit *test)
78 { 121 {
79 KUNIT_EXPECT_EQ(test, 0, add(- 122 KUNIT_EXPECT_EQ(test, 0, add(-1, 1));
80 } 123 }
81 124
82 void add_test_max(struct kunit *test) 125 void add_test_max(struct kunit *test)
83 { 126 {
84 KUNIT_EXPECT_EQ(test, INT_MAX, 127 KUNIT_EXPECT_EQ(test, INT_MAX, add(0, INT_MAX));
85 KUNIT_EXPECT_EQ(test, -1, add( 128 KUNIT_EXPECT_EQ(test, -1, add(INT_MAX, INT_MIN));
86 } 129 }
87 130
88 void add_test_overflow(struct kunit *t 131 void add_test_overflow(struct kunit *test)
89 { 132 {
90 KUNIT_EXPECT_EQ(test, INT_MIN, 133 KUNIT_EXPECT_EQ(test, INT_MIN, add(INT_MAX, 1));
91 } 134 }
92 135
>> 136 Notice how it is immediately obvious what all the properties that we are testing
>> 137 for are.
>> 138
93 Assertions 139 Assertions
94 ~~~~~~~~~~ 140 ~~~~~~~~~~
95 141
96 An assertion is like an expectation, except th !! 142 KUnit also has the concept of an *assertion*. An assertion is just like an
97 terminates the test case if the condition is n !! 143 expectation except the assertion immediately terminates the test case if it is
98 !! 144 not satisfied.
99 .. code-block:: c <<
100 <<
101 static void test_sort(struct kunit *te <<
102 { <<
103 int *a, i, r = 1; <<
104 a = kunit_kmalloc_array(test, <<
105 KUNIT_ASSERT_NOT_ERR_OR_NULL(t <<
106 for (i = 0; i < TEST_LEN; i++) <<
107 r = (r * 725861) % 659 <<
108 a[i] = r; <<
109 } <<
110 sort(a, TEST_LEN, sizeof(*a), <<
111 for (i = 0; i < TEST_LEN-1; i+ <<
112 KUNIT_EXPECT_LE(test, <<
113 } <<
114 <<
115 In this example, we need to be able to allocat <<
116 function. So we use ``KUNIT_ASSERT_NOT_ERR_OR_ <<
117 there's an allocation error. <<
118 <<
119 .. note:: <<
120 In other test frameworks, ``ASSERT`` macros <<
121 ``return`` so they only work from the test <<
122 current kthread on failure, so you can call <<
123 <<
124 .. note:: <<
125 Warning: There is an exception to the above <<
126 in the suite's exit() function, or in the f <<
127 run when a test is shutting down, and an as <<
128 cleanup code from running, potentially lead <<
129 <<
130 Customizing error messages <<
131 -------------------------- <<
132 145
133 Each of the ``KUNIT_EXPECT`` and ``KUNIT_ASSER !! 146 For example:
134 variant. These take a format string and argum <<
135 context to the automatically generated error m <<
136 <<
137 .. code-block:: c <<
138 <<
139 char some_str[41]; <<
140 generate_sha1_hex_string(some_str); <<
141 <<
142 /* Before. Not easy to tell why the te <<
143 KUNIT_EXPECT_EQ(test, strlen(some_str) <<
144 <<
145 /* After. Now we see the offending str <<
146 KUNIT_EXPECT_EQ_MSG(test, strlen(some_ <<
147 <<
148 Alternatively, one can take full control over <<
149 ``KUNIT_FAIL()``, e.g. <<
150 147
151 .. code-block:: c 148 .. code-block:: c
152 149
153 /* Before */ !! 150 static void mock_test_do_expect_default_return(struct kunit *test)
154 KUNIT_EXPECT_EQ(test, some_setup_funct !! 151 {
155 !! 152 struct mock_test_context *ctx = test->priv;
156 /* After: full control over the failur !! 153 struct mock *mock = ctx->mock;
157 if (some_setup_function()) !! 154 int param0 = 5, param1 = -5;
158 KUNIT_FAIL(test, "Failed to se !! 155 const char *two_param_types[] = {"int", "int"};
159 !! 156 const void *two_params[] = {¶m0, ¶m1};
>> 157 const void *ret;
>> 158
>> 159 ret = mock->do_expect(mock,
>> 160 "test_printk", test_printk,
>> 161 two_param_types, two_params,
>> 162 ARRAY_SIZE(two_params));
>> 163 KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ret);
>> 164 KUNIT_EXPECT_EQ(test, -4, *((int *) ret));
>> 165 }
>> 166
>> 167 In this example, the method under test should return a pointer to a value, so
>> 168 if the pointer returned by the method is null or an errno, we don't want to
>> 169 bother continuing the test since the following expectation could crash the test
>> 170 case. `ASSERT_NOT_ERR_OR_NULL(...)` allows us to bail out of the test case if
>> 171 the appropriate conditions have not been satisfied to complete the test.
160 172
161 Test Suites 173 Test Suites
162 ~~~~~~~~~~~ 174 ~~~~~~~~~~~
163 175
164 We need many test cases covering all the unit' !! 176 Now obviously one unit test isn't very helpful; the power comes from having
165 many similar tests. In order to reduce duplica !! 177 many test cases covering all of a unit's behaviors. Consequently it is common
166 tests, most unit testing frameworks (including !! 178 to have many *similar* tests; in order to reduce duplication in these closely
167 *test suite*. A test suite is a collection of !! 179 related tests most unit testing frameworks - including KUnit - provide the
168 with optional setup and teardown functions tha !! 180 concept of a *test suite*. A *test suite* is just a collection of test cases
169 suite and/or every test case. !! 181 for a unit of code with a set up function that gets invoked before every test
>> 182 case and then a tear down function that gets invoked after every test case
>> 183 completes.
