xref: /dragonfly/usr.bin/dfregress/dfregress.8 (revision 33311965)
1.\"
2.\" Copyright (c) 2011
3.\"	The DragonFly Project.  All rights reserved.
4.\"
5.\" Redistribution and use in source and binary forms, with or without
6.\" modification, are permitted provided that the following conditions
7.\" are met:
8.\"
9.\" 1. Redistributions of source code must retain the above copyright
10.\"    notice, this list of conditions and the following disclaimer.
11.\" 2. Redistributions in binary form must reproduce the above copyright
12.\"    notice, this list of conditions and the following disclaimer in
13.\"    the documentation and/or other materials provided with the
14.\"    distribution.
15.\" 3. Neither the name of The DragonFly Project nor the names of its
16.\"    contributors may be used to endorse or promote products derived
17.\"    from this software without specific, prior written permission.
18.\"
19.\" THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
20.\" ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
21.\" LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
22.\" FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE
23.\" COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
24.\" INCIDENTAL, SPECIAL, EXEMPLARY OR CONSEQUENTIAL DAMAGES (INCLUDING,
25.\" BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
26.\" LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
27.\" AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
28.\" OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
29.\" OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
30.\" SUCH DAMAGE.
31.\"
32.Dd November 17, 2011
33.Dt DFREGRESS 8
34.Os
35.Sh NAME
36.Nm dfregress
37.Nd an automation test driver and framework
38.Sh SYNOPSIS
39.Nm
40.Op Fl o Ar output_plist
41.Op Fl t Ar testcase_dir
42.Op Fl p Ar prepost_dir
43.Ar runlist_file
44.Nm
45.Fl h
46.Sh DESCRIPTION
47.Nm
48is a regression test framework and automation driver.
49It executes a series of testcases as specified in the
50.Ar runlist_file
51and collects the results.
52.Pp
53The path to the testcase collection is specified via the
54.Ar testcase_dir
55argument.
56If this argument is not specified, the default is assumed to be the
57same directory as that of the runlist.
58For example if the used runlist is
59.Pa /usr/src/test/testcases/sample.run
60the default testcase directory, unless otherwise specified, will be
61.Pa /usr/src/test/testcases .
62.Pp
63Similarly the path to the pre- and post commands is
64specified via
65.Ar prepost_dir .
66The
67.Ar prepost_dir
68only needs to be specified if the runlist contains custom pre or
69post commands.
70.Pp
71The output is saved in plist format to the
72.Ar output_plist
73file, or if none is specified, to a file with the same base name as
74the runlist, but in the current working directory and with a
75.Pa .plist
76extension.
77For example if the used runlist is
78.Pa /usr/src/test/testcases/sample.run
79the default output, unless otherwise specified, will be
80.Pa ./sample.plist .
81Other tools, known as frontends, can parse the plist output into
82an easier to read form.
83.Pp
84The following is a summary of the optional parameters:
85.Bl -tag -width indent
86.It Fl o Ar output_plist
87Specifies the file to which to write the results.
88The resulting file,
89.Ar output_plist ,
90will be in plist format.
91.It Fl t Ar testcase_dir
92Specifies the directory in which to find the testcases specified in the runlist.
93.It Fl p Ar prepost_dir
94Specifies the directory in which to find the pre- and post commands used
95in the runlist.
96This argument is only required when the runlist uses custom pre- or post
97commands.
98.It Fl h
99Prints a short synopsis.
100.El
101.Sh RUNLIST SYNTAX
102Testcases are specified one testcase per line, with whitespace separated
103values.
104Empty lines and lines beginning with a
105.Dq #
106are ignored.
107.Pp
108Runlist lines are of the following format:
109.Bd -literal -offset indent
110.Ic testcase type options Cm arguments ...
111.Ed
112.Pp
113The
114.Ic testcase
115part needs to be a relative path from the testcase base directory specified
116by the
117.Fl t
118argument to the resulting (after building the testcase) testcase executable.
119The testcase will be executed with the
120.Cm arguments
121passed as command line arguments to it.
122.Pp
123Valid types are
124.Ic userland ,
125.Ic kernel
126and
127.Ic buildonly :
128.Bl -tag -width indent -offset indent
129.It Ic userland
130A userland testcase is a normal userland executable that returns a non-zero
131exit value on test failure.
