platform/external/ffmpeg.git - Unnamed repository; edit this file 'description' to name the repository.

1 \input texinfo @c -*- texinfo -*-
2 @documentencoding UTF-8
3
4 @settitle FFmpeg Automated Testing Environment
5 @titlepage
6 @center @titlefont{FFmpeg Automated Testing Environment}
7 @end titlepage
8
9 @node Top
10 @top
11
12 @contents
13
14 @chapter Introduction
15
16 FATE is an extended regression suite on the client-side and a means
17 for results aggregation and presentation on the server-side.
18
19 The first part of this document explains how you can use FATE from
20 your FFmpeg source directory to test your ffmpeg binary. The second
21 part describes how you can run FATE to submit the results to FFmpeg's
22 FATE server.
23
24 In any way you can have a look at the publicly viewable FATE results
25 by visiting this website:
26
27 @url{http://fate.ffmpeg.org/}
28
29 This is especially recommended for all people contributing source
30 code to FFmpeg, as it can be seen if some test on some platform broke
31 with their recent contribution. This usually happens on the platforms
32 the developers could not test on.
33
34 The second part of this document describes how you can run FATE to
35 submit your results to FFmpeg's FATE server. If you want to submit your
36 results be sure to check that your combination of CPU, OS and compiler
37 is not already listed on the above mentioned website.
38
39 In the third part you can find a comprehensive listing of FATE makefile
40 targets and variables.
41
42
43 @chapter Using FATE from your FFmpeg source directory
44
45 If you want to run FATE on your machine you need to have the samples
46 in place. You can get the samples via the build target fate-rsync.
47 Use this command from the top-level source directory:
48
49 @example
50 make fate-rsync SAMPLES=fate-suite/
51 make fate       SAMPLES=fate-suite/
52 @end example
53
54 The above commands set the samples location by passing a makefile
55 variable via command line. It is also possible to set the samples
56 location at source configuration time by invoking configure with
57 @option{--samples=<path to the samples directory>}. Afterwards you can
58 invoke the makefile targets without setting the @var{SAMPLES} makefile
59 variable. This is illustrated by the following commands:
60
61 @example
62 ./configure --samples=fate-suite/
63 make fate-rsync
64 make fate
65 @end example
66
67 Yet another way to tell FATE about the location of the sample
68 directory is by making sure the environment variable FATE_SAMPLES
69 contains the path to your samples directory. This can be achieved
70 by e.g. putting that variable in your shell profile or by setting
71 it in your interactive session.
72
73 @example
74 FATE_SAMPLES=fate-suite/ make fate
75 @end example
76
77 @float NOTE
78 Do not put a '~' character in the samples path to indicate a home
79 directory. Because of shell nuances, this will cause FATE to fail.
80 @end float
81
82 To use a custom wrapper to run the test, pass @option{--target-exec} to
83 @command{configure} or set the @var{TARGET_EXEC} Make variable.
84
85
86 @chapter Submitting the results to the FFmpeg result aggregation server
87
88 To submit your results to the server you should run fate through the
89 shell script @file{tests/fate.sh} from the FFmpeg sources. This script needs
90 to be invoked with a configuration file as its first argument.
91
92 @example
93 tests/fate.sh /path/to/fate_config
94 @end example
95
96 A configuration file template with comments describing the individual
97 configuration variables can be found at @file{doc/fate_config.sh.template}.
98
99 @ifhtml
100 The mentioned configuration template is also available here:
101 @verbatiminclude fate_config.sh.template
102 @end ifhtml
103
104 Create a configuration that suits your needs, based on the configuration
105 template. The @env{slot} configuration variable can be any string that is not
106 yet used, but it is suggested that you name it adhering to the following
107 pattern @samp{@var{arch}-@var{os}-@var{compiler}-@var{compiler version}}. The
108 configuration file itself will be sourced in a shell script, therefore all
109 shell features may be used. This enables you to setup the environment as you
110 need it for your build.
111
112 For your first test runs the @env{fate_recv} variable should be empty or
113 commented out. This will run everything as normal except that it will omit
114 the submission of the results to the server. The following files should be
115 present in $workdir as specified in the configuration file:
116
117 @itemize
118     @item configure.log
119     @item compile.log
120     @item test.log
121     @item report
122     @item version
123 @end itemize
124
125 When you have everything working properly you can create an SSH key pair
126 and send the public key to the FATE server administrator who can be contacted
127 at the email address @email{fate-admin@@ffmpeg.org}.
128
129 Configure your SSH client to use public key authentication with that key
130 when connecting to the FATE server. Also do not forget to check the identity
131 of the server and to accept its host key. This can usually be achieved by
132 running your SSH client manually and killing it after you accepted the key.