170 184
171 .. note:: !! 185 Example:
172 A test case will only run if it is associat <<
173 <<
174 For example: <<
175 186
176 .. code-block:: c 187 .. code-block:: c
177 188
178 static struct kunit_case example_test_ 189 static struct kunit_case example_test_cases[] = {
179 KUNIT_CASE(example_test_foo), 190 KUNIT_CASE(example_test_foo),
180 KUNIT_CASE(example_test_bar), 191 KUNIT_CASE(example_test_bar),
181 KUNIT_CASE(example_test_baz), 192 KUNIT_CASE(example_test_baz),
182 {} 193 {}
183 }; 194 };
184 195
185 static struct kunit_suite example_test 196 static struct kunit_suite example_test_suite = {
186 .name = "example", 197 .name = "example",
187 .init = example_test_init, 198 .init = example_test_init,
188 .exit = example_test_exit, 199 .exit = example_test_exit,
189 .suite_init = example_suite_in <<
190 .suite_exit = example_suite_ex <<
191 .test_cases = example_test_cas 200 .test_cases = example_test_cases,
192 }; 201 };
193 kunit_test_suite(example_test_suite); 202 kunit_test_suite(example_test_suite);
194 203
195 In the above example, the test suite ``example !! 204 In the above example the test suite, ``example_test_suite``, would run the test
196 ``example_suite_init``, then run the test case !! 205 cases ``example_test_foo``, ``example_test_bar``, and ``example_test_baz``;
197 ``example_test_bar``, and ``example_test_baz`` !! 206 each would have ``example_test_init`` called immediately before it and would
198 ``example_test_init`` called immediately befor !! 207 have ``example_test_exit`` called immediately after it.
199 called immediately after it. Finally, ``exampl !! 208 ``kunit_test_suite(example_test_suite)`` registers the test suite with the
200 after everything else. ``kunit_test_suite(exam !! 209 KUnit test framework.
201 test suite with the KUnit test framework. <<
202 210
203 .. note:: 211 .. note::
204 The ``exit`` and ``suite_exit`` functions w !! 212 A test case will only be run if it is associated with a test suite.
205 ``suite_init`` fail. Make sure that they ca <<
206 state which may result from ``init`` or ``s <<
207 or exiting early. <<
208 <<
209 ``kunit_test_suite(...)`` is a macro which tel <<
210 specified test suite in a special linker secti <<
211 either after ``late_init``, or when the test m <<
212 built as a module). <<
213 213
214 For more information, see Documentation/dev-to !! 214 ``kunit_test_suite(...)`` is a macro which tells the linker to put the specified
>> 215 test suite in a special linker section so that it can be run by KUnit either
>> 216 after late_init, or when the test module is loaded (depending on whether the
>> 217 test was built in or not).
215 218
216 .. _kunit-on-non-uml: !! 219 For more information on these types of things see the :doc:`api/test`.
217 <<
218 Writing Tests For Other Architectures <<
219 ------------------------------------- <<
220 <<
221 It is better to write tests that run on UML to <<
222 particular architecture. It is better to write <<
223 another easy to obtain (and monetarily free) s <<
224 piece of hardware. <<
225 <<
226 Nevertheless, there are still valid reasons to <<
227 or hardware specific. For example, we might wa <<
228 belongs in ``arch/some-arch/*``. Even so, try <<
229 not depend on physical hardware. Some of our t <<
230 only few tests actually require the hardware t <<
231 available, instead of disabling tests, we can <<
232 <<
233 Now that we have narrowed down exactly what bi <<
234 actual procedure for writing and running the t <<
235 KUnit tests. <<
236 <<
237 .. important:: <<
238 We may have to reset hardware state. If thi <<
239 be able to run one test case per invocation <<
240 <<
241 .. TODO(brendanhiggins@google.com): Add an act <<
242 dependent KUnit test. <<
243 220
244 Common Patterns 221 Common Patterns
245 =============== 222 ===============
246 223
247 Isolating Behavior 224 Isolating Behavior
248 ------------------ 225 ------------------
249 226
250 Unit testing limits the amount of code under t !! 227 The most important aspect of unit testing that other forms of testing do not
251 what code gets run when the unit under test ca !! 228 provide is the ability to limit the amount of code under test to a single unit.
252 is exposed as part of an API such that the def !! 229 In practice, this is only possible by being able to control what code gets run
253 changed without affecting the rest of the code !! 230 when the unit under test calls a function and this is usually accomplished
254 from two constructs: classes, which are struct !! 231 through some sort of indirection where a function is exposed as part of an API
255 provided by the implementer, and architecture- !! 232 such that the definition of that function can be changed without affecting the
256 definitions selected at compile time. !! 233 rest of the code base. In the kernel this primarily comes from two constructs,
>> 234 classes, structs that contain function pointers that are provided by the
>> 235 implementer, and architecture-specific functions which have definitions selected
>> 236 at compile time.
257 237
258 Classes 238 Classes
259 ~~~~~~~ 239 ~~~~~~~
260 240
261 Classes are not a construct that is built into 241 Classes are not a construct that is built into the C programming language;
262 however, it is an easily derived concept. Acco !! 242 however, it is an easily derived concept. Accordingly, pretty much every project
263 project that does not use a standardized objec !! 243 that does not use a standardized object oriented library (like GNOME's GObject)
264 GObject) has their own slightly different way !! 244 has their own slightly different way of doing object oriented programming; the
265 programming; the Linux kernel is no exception. !! 245 Linux kernel is no exception.
266 246
267 The central concept in kernel object oriented 247 The central concept in kernel object oriented programming is the class. In the
268 kernel, a *class* is a struct that contains fu 248 kernel, a *class* is a struct that contains function pointers. This creates a
269 contract between *implementers* and *users* si 249 contract between *implementers* and *users* since it forces them to use the
270 same function signature without having to call !! 250 same function signature without having to call the function directly. In order
271 class, the function pointers must specify that !! 251 for it to truly be a class, the function pointers must specify that a pointer
272 a *class handle*, be one of the parameters. Th !! 252 to the class, known as a *class handle*, be one of the parameters; this makes
273 known as *methods*) have access to member vari !! 253 it possible for the member functions (also known as *methods*) to have access
274 allowing the same implementation to have multi !! 254 to member variables (more commonly known as *fields*) allowing the same
275 !! 255 implementation to have multiple *instances*.