132.It Ic kernel
133A kernel testcase is run with the kernel test bridge inside the kernel.
134.It Ic buildonly
135A buildonly test passes when the build for the given testcase succeeds.
136.El
137.Pp
138Valid options are
139.Ic defaults ,
140.Ic make ,
141.Ic timeout ,
142.Ic nobuild ,
143.Ic runas ,
144.Ic intpre ,
145.Ic intpost ,
146.Ic pre ,
147and
148.Ic post :
149.Bl -tag -width indent -offset indent
150.It Ic defaults
151This option does nothing.
152All default options are applied.
153.It Ic make Ar make_command
154Uses
155.Ar make_command
156instead of
157.Xr make 1
158to build the testcase.
159.It Ic timeout Ar timeout
160Sets the timeout for the testcase after which the testcase will be aborted to
161.Ar timeout
162seconds.
163.It Ic nobuild
164If this option is set, the build stage and cleanup stage of the test case
165are not run.
166.It Ic runas Ar uid
167Runs the testcase as the user identified by the given
168.Ar uid .
169.It Ic intpre
170Executes the testcase command with the argument
171.Dq pre
172during the pre-run command stage.
173.It Ic intpost
174Executes the testcase command with the argument
175.Dq post
176during the post-run command stage.
177.It Ic pre Ar precmd
178Executes the command specified by
179.Ar precmd
180during the pre-run command stage.
181.It Ic pre Ar postcmd
182Executes the command specified by
183.Ar postcmd
184during the post-run command stage.
185.El
186.Sh TESTCASE EXECUTION
187.Bl -enum -width 3n -offset indent
188.It
189.Tn "CHDIR"
190to the testcase directory.
191If it fails, the result will be set to
192.Dv RESULT_PREFAIL
193and the
194.Ar sysbuf
195buffer will provide further details.
196.It
197.Tn "BUILD"
198the testcase using the command specified by the
199.Ic make
200option or, if not specified,
201.Xr make 1 ,
202unless the
203.Ic nobuild
204option was specified.
205If there is an internal driver error, the result will be set to
206.Dv RESULT_PREFAIL ,
207the next step to be performed will be
208.Tn "CLEANUP"
209and the
210.Ar sysbuf
211buffer will provide further details.
212If no internal error occurs, the
213.Ar buildbuf
214will contain the build log.
215.It
216.Tn "RUN PRE COMMAND"
217if
218.Ic intpre
219or
220.Ic pre
221are set.
222If there is an internal driver error, the result will be set to
223.Dv RESULT_PREFAIL ,
224the next step to be performed will be
225.Tn "CLEANUP"
226and the
227.Ar sysbuf
228buffer will provide further details.
229If the pre command has a non-zero exit value, the result will be set to
230.Dv RESULT_PREFAIL
231and the
232next step to be performed will be
233.Tn "CLEANUP" .
234In this case and in the case where the command succeeds, the
235.Ar precmd_buf
236will contain the execution log.
237.It
238.Tn "RUN TESTCASE"
239depending on type:
240.Bl -tag -width 2n -compact
241.It "buildonly"
242testcases will get their result set to
243.Dv RESULT_PASS at this point, since the build must have succeeded already.
244.It "userland"
245testcases will get executed in a separate child process, possibly as a
246different user, depending on whether the
247.Ic runas
248option was specified.
249The result will be set to
250.Dv RESULT_TIMEOUT
251if the timeout is reached before the testcase finishes,
252.Dv RESULT_SIGNALLED
253if the testcase dies because of an unhandled signal (often due to crashing),
254.Dv RESULT_NOTRUN
255if the testcase could not be executed,
256.Dv RESULT_FAIL
257if the testcase completes but returns failure or
258.Dv RESULT_PASS
259if the testcase completes and returns success.
260.It "kernel"
261testcases will be executed in kernel space by loading a given module and
262running the testcase entry point function.
263The result will be set to
264.Dv RESULT_NOTRUN
265if the testcase could not be executed.
266Otherwise the result will be set according to the kernel test case to
267one of
268.Dv RESULT_TIMEOUT ,
269.Dv RESULT_FAIL ,
270or
271.Dv RESULT_PASS .