133 The FATE server's fingerprint is:
134
135 @table @samp
136 @item RSA
137    d3:f1:83:97:a4:75:2b:a6:fb:d6:e8:aa:81:93:97:51
138 @item ECDSA
139    76:9f:68:32:04:1e:d5:d4:ec:47:3f:dc:fc:18:17:86
140 @end table
141
142 If you have problems connecting to the FATE server, it may help to try out
143 the @command{ssh} command with one or more @option{-v} options. You should
144 get detailed output concerning your SSH configuration and the authentication
145 process.
146
147 The only thing left is to automate the execution of the fate.sh script and
148 the synchronisation of the samples directory.
149
150
151 @chapter FATE makefile targets and variables
152
153 @section Makefile targets
154
155 @table @option
156 @item fate-rsync
157 Download/synchronize sample files to the configured samples directory.
158
159 @item fate-list
160 Will list all fate/regression test targets.
161
162 @item fate
163 Run the FATE test suite (requires the fate-suite dataset).
164 @end table
165
166 @section Makefile variables
167
168 @table @env
169 @item V
170 Verbosity level, can be set to 0, 1 or 2.
171     @itemize
172         @item 0: show just the test arguments
173         @item 1: show just the command used in the test
174         @item 2: show everything
175     @end itemize
176
177 @item SAMPLES
178 Specify or override the path to the FATE samples at make time, it has a
179 meaning only while running the regression tests.
180
181 @item THREADS
182 Specify how many threads to use while running regression tests, it is
183 quite useful to detect thread-related regressions.
184
185 @item THREAD_TYPE
186 Specify which threading strategy test, either @samp{slice} or @samp{frame},
187 by default @samp{slice+frame}
188
189 @item CPUFLAGS
190 Specify CPU flags.
191
192 @item TARGET_EXEC
193 Specify or override the wrapper used to run the tests.
194 The @env{TARGET_EXEC} option provides a way to run FATE wrapped in
195 @command{valgrind}, @command{qemu-user} or @command{wine} or on remote targets
196 through @command{ssh}.
197
198 @item GEN
199 Set to @samp{1} to generate the missing or mismatched references.
200
201 @item HWACCEL
202 Specify which hardware acceleration to use while running regression tests,
203 by default @samp{none} is used.
204
205 @end table
206
207 @section Examples
208
209 @example
210 make V=1 SAMPLES=/var/fate/samples THREADS=2 CPUFLAGS=mmx fate
211 @end example
212
1	\input texinfo @c -- texinfo --
2	@documentencoding UTF-8
3
4	@settitle FFmpeg Automated Testing Environment
5	@titlepage
6	@center @titlefont{FFmpeg Automated Testing Environment}
7	@end titlepage
8
9	@node Top
10	@top
11
12	@contents
13
14	@chapter Introduction
15
16	FATE is an extended regression suite on the client-side and a means
17	for results aggregation and presentation on the server-side.
18
19	The first part of this document explains how you can use FATE from
20	your FFmpeg source directory to test your ffmpeg binary. The second
21	part describes how you can run FATE to submit the results to FFmpeg's
22	FATE server.
23
24	In any way you can have a look at the publicly viewable FATE results
25	by visiting this website:
26
27	@url{http://fate.ffmpeg.org/}
28
29	This is especially recommended for all people contributing source
30	code to FFmpeg, as it can be seen if some test on some platform broke
31	with their recent contribution. This usually happens on the platforms
32	the developers could not test on.
33
34	The second part of this document describes how you can run FATE to
35	submit your results to FFmpeg's FATE server. If you want to submit your
36	results be sure to check that your combination of CPU, OS and compiler
37	is not already listed on the above mentioned website.
38
39	In the third part you can find a comprehensive listing of FATE makefile
40	targets and variables.
41
42
43	@chapter Using FATE from your FFmpeg source directory
44
45	If you want to run FATE on your machine you need to have the samples
46	in place. You can get the samples via the build target fate-rsync.
47	Use this command from the top-level source directory:
48
49	@example
50	make fate-rsync SAMPLES=fate-suite/
51	make fate SAMPLES=fate-suite/
52	@end example
53
54	The above commands set the samples location by passing a makefile
55	variable via command line. It is also possible to set the samples
56	location at source configuration time by invoking configure with
57	@option{--samples=<path to the samples directory>}. Afterwards you can
58	invoke the makefile targets without setting the @var{SAMPLES} makefile
59	variable. This is illustrated by the following commands:
60
61	@example
62	./configure --samples=fate-suite/
63	make fate-rsync
64	make fate
65	@end example
66
67	Yet another way to tell FATE about the location of the sample
68	directory is by making sure the environment variable FATE_SAMPLES
69	contains the path to your samples directory. This can be achieved
70	by e.g. putting that variable in your shell profile or by setting
71	it in your interactive session.