276 A class can be *overridden* by *child classes* !! 256
277 in the child class. Then when the child class !! 257 Typically a class can be *overridden* by *child classes* by embedding the
278 implementation knows that the pointer passed t !! 258 *parent class* in the child class. Then when a method provided by the child
279 within the child. Thus, the child can compute !! 259 class is called, the child implementation knows that the pointer passed to it is
280 pointer to the parent is always a fixed offset !! 260 of a parent contained within the child; because of this, the child can compute
281 This offset is the offset of the parent contai !! 261 the pointer to itself because the pointer to the parent is always a fixed offset
282 example: !! 262 from the pointer to the child; this offset is the offset of the parent contained
>> 263 in the child struct. For example:
283 264
284 .. code-block:: c 265 .. code-block:: c
285 266
286 struct shape { 267 struct shape {
287 int (*area)(struct shape *this 268 int (*area)(struct shape *this);
288 }; 269 };
289 270
290 struct rectangle { 271 struct rectangle {
291 struct shape parent; 272 struct shape parent;
292 int length; 273 int length;
293 int width; 274 int width;
294 }; 275 };
295 276
296 int rectangle_area(struct shape *this) 277 int rectangle_area(struct shape *this)
297 { 278 {
298 struct rectangle *self = conta !! 279 struct rectangle *self = container_of(this, struct shape, parent);
299 280
300 return self->length * self->wi 281 return self->length * self->width;
301 }; 282 };
302 283
303 void rectangle_new(struct rectangle *s 284 void rectangle_new(struct rectangle *self, int length, int width)
304 { 285 {
305 self->parent.area = rectangle_ 286 self->parent.area = rectangle_area;
306 self->length = length; 287 self->length = length;
307 self->width = width; 288 self->width = width;
308 } 289 }
309 290
310 In this example, computing the pointer to the !! 291 In this example (as in most kernel code) the operation of computing the pointer
311 parent is done by ``container_of``. !! 292 to the child from the pointer to the parent is done by ``container_of``.
312 293
313 Faking Classes 294 Faking Classes
314 ~~~~~~~~~~~~~~ 295 ~~~~~~~~~~~~~~
315 296
316 In order to unit test a piece of code that cal 297 In order to unit test a piece of code that calls a method in a class, the
317 behavior of the method must be controllable, o 298 behavior of the method must be controllable, otherwise the test ceases to be a
318 unit test and becomes an integration test. 299 unit test and becomes an integration test.
319 300
320 A fake class implements a piece of code that i !! 301 A fake just provides an implementation of a piece of code that is different than
321 production instance, but behaves identical fro !! 302 what runs in a production instance, but behaves identically from the standpoint
322 This is done to replace a dependency that is h !! 303 of the callers; this is usually done to replace a dependency that is hard to
323 example, implementing a fake EEPROM that store !! 304 deal with, or is slow.
324 internal buffer. Assume we have a class that r !! 305
>> 306 A good example for this might be implementing a fake EEPROM that just stores the
>> 307 "contents" in an internal buffer. For example, let's assume we have a class that
>> 308 represents an EEPROM:
325 309
326 .. code-block:: c 310 .. code-block:: c
327 311
328 struct eeprom { 312 struct eeprom {
329 ssize_t (*read)(struct eeprom 313 ssize_t (*read)(struct eeprom *this, size_t offset, char *buffer, size_t count);
330 ssize_t (*write)(struct eeprom 314 ssize_t (*write)(struct eeprom *this, size_t offset, const char *buffer, size_t count);
331 }; 315 };
332 316
333 And we want to test code that buffers writes t !! 317 And we want to test some code that buffers writes to the EEPROM:
334 318
335 .. code-block:: c 319 .. code-block:: c
336 320
337 struct eeprom_buffer { 321 struct eeprom_buffer {
338 ssize_t (*write)(struct eeprom 322 ssize_t (*write)(struct eeprom_buffer *this, const char *buffer, size_t count);
339 int flush(struct eeprom_buffer 323 int flush(struct eeprom_buffer *this);
340 size_t flush_count; /* Flushes 324 size_t flush_count; /* Flushes when buffer exceeds flush_count. */
341 }; 325 };
342 326
343 struct eeprom_buffer *new_eeprom_buffe 327 struct eeprom_buffer *new_eeprom_buffer(struct eeprom *eeprom);
344 void destroy_eeprom_buffer(struct eepr 328 void destroy_eeprom_buffer(struct eeprom *eeprom);
345 329
346 We can test this code by *faking out* the unde !! 330 We can easily test this code by *faking out* the underlying EEPROM:
347 331
348 .. code-block:: c 332 .. code-block:: c
349 333
350 struct fake_eeprom { 334 struct fake_eeprom {
351 struct eeprom parent; 335 struct eeprom parent;
352 char contents[FAKE_EEPROM_CONT 336 char contents[FAKE_EEPROM_CONTENTS_SIZE];
353 }; 337 };
354 338
355 ssize_t fake_eeprom_read(struct eeprom 339 ssize_t fake_eeprom_read(struct eeprom *parent, size_t offset, char *buffer, size_t count)
356 { 340 {
357 struct fake_eeprom *this = con 341 struct fake_eeprom *this = container_of(parent, struct fake_eeprom, parent);
358 342
359 count = min(count, FAKE_EEPROM 343 count = min(count, FAKE_EEPROM_CONTENTS_SIZE - offset);
360 memcpy(buffer, this->contents 344 memcpy(buffer, this->contents + offset, count);
361 345
362 return count; 346 return count;
363 } 347 }
364 348
365 ssize_t fake_eeprom_write(struct eepro 349 ssize_t fake_eeprom_write(struct eeprom *parent, size_t offset, const char *buffer, size_t count)
366 { 350 {
367 struct fake_eeprom *this = con 351 struct fake_eeprom *this = container_of(parent, struct fake_eeprom, parent);
368 352
369 count = min(count, FAKE_EEPROM 353 count = min(count, FAKE_EEPROM_CONTENTS_SIZE - offset);
370 memcpy(this->contents + offset 354 memcpy(this->contents + offset, buffer, count);
371 355
372 return count; 356 return count;
373 } 357 }
374 358