272.El
273The output will be logged separately for stdout and stderr to the
274.Ar stdout_buf
275and
276.Ar stderr_buf
277respectively.
278If the result was
279.Dv RESULT_NOTRUN
280the
281.Ar sysbuf
282will contain more information.
283.It
284.Tn "RUN POST COMMAND"
285if
286.Ic intpost
287or
288.Ic post
289are set.
290If there is an internal driver error, the result will be set to
291.Dv RESULT_POSTFAIL ,
292the next step to be performed will be
293.Tn "CLEANUP"
294and the
295.Ar sysbuf
296buffer will provide further details.
297If the post command has a non-zero exit value, the result will be set to
298.Dv RESULT_POSTFAIL
299and the
300next step to be performed will be
301.Tn "CLEANUP" .
302In this case and in the case where the command succeeds, the
303.Ar postcmd_buf
304will contain the execution log.
305.It
306.Tn "CLEANUP"
307the testcase execution using the command specified by the
308.Ic make
309option or, if not specified,
310.Xr make 1
311with the parameter
312.Dq clean ,
313unless the
314.Ic nobuild
315option was specified.
316If there is an internal driver error the
317.Ar sysbuf
318buffer will contain more information.
319If no internal error occurs, the
320.Ar cleanu_pbuf
321will contain the cleanup log.
322.El
323.Sh FRONTENDS
324The output of
325.Nm
326is in the Apple plist serialized object format.
327This format can be easily parsed by using
328.Xr proplib 3
329when using C.
330Ruby and Python also have parsers for the plist format.
331.Pp
332A frontend for
333.Nm
334parses the intermediate output plist into a more easily readable format
335such as plain text or websites.
336.Pp
337By default
338.Nm
339ships only with the
340.Xr dfr2text 8
341text-based frontend.
342.Sh HOW TO WRITE A TESTCASE
343A userland testcase is a simple program that prints some relevant
344information to stdout and stderr, both of which are captured by the test
345driver, and returns an exit value of 0 if the test passed, or any other
346non-zero exit value to signal a failure.
347The exact exit value is recorded by
348.Nm .
349All signals/exceptions not explicitly caught by the testcase will abort
350the execution of the testcase and the result will be set appropriately and
351the signal number will be recorded.
352.Pp
353A kernel testcase is a simple kernel module that defines two methods:
354a
355.Fn run
356method as well as an optional
357.Fn abort
358method.
359The
360.Fn run
361method will be run from a separate kernel thread.
362The testcase will need to call
363.Xr tbridge 9
364functions to record output and to notify of testcase completion.
365Refer to the
366.Xr tbridge 9
367manual page for more details.
368.Pp
369For all testcase types the build stage is common.
370Every testcase should either have the
371.Ic nobuild
372option set, or have a Makefile or similar in its directory.
373By default
374.Nm
375assumes it is a standard
376.Xr make 1
377Makefile.
378If that is not the case, the
379.Ic build
380option needs to be adjusted accordingly.
381.Sh GENERAL ADVICE ON WRITING TESTCASES
382Test only one thing at a time, it is not good practice to test multiple
383things in the same testcase as it makes it less obvious what's going on.
384.Pp
385Keep it short, simple and well documented on what the requirements are,
386what the preconditions need to be, what exactly is being tested, ideally
387with a reference to a particular bug if that exists, and most importantly
388what the expected outcomes are.
389.Pp
390Make sure your testcase doesn't depend on any other being run previously
391as well as that it won't hinder any other testcase from running.
392This effectively means that your testcase should make no assumptions as
393to what has been run previously unless it has a registered pre-command
394and that the system should be left as found before your testcase.
395.Sh EXAMPLES
396An example runlist can be found under
397.Pa test/testcases/sample.run .
398.Pp
399Several example testcases for both userland and kernel are under
400.Pa test/testcases/sample .
401.Sh SEE ALSO
402.Xr dfr2text 8 ,
403.Xr tbridge 9
404.Sh HISTORY
405The
406.Nm
407utility appeared in
408.Dx 2.13 .
409.Sh AUTHORS
410.An Alex Hornung
411