72
73	@example
74	FATE_SAMPLES=fate-suite/ make fate
75	@end example
76
77	@float NOTE
78	Do not put a '~' character in the samples path to indicate a home
79	directory. Because of shell nuances, this will cause FATE to fail.
80	@end float
81
82	To use a custom wrapper to run the test, pass @option{--target-exec} to
83	@command{configure} or set the @var{TARGET_EXEC} Make variable.
84
85
86	@chapter Submitting the results to the FFmpeg result aggregation server
87
88	To submit your results to the server you should run fate through the
89	shell script @file{tests/fate.sh} from the FFmpeg sources. This script needs
90	to be invoked with a configuration file as its first argument.
91
92	@example
93	tests/fate.sh /path/to/fate_config
94	@end example
95
96	A configuration file template with comments describing the individual
97	configuration variables can be found at @file{doc/fate_config.sh.template}.
98
99	@ifhtml
100	The mentioned configuration template is also available here:
101	@verbatiminclude fate_config.sh.template
102	@end ifhtml
103
104	Create a configuration that suits your needs, based on the configuration
105	template. The @env{slot} configuration variable can be any string that is not
106	yet used, but it is suggested that you name it adhering to the following
107	pattern @samp{@var{arch}-@var{os}-@var{compiler}-@var{compiler version}}. The
108	configuration file itself will be sourced in a shell script, therefore all
109	shell features may be used. This enables you to setup the environment as you
110	need it for your build.
111
112	For your first test runs the @env{fate_recv} variable should be empty or
113	commented out. This will run everything as normal except that it will omit
114	the submission of the results to the server. The following files should be
115	present in $workdir as specified in the configuration file:
116
117	@itemize
118	@item configure.log
119	@item compile.log
120	@item test.log
121	@item report
122	@item version
123	@end itemize
124
125	When you have everything working properly you can create an SSH key pair
126	and send the public key to the FATE server administrator who can be contacted
127	at the email address @email{fate-admin@@ffmpeg.org}.
128
129	Configure your SSH client to use public key authentication with that key
130	when connecting to the FATE server. Also do not forget to check the identity
131	of the server and to accept its host key. This can usually be achieved by
132	running your SSH client manually and killing it after you accepted the key.
133	The FATE server's fingerprint is:
134
135	@table @samp
136	@item RSA
137	d3:f1:83:97:a4:75:2b:a6:fb:d6:e8:aa:81:93:97:51
138	@item ECDSA
139	76:9f:68:32:04:1e:d5:d4:ec:47:3f:dc:fc:18:17:86
140	@end table
141
142	If you have problems connecting to the FATE server, it may help to try out
143	the @command{ssh} command with one or more @option{-v} options. You should
144	get detailed output concerning your SSH configuration and the authentication
145	process.
146
147	The only thing left is to automate the execution of the fate.sh script and
148	the synchronisation of the samples directory.
149
150
151	@chapter FATE makefile targets and variables
152
153	@section Makefile targets
154
155	@table @option
156	@item fate-rsync
157	Download/synchronize sample files to the configured samples directory.
158
159	@item fate-list
160	Will list all fate/regression test targets.
161
162	@item fate
163	Run the FATE test suite (requires the fate-suite dataset).
164	@end table
165
166	@section Makefile variables
167
168	@table @env
169	@item V
170	Verbosity level, can be set to 0, 1 or 2.
171	@itemize
172	@item 0: show just the test arguments
173	@item 1: show just the command used in the test
174	@item 2: show everything
175	@end itemize
176
177	@item SAMPLES
178	Specify or override the path to the FATE samples at make time, it has a
179	meaning only while running the regression tests.
180
181	@item THREADS
182	Specify how many threads to use while running regression tests, it is
183	quite useful to detect thread-related regressions.
184
185	@item THREAD_TYPE
186	Specify which threading strategy test, either @samp{slice} or @samp{frame},
187	by default @samp{slice+frame}
188
189	@item CPUFLAGS
190	Specify CPU flags.
191
192	@item TARGET_EXEC
193	Specify or override the wrapper used to run the tests.
194	The @env{TARGET_EXEC} option provides a way to run FATE wrapped in
195	@command{valgrind}, @command{qemu-user} or @command{wine} or on remote targets
196	through @command{ssh}.
197
198	@item GEN
199	Set to @samp{1} to generate the missing or mismatched references.
200
201	@item HWACCEL
202	Specify which hardware acceleration to use while running regression tests,
203	by default @samp{none} is used.
204
205	@end table
206
207	@section Examples
208
209	@example
210	make V=1 SAMPLES=/var/fate/samples THREADS=2 CPUFLAGS=mmx fate
211	@end example
212