375 void fake_eeprom_init(struct fake_eepr 359 void fake_eeprom_init(struct fake_eeprom *this)
376 { 360 {
377 this->parent.read = fake_eepro 361 this->parent.read = fake_eeprom_read;
378 this->parent.write = fake_eepr 362 this->parent.write = fake_eeprom_write;
379 memset(this->contents, 0, FAKE 363 memset(this->contents, 0, FAKE_EEPROM_CONTENTS_SIZE);
380 } 364 }
381 365
382 We can now use it to test ``struct eeprom_buff 366 We can now use it to test ``struct eeprom_buffer``:
383 367
384 .. code-block:: c 368 .. code-block:: c
385 369
386 struct eeprom_buffer_test { 370 struct eeprom_buffer_test {
387 struct fake_eeprom *fake_eepro 371 struct fake_eeprom *fake_eeprom;
388 struct eeprom_buffer *eeprom_b 372 struct eeprom_buffer *eeprom_buffer;
389 }; 373 };
390 374
391 static void eeprom_buffer_test_does_no 375 static void eeprom_buffer_test_does_not_write_until_flush(struct kunit *test)
392 { 376 {
393 struct eeprom_buffer_test *ctx 377 struct eeprom_buffer_test *ctx = test->priv;
394 struct eeprom_buffer *eeprom_b 378 struct eeprom_buffer *eeprom_buffer = ctx->eeprom_buffer;
395 struct fake_eeprom *fake_eepro 379 struct fake_eeprom *fake_eeprom = ctx->fake_eeprom;
396 char buffer[] = {0xff}; 380 char buffer[] = {0xff};
397 381
398 eeprom_buffer->flush_count = S 382 eeprom_buffer->flush_count = SIZE_MAX;
399 383
400 eeprom_buffer->write(eeprom_bu 384 eeprom_buffer->write(eeprom_buffer, buffer, 1);
401 KUNIT_EXPECT_EQ(test, fake_eep 385 KUNIT_EXPECT_EQ(test, fake_eeprom->contents[0], 0);
402 386
403 eeprom_buffer->write(eeprom_bu 387 eeprom_buffer->write(eeprom_buffer, buffer, 1);
404 KUNIT_EXPECT_EQ(test, fake_eep 388 KUNIT_EXPECT_EQ(test, fake_eeprom->contents[1], 0);
405 389
406 eeprom_buffer->flush(eeprom_bu 390 eeprom_buffer->flush(eeprom_buffer);
407 KUNIT_EXPECT_EQ(test, fake_eep 391 KUNIT_EXPECT_EQ(test, fake_eeprom->contents[0], 0xff);
408 KUNIT_EXPECT_EQ(test, fake_eep 392 KUNIT_EXPECT_EQ(test, fake_eeprom->contents[1], 0xff);
409 } 393 }
410 394
411 static void eeprom_buffer_test_flushes 395 static void eeprom_buffer_test_flushes_after_flush_count_met(struct kunit *test)
412 { 396 {
413 struct eeprom_buffer_test *ctx 397 struct eeprom_buffer_test *ctx = test->priv;
414 struct eeprom_buffer *eeprom_b 398 struct eeprom_buffer *eeprom_buffer = ctx->eeprom_buffer;
415 struct fake_eeprom *fake_eepro 399 struct fake_eeprom *fake_eeprom = ctx->fake_eeprom;
416 char buffer[] = {0xff}; 400 char buffer[] = {0xff};
417 401
418 eeprom_buffer->flush_count = 2 402 eeprom_buffer->flush_count = 2;
419 403
420 eeprom_buffer->write(eeprom_bu 404 eeprom_buffer->write(eeprom_buffer, buffer, 1);
421 KUNIT_EXPECT_EQ(test, fake_eep 405 KUNIT_EXPECT_EQ(test, fake_eeprom->contents[0], 0);
422 406
423 eeprom_buffer->write(eeprom_bu 407 eeprom_buffer->write(eeprom_buffer, buffer, 1);
424 KUNIT_EXPECT_EQ(test, fake_eep 408 KUNIT_EXPECT_EQ(test, fake_eeprom->contents[0], 0xff);
425 KUNIT_EXPECT_EQ(test, fake_eep 409 KUNIT_EXPECT_EQ(test, fake_eeprom->contents[1], 0xff);
426 } 410 }
427 411
428 static void eeprom_buffer_test_flushes 412 static void eeprom_buffer_test_flushes_increments_of_flush_count(struct kunit *test)
429 { 413 {
430 struct eeprom_buffer_test *ctx 414 struct eeprom_buffer_test *ctx = test->priv;
431 struct eeprom_buffer *eeprom_b 415 struct eeprom_buffer *eeprom_buffer = ctx->eeprom_buffer;
432 struct fake_eeprom *fake_eepro 416 struct fake_eeprom *fake_eeprom = ctx->fake_eeprom;
433 char buffer[] = {0xff, 0xff}; 417 char buffer[] = {0xff, 0xff};
434 418
435 eeprom_buffer->flush_count = 2 419 eeprom_buffer->flush_count = 2;
436 420
437 eeprom_buffer->write(eeprom_bu 421 eeprom_buffer->write(eeprom_buffer, buffer, 1);
438 KUNIT_EXPECT_EQ(test, fake_eep 422 KUNIT_EXPECT_EQ(test, fake_eeprom->contents[0], 0);
439 423
440 eeprom_buffer->write(eeprom_bu 424 eeprom_buffer->write(eeprom_buffer, buffer, 2);
441 KUNIT_EXPECT_EQ(test, fake_eep 425 KUNIT_EXPECT_EQ(test, fake_eeprom->contents[0], 0xff);
442 KUNIT_EXPECT_EQ(test, fake_eep 426 KUNIT_EXPECT_EQ(test, fake_eeprom->contents[1], 0xff);
443 /* Should have only flushed th 427 /* Should have only flushed the first two bytes. */
444 KUNIT_EXPECT_EQ(test, fake_eep 428 KUNIT_EXPECT_EQ(test, fake_eeprom->contents[2], 0);
445 } 429 }
446 430
447 static int eeprom_buffer_test_init(str 431 static int eeprom_buffer_test_init(struct kunit *test)
448 { 432 {
449 struct eeprom_buffer_test *ctx 433 struct eeprom_buffer_test *ctx;
450 434
451 ctx = kunit_kzalloc(test, size 435 ctx = kunit_kzalloc(test, sizeof(*ctx), GFP_KERNEL);
452 KUNIT_ASSERT_NOT_ERR_OR_NULL(t 436 KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx);
453 437
454 ctx->fake_eeprom = kunit_kzall 438 ctx->fake_eeprom = kunit_kzalloc(test, sizeof(*ctx->fake_eeprom), GFP_KERNEL);
455 KUNIT_ASSERT_NOT_ERR_OR_NULL(t 439 KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx->fake_eeprom);
456 fake_eeprom_init(ctx->fake_eep 440 fake_eeprom_init(ctx->fake_eeprom);
457 441
458 ctx->eeprom_buffer = new_eepro 442 ctx->eeprom_buffer = new_eeprom_buffer(&ctx->fake_eeprom->parent);
459 KUNIT_ASSERT_NOT_ERR_OR_NULL(t 443 KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ctx->eeprom_buffer);
460 444
461 test->priv = ctx; 445 test->priv = ctx;
462 446
463 return 0; 447 return 0;
464 } 448 }
465 449
466 static void eeprom_buffer_test_exit(st 450 static void eeprom_buffer_test_exit(struct kunit *test)
467 { 451 {
468 struct eeprom_buffer_test *ctx 452 struct eeprom_buffer_test *ctx = test->priv;
469 453
470 destroy_eeprom_buffer(ctx->eep 454 destroy_eeprom_buffer(ctx->eeprom_buffer);
471 } 455 }
472 456
473 Testing Against Multiple Inputs !! 457 Testing against multiple inputs
474 ------------------------------- 458 -------------------------------
475 459
476 Testing just a few inputs is not enough to ens !! 460 Testing just a few inputs might not be enough to have confidence that the code
477 for example: testing a hash function. !! 461 works correctly, e.g. for a hash function.
478 462
479 We can write a helper macro or function. The f !! 463 In such cases, it can be helpful to have a helper macro or function, e.g. this
480 For example, to test ``sha1sum(1)``, we can wr !! 464 fictitious example for ``sha1sum(1)``
481 465
482 .. code-block:: c 466 .. code-block:: c
483 467
>> 468 /* Note: the cast is to satisfy overly strict type-checking. */
484 #define TEST_SHA1(in, want) \ 469 #define TEST_SHA1(in, want) \
485 sha1sum(in, out); \ 470 sha1sum(in, out); \
486 KUNIT_EXPECT_STREQ_MSG(test, o !! 471 KUNIT_EXPECT_STREQ_MSG(test, (char *)out, want, "sha1sum(%s)", in);
487 472
488 char out[40]; 473 char out[40];
489 TEST_SHA1("hello world", "2aae6c35c94 474 TEST_SHA1("hello world", "2aae6c35c94fcfb415dbe95f408b9ce91ee846ed");
490 TEST_SHA1("hello world!", "430ce34d020 475 TEST_SHA1("hello world!", "430ce34d020724ed75a196dfc2ad67c77772d169");
491 476
492 Note the use of the ``_MSG`` version of ``KUNI <<
493 detailed error and make the assertions clearer <<
494 477
495 The ``_MSG`` variants are useful when the same !! 478 Note the use of ``KUNIT_EXPECT_STREQ_MSG`` to give more context when it fails
496 times (in a loop or helper function) and thus !! 479 and make it easier to track down. (Yes, in this example, ``want`` is likely
497 identify what failed, as shown below. !! 480 going to be unique enough on its own).
>> 481
>> 482 The ``_MSG`` variants are even more useful when the same expectation is called
>> 483 multiple times (in a loop or helper function) and thus the line number isn't
>> 484 enough to identify what failed, like below.
498 485
499 In complicated cases, we recommend using a *ta !! 486 In some cases, it can be helpful to write a *table-driven test* instead, e.g.
500 helper macro variation, for example: <<
501 487
502 .. code-block:: c 488 .. code-block:: c
503 489
504 int i; 490 int i;
505 char out[40]; 491 char out[40];
506 492
507 struct sha1_test_case { 493 struct sha1_test_case {
508 const char *str; 494 const char *str;
509 const char *sha1; 495 const char *sha1;
510 }; 496 };
511 497
512 struct sha1_test_case cases[] = { 498 struct sha1_test_case cases[] = {
513 { 499 {
514 .str = "hello world", 500 .str = "hello world",
515 .sha1 = "2aae6c35c94fc 501 .sha1 = "2aae6c35c94fcfb415dbe95f408b9ce91ee846ed",
516 }, 502 },
517 { 503 {
518 .str = "hello world!", 504 .str = "hello world!",
519 .sha1 = "430ce34d02072 505 .sha1 = "430ce34d020724ed75a196dfc2ad67c77772d169",
520 }, 506 },
521 }; 507 };
522 for (i = 0; i < ARRAY_SIZE(cases); ++i 508 for (i = 0; i < ARRAY_SIZE(cases); ++i) {
523 sha1sum(cases[i].str, out); 509 sha1sum(cases[i].str, out);
524 KUNIT_EXPECT_STREQ_MSG(test, o !! 510 KUNIT_EXPECT_STREQ_MSG(test, (char *)out, cases[i].sha1,
525 "sha1sum 511 "sha1sum(%s)", cases[i].str);
526 } 512 }
527 513
528 514
529 There is more boilerplate code involved, but i !! 515 There's more boilerplate involved, but it can:
530 <<
531 * be more readable when there are multiple inp <<
532 516
533 * For example, see ``fs/ext4/inode-test.c``. !! 517 * be more readable when there are multiple inputs/outputs thanks to field names,
534 518
535 * reduce duplication if test cases are shared !! 519 * E.g. see ``fs/ext4/inode-test.c`` for an example of both.
>> 520 * reduce duplication if test cases can be shared across multiple tests.
536 521
537 * For example: if we want to test ``sha256su !! 522 * E.g. if we wanted to also test ``sha256sum``, we could add a ``sha256``
538 field and reuse ``cases``. 523 field and reuse ``cases``.
539 524
540 * be converted to a "parameterized test". !! 525 * be converted to a "parameterized test", see below.
541 526
542 Parameterized Testing 527 Parameterized Testing
543 ~~~~~~~~~~~~~~~~~~~~~ 528 ~~~~~~~~~~~~~~~~~~~~~
544 529
545 The table-driven testing pattern is common eno 530 The table-driven testing pattern is common enough that KUnit has special
546 support for it. 531 support for it.
547 532
548 By reusing the same ``cases`` array from above !! 533 Reusing the same ``cases`` array from above, we can write the test as a
549 "parameterized test" with the following. 534 "parameterized test" with the following.
550 535
551 .. code-block:: c 536 .. code-block:: c
552 537
553 // This is copy-pasted from above. 538 // This is copy-pasted from above.
554 struct sha1_test_case { 539 struct sha1_test_case {
555 const char *str; 540 const char *str;
556 const char *sha1; 541 const char *sha1;
557 }; 542 };
558 const struct sha1_test_case cases[] = !! 543 struct sha1_test_case cases[] = {
559 { 544 {
560 .str = "hello world", 545 .str = "hello world",
561 .sha1 = "2aae6c35c94fc 546 .sha1 = "2aae6c35c94fcfb415dbe95f408b9ce91ee846ed",
562 }, 547 },
563 { 548 {
564 .str = "hello world!", 549 .str = "hello world!",
565 .sha1 = "430ce34d02072 550 .sha1 = "430ce34d020724ed75a196dfc2ad67c77772d169",
566 }, 551 },
567 }; 552 };
568 553
569 // Creates `sha1_gen_params()` to iter !! 554 // Need a helper function to generate a name for each test case.
570 // the struct member `str` for the cas !! 555 static void case_to_desc(const struct sha1_test_case *t, char *desc)
571 KUNIT_ARRAY_PARAM_DESC(sha1, cases, st !! 556 {
>> 557 strcpy(desc, t->str);
>> 558 }
>> 559 // Creates `sha1_gen_params()` to iterate over `cases`.
>> 560 KUNIT_ARRAY_PARAM(sha1, cases, case_to_desc);
572 561
573 // Looks no different from a normal te 562 // Looks no different from a normal test.
574 static void sha1_test(struct kunit *te 563 static void sha1_test(struct kunit *test)
575 { 564 {
576 // This function can just cont 565 // This function can just contain the body of the for-loop.
577 // The former `cases[i]` is ac 566 // The former `cases[i]` is accessible under test->param_value.
578 char out[40]; 567 char out[40];
579 struct sha1_test_case *test_pa 568 struct sha1_test_case *test_param = (struct sha1_test_case *)(test->param_value);
580 569
581 sha1sum(test_param->str, out); 570 sha1sum(test_param->str, out);
582 KUNIT_EXPECT_STREQ_MSG(test, o !! 571 KUNIT_EXPECT_STREQ_MSG(test, (char *)out, test_param->sha1,
583 "sha1sum 572 "sha1sum(%s)", test_param->str);
584 } 573 }
585 574
586 // Instead of KUNIT_CASE, we use KUNIT 575 // Instead of KUNIT_CASE, we use KUNIT_CASE_PARAM and pass in the
587 // function declared by KUNIT_ARRAY_PA !! 576 // function declared by KUNIT_ARRAY_PARAM.
588 static struct kunit_case sha1_test_cas 577 static struct kunit_case sha1_test_cases[] = {
589 KUNIT_CASE_PARAM(sha1_test, sh 578 KUNIT_CASE_PARAM(sha1_test, sha1_gen_params),
590 {} 579 {}
591 }; 580 };
592 581
593 Allocating Memory !! 582 .. _kunit-on-non-uml:
594 ----------------- <<
595 <<
596 Where you might use ``kzalloc``, you can inste <<
597 will then ensure that the memory is freed once <<
598 <<
599 This is useful because it lets us use the ``KU <<
600 early from a test without having to worry abou <<
601 For example: <<
602 <<
603 .. code-block:: c <<
604 <<
605 void example_test_allocation(struct ku <<
606 { <<
607 char *buffer = kunit_kzalloc(t <<
608 /* Ensure allocation succeeded <<
609 KUNIT_ASSERT_NOT_ERR_OR_NULL(t <<
610 <<
611 KUNIT_ASSERT_STREQ(test, buffe <<
612 } <<
613 <<
614 Registering Cleanup Actions <<
615 --------------------------- <<
616 <<
617 If you need to perform some cleanup beyond sim <<
618 you can register a custom "deferred action", w <<
619 run when the test exits (whether cleanly, or v <<
620 <<
621 Actions are simple functions with no return va <<
622 context argument, and fulfill the same role as <<
623 and Go tests, "defer" statements in languages <<
624 (in some cases) destructors in RAII languages. <<
625 <<
626 These are very useful for unregistering things <<
627 files or other resources, or freeing resources <<
628 <<
629 For example: <<
630 <<
631 .. code-block:: C <<
632 <<
633 static void cleanup_device(void *ctx) <<
634 { <<
635 struct device *dev = (struct d <<
636 <<
637 device_unregister(dev); <<
638 } <<
639 <<
640 void example_device_test(struct kunit <<
641 { <<
642 struct my_device dev; <<
643 <<
644 device_register(&dev); <<
645 <<
646 kunit_add_action(test, &cleanu <<
647 } <<
648 <<
649 Note that, for functions like device_unregiste <<
650 pointer-sized argument, it's possible to autom <<
651 with the ``KUNIT_DEFINE_ACTION_WRAPPER()`` mac <<
652 <<
653 .. code-block:: C <<
654 <<
655 KUNIT_DEFINE_ACTION_WRAPPER(device_unr <<
656 kunit_add_action(test, &device_unregis <<
657 <<
658 You should do this in preference to manually c <<
659 as casting function pointers will break Contro <<
660 <<
661 ``kunit_add_action`` can fail if, for example, <<
662 You can use ``kunit_add_action_or_reset`` inst <<
663 immediately if it cannot be deferred. <<
664 <<
665 If you need more control over when the cleanup <<
666 can trigger it early using ``kunit_release_act <<
667 with ``kunit_remove_action``. <<
668 <<
669 <<
670 Testing Static Functions <<
671 ------------------------ <<
672 <<
673 If we do not want to expose functions or varia <<
674 conditionally export the used symbol. For exam <<
675 <<
676 .. code-block:: c <<
677 <<
678 /* In my_file.c */ <<
679 <<
680 VISIBLE_IF_KUNIT int do_interesting_th <<
681 EXPORT_SYMBOL_IF_KUNIT(do_interesting_ <<
682 <<
683 /* In my_file.h */ <<
684 <<
685 #if IS_ENABLED(CONFIG_KUNIT) <<
686 int do_interesting_thing(void) <<
687 #endif <<
688 <<
689 Alternatively, you could conditionally ``#incl <<
690 your .c file. For example: <<
691 <<
692 .. code-block:: c <<
693 <<
694 /* In my_file.c */ <<
695 <<
696 static int do_interesting_thing(); <<
697 <<
698 #ifdef CONFIG_MY_KUNIT_TEST <<
699 #include "my_kunit_test.c" <<
700 #endif <<
701 <<
702 Injecting Test-Only Code <<
703 ------------------------ <<
704 <<
705 Similar to as shown above, we can add test-spe <<
706 <<
707 .. code-block:: c <<
708 <<
709 /* In my_file.h */ <<
710 583
711 #ifdef CONFIG_MY_KUNIT_TEST !! 584 KUnit on non-UML architectures
712 /* Defined in my_kunit_test.c */ !! 585 ==============================
713 void test_only_hook(void); <<
714 #else <<
715 void test_only_hook(void) { } <<
716 #endif <<
717 586
718 This test-only code can be made more useful by !! 587 By default KUnit uses UML as a way to provide dependencies for code under test.
719 as shown in next section: *Accessing The Curre !! 588 Under most circumstances KUnit's usage of UML should be treated as an
>> 589 implementation detail of how KUnit works under the hood. Nevertheless, there
>> 590 are instances where being able to run architecture-specific code or test
>> 591 against real hardware is desirable. For these reasons KUnit supports running on
>> 592 other architectures.
720 593
721 Accessing The Current Test !! 594 Running existing KUnit tests on non-UML architectures
722 -------------------------- !! 595 -----------------------------------------------------
723 596
724 In some cases, we need to call test-only code !! 597 There are some special considerations when running existing KUnit tests on
725 is helpful, for example, when providing a fake !! 598 non-UML architectures:
726 to fail any current test from within an error <<
727 We can do this via the ``kunit_test`` field in <<
728 access using the ``kunit_get_current_test()`` <<
729 599
730 ``kunit_get_current_test()`` is safe to call e !! 600 * Hardware may not be deterministic, so a test that always passes or fails
731 KUnit is not enabled, or if no test is running !! 601 when run under UML may not always do so on real hardware.
732 return ``NULL``. This compiles down to either !! 602 * Hardware and VM environments may not be hermetic. KUnit tries its best to
733 so will have a negligible performance impact w !! 603 provide a hermetic environment to run tests; however, it cannot manage state
>> 604 that it doesn't know about outside of the kernel. Consequently, tests that
>> 605 may be hermetic on UML may not be hermetic on other architectures.
>> 606 * Some features and tooling may not be supported outside of UML.
>> 607 * Hardware and VMs are slower than UML.
734 608
735 The example below uses this to implement a "mo !! 609 None of these are reasons not to run your KUnit tests on real hardware; they are
>> 610 only things to be aware of when doing so.
736 611
737 .. code-block:: c !! 612 The biggest impediment will likely be that certain KUnit features and
>> 613 infrastructure may not support your target environment. For example, at this
>> 614 time the KUnit Wrapper (``tools/testing/kunit/kunit.py``) does not work outside
>> 615 of UML. Unfortunately, there is no way around this. Using UML (or even just a
>> 616 particular architecture) allows us to make a lot of assumptions that make it
>> 617 possible to do things which might otherwise be impossible.
738 618
739 #include <kunit/test-bug.h> /* for kun !! 619 Nevertheless, all core KUnit framework features are fully supported on all
>> 620 architectures, and using them is straightforward: all you need to do is to take
>> 621 your kunitconfig, your Kconfig options for the tests you would like to run, and
>> 622 merge them into whatever config your are using for your platform. That's it!
740 623
741 struct test_data { !! 624 For example, let's say you have the following kunitconfig:
742 int foo_result; <<
743 int want_foo_called_with; <<
744 }; <<
745 625
746 static int fake_foo(int arg) !! 626 .. code-block:: none
747 { <<
748 struct kunit *test = kunit_get <<
749 struct test_data *test_data = <<
750 627
751 KUNIT_EXPECT_EQ(test, test_dat !! 628 CONFIG_KUNIT=y
752 return test_data->foo_result; !! 629 CONFIG_KUNIT_EXAMPLE_TEST=y
753 } <<
754 630
755 static void example_simple_test(struct !! 631 If you wanted to run this test on an x86 VM, you might add the following config
756 { !! 632 options to your ``.config``:
757 /* Assume priv (private, a mem <<
758 * the init function) is alloc <<
759 struct test_data *test_data = <<
760 633
761 test_data->foo_result = 42; !! 634 .. code-block:: none
762 test_data->want_foo_called_wit <<
763 635
764 /* In a real test, we'd probab !! 636 CONFIG_KUNIT=y
765 * like an ops struct, etc. in !! 637 CONFIG_KUNIT_EXAMPLE_TEST=y
766 KUNIT_EXPECT_EQ(test, fake_foo !! 638 CONFIG_SERIAL_8250=y
767 } !! 639 CONFIG_SERIAL_8250_CONSOLE=y
768 640
769 In this example, we are using the ``priv`` mem !! 641 All these new options do is enable support for a common serial console needed
770 of passing data to the test from the init func !! 642 for logging.
771 pointer that can be used for any user data. Th <<
772 variables, as it avoids concurrency issues. <<
773 643
774 Had we wanted something more flexible, we coul !! 644 Next, you could build a kernel with these tests as follows:
775 Each test can have multiple resources which ha <<
776 flexibility as a ``priv`` member, but also, fo <<
777 functions to create resources without conflict <<
778 possible to define a clean up function for eac <<
779 avoid resource leaks. For more information, se <<
780 645
781 Failing The Current Test <<
782 ------------------------ <<
783 646
784 If we want to fail the current test, we can us !! 647 .. code-block:: bash
785 which is defined in ``<kunit/test-bug.h>`` and <<
786 For example, we have an option to enable some <<
787 structures as shown below: <<
788 648
789 .. code-block:: c !! 649 make ARCH=x86 olddefconfig
>> 650 make ARCH=x86
790 651
791 #include <kunit/test-bug.h> !! 652 Once you have built a kernel, you could run it on QEMU as follows:
792 653
793 #ifdef CONFIG_EXTRA_DEBUG_CHECKS !! 654 .. code-block:: bash
794 static void validate_my_data(struct da <<
795 { <<
796 if (is_valid(data)) <<
797 return; <<
798 655
799 kunit_fail_current_test("data !! 656 qemu-system-x86_64 -enable-kvm \
>> 657 -m 1024 \
>> 658 -kernel arch/x86_64/boot/bzImage \
>> 659 -append 'console=ttyS0' \
>> 660 --nographic
800 661
801 /* Normal, non-KUnit, error re !! 662 Interspersed in the kernel logs you might see the following:
802 } <<
803 #else <<
804 static void my_debug_function(void) { <<
805 #endif <<
806 663
807 ``kunit_fail_current_test()`` is safe to call !! 664 .. code-block:: none
808 KUnit is not enabled, or if no test is running <<
809 nothing. This compiles down to either a no-op <<
810 have a negligible performance impact when no t <<
811 665
812 Managing Fake Devices and Drivers !! 666 TAP version 14
813 --------------------------------- !! 667 # Subtest: example
>> 668 1..1
>> 669 # example_simple_test: initializing
>> 670 ok 1 - example_simple_test
>> 671 ok 1 - example
814 672
815 When testing drivers or code which interacts w !! 673 Congratulations, you just ran a KUnit test on the x86 architecture!
816 require a ``struct device`` or ``struct device <<
817 up a real device is not required to test any g <<
818 can be used instead. <<
819 674
820 KUnit provides helper functions to create and !! 675 In a similar manner, kunit and kunit tests can also be built as modules,
821 are internally of type ``struct kunit_device`` !! 676 so if you wanted to run tests in this way you might add the following config
822 ``kunit_bus``. These devices support managed d !! 677 options to your ``.config``:
823 described in Documentation/driver-api/driver-m <<
824 678
825 To create a KUnit-managed ``struct device_driv !! 679 .. code-block:: none
826 which will create a driver with the given name <<
827 will automatically be destroyed when the corre <<
828 be manually destroyed with ``driver_unregister <<
829 680
830 To create a fake device, use the ``kunit_devic !! 681 CONFIG_KUNIT=m
831 and register a device, using a new KUnit-manag !! 682 CONFIG_KUNIT_EXAMPLE_TEST=m
832 To provide a specific, non-KUnit-managed drive <<
833 instead. Like with managed drivers, KUnit-mana <<
834 cleaned up when the test finishes, but can be <<
835 ``kunit_device_unregister()``. <<
836 683
837 The KUnit devices should be used in preference !! 684 Once the kernel is built and installed, a simple
838 instead of ``platform_device_register()`` in c <<
839 a platform device. <<
840 685
841 For example: !! 686 .. code-block:: bash
842 687
843 .. code-block:: c !! 688 modprobe example-test
844 689
845 #include <kunit/device.h> !! 690 ...will run the tests.
846 691
847 static void test_my_device(struct kuni !! 692 .. note::
848 { !! 693 Note that you should make sure your test depends on ``KUNIT=y`` in Kconfig
849 struct device *fake_device; !! 694 if the test does not support module build. Otherwise, it will trigger
850 const char *dev_managed_string !! 695 compile errors if ``CONFIG_KUNIT`` is ``m``.
>> 696
>> 697 Writing new tests for other architectures
>> 698 -----------------------------------------
>> 699
>> 700 The first thing you must do is ask yourself whether it is necessary to write a
>> 701 KUnit test for a specific architecture, and then whether it is necessary to
>> 702 write that test for a particular piece of hardware. In general, writing a test
>> 703 that depends on having access to a particular piece of hardware or software (not
>> 704 included in the Linux source repo) should be avoided at all costs.
>> 705
>> 706 Even if you only ever plan on running your KUnit test on your hardware
>> 707 configuration, other people may want to run your tests and may not have access
>> 708 to your hardware. If you write your test to run on UML, then anyone can run your
>> 709 tests without knowing anything about your particular setup, and you can still
>> 710 run your tests on your hardware setup just by compiling for your architecture.
851 711
852 // Create a fake device. !! 712 .. important::
853 fake_device = kunit_device_reg !! 713 Always prefer tests that run on UML to tests that only run under a particular
854 KUNIT_ASSERT_NOT_ERR_OR_NULL(t !! 714 architecture, and always prefer tests that run under QEMU or another easy
>> 715 (and monetarily free) to obtain software environment to a specific piece of
>> 716 hardware.
>> 717
>> 718 Nevertheless, there are still valid reasons to write an architecture or hardware
>> 719 specific test: for example, you might want to test some code that really belongs
>> 720 in ``arch/some-arch/*``. Even so, try your best to write the test so that it
>> 721 does not depend on physical hardware: if some of your test cases don't need the
>> 722 hardware, only require the hardware for tests that actually need it.
>> 723
>> 724 Now that you have narrowed down exactly what bits are hardware specific, the
>> 725 actual procedure for writing and running the tests is pretty much the same as
>> 726 writing normal KUnit tests. One special caveat is that you have to reset
>> 727 hardware state in between test cases; if this is not possible, you may only be
>> 728 able to run one test case per invocation.
855 729
856 // Pass it to functions which !! 730 .. TODO(brendanhiggins@google.com): Add an actual example of an architecture-
857 dev_managed_string = devm_kstr !! 731 dependent KUnit test.
858 732
859 // Everything is cleaned up au !! 733 KUnit debugfs representation
860 } !! 734 ============================
>> 735 When kunit test suites are initialized, they create an associated directory
>> 736 in ``/sys/kernel/debug/kunit/<test-suite>``. The directory contains one file
>> 737
>> 738 - results: "cat results" displays results of each test case and the results
>> 739 of the entire suite for the last test run.
>> 740
>> 741 The debugfs representation is primarily of use when kunit test suites are
>> 742 run in a native environment, either as modules or builtin. Having a way
>> 743 to display results like this is valuable as otherwise results can be
>> 744 intermixed with other events in dmesg output. The maximum size of each
>> 745 results file is KUNIT_LOG_SIZE bytes (defined in ``include/kunit/test.h``).
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.