xref: /aosp_15_r20/external/oboe/apps/OboeTester/docs/AutomatedTesting.md (revision 05767d913155b055644481607e6fa1e35e2fe72c)
1[Home](README.md)
2
3# Automated Testing
4
5OboeTester can be used to measure the round trip latency and glitches.
6It can be launched from a shell script by using an Android Intent.
7
8Before running the app from an Intent, it should be launched manually and a Round Trip Latency test run. Then you can give permission for using the microphone to record the looped back sound, and give permission to write to external storage for saving the test result.
9
10## Requirements
11
12All tests require:
13
14* host computer
15* ADB installed
16* ADB USB cable
17
18The latency, glitch and data_paths tests also need:
19
20* [loopback adapter](https://source.android.com/devices/audio/latency/loopback)
21* a 3.5 mm jack on the phone*
22
23\* If you don't have a 3.5 mm jack then you can use a USB-C to 3.5mm adapter.
24In order to use ADB at the same time you will also need a USB switching device
25or  use ADB/Wifi.
26
27## Start App from Intent
28
29The app can be started by sending a Start comment to the OboeTester class.
30The app will run and the results will be written to a file.
31
32    adb shell am start -n com.mobileer.oboetester/.MainActivity {parameters}
33
34String parameters are sent using:
35
36    --es {parameterName} {parameterValue}
37
38For example:
39
40    --es test latency
41
42Integer parameters are sent using:
43
44    --ei {parameterName} {parameterValue}
45
46For example:
47
48    --ei buffer_bursts 8
49
50Boolean parameters are sent using:
51
52    --ez {parameterName} {parameterValue}
53
54For example:
55
56    --ez use_input_presets false
57
58## Parameters
59
60There are two required parameters for all tests:
61
62    --es test {latency, glitch, data_paths, input, output}
63            The "latency" test will perform a Round Trip Latency test.
64            It will request EXCLUSIVE mode for minimal latency.
65            The "glitch" test will perform a single Glitch test.
66            The "data_paths" test will verify input and output streams in many possible configurations.
67            The "input" test will open and start an input stream.
68            The "output" test will open and start an output stream.
69
70    --es file {name of resulting file}
71
72The file will be stored in a directory that can be written by OboeTester without any special permissions.
73This is typically "/storage/emulated/0/Android/data/com.mobileer.oboetester/files/".
74
75There are some optional parameter in common for all tests:
76
77    --ef volume             {volume} // normalized volume in the range of 0.0 to 1.0
78    --es volume_type        {"accessibility", "alarm", "dtmf", "music", "notification", "ring", "system", "voice_call"}
79                            Stream type for the setStreamVolume() call. Default is "music".
80    --ez background         {"true", 1, "false", 0} // if true then Oboetester will continue to run in the background
81
82There are several optional parameters in common for glitch, latency, input, and output tests:
83
84    --ei buffer_bursts      {bursts}     // number of bursts in the buffer, 2 for "double buffered"
85    --es in_api             {"unspecified", "opensles", "aaudio"}  // native input API, default is "unspecified"
86    --es out_api            {"unspecified", "opensles", "aaudio"}  // native output API, default is "unspecified"
87    --es in_channel_mask    {"mono", "stereo", "2.1", "tri", "triBack", "3.1", "2.0.2", "2.1.2", "3.0.2", "3.1.2", "quad", "quadSide", "surround", "penta", "5.1", "5.1Side", "6.1", "7.1", "5.1.2", "5.1.4", "7.1.2", "7.1.4", "9.1.4", "9.1.6", "frontBack"}
88    --es out_channel_mask    {"mono", "stereo", "2.1", "tri", "triBack", "3.1", "2.0.2", "2.1.2", "3.0.2", "3.1.2", "quad", "quadSide", "surround", "penta", "5.1", "5.1Side", "6.1", "7.1", "5.1.2", "5.1.4", "7.1.2", "7.1.4", "9.1.4", "9.1.6", "frontBack"}
89    --ei in_channels        {samples}    // number of input channels, default is 2. This is ignored if in_channel_mask is set.
90    --ei out_channels       {samples}    // number of output channels, default is 2. This is ignored if out_channel_mask is set.
91    --ei sample_rate        {hertz}
92    --es in_perf            {"none", "lowlat", "powersave"}  // input performance mode, default is "lowlat"
93    --es out_perf           {"none", "lowlat", "powersave"}  // output performance mode, default is "lowlat"
94    --es out_usage          {"media", "voice_communication", "alarm", "notification", "game"} // default is media
95    --es in_sharing         {"shared", "exclusive"} // input sharing mode, default is "exclusive"
96    --es out_sharing        {"shared", "exclusive"} // output sharing mode, default is "exclusive"
97    --ez in_use_mmap        {"true", 1, "false", 0} // if true then MMAP is allowed, if false then MMAP will be disabled
98    --ez out_use_mmap       {"true", 1, "false", 0} // if true then MMAP is allowed, if false then MMAP will be disabled
99
100There are some optional parameters in common for glitch, input, and output tests:
101
102    --ei duration           {seconds}    // glitch test duration, default is 10 seconds
103
104There are several optional parameters for just the "glitch" test:
105
106    --ef tolerance          {tolerance}  // amount of deviation from expected that is considered a glitch
107                                         // Range of tolerance is 0.0 to 1.0. Default is 0.1. Note use of "-ef".
108                            // input preset, default is "voicerec"
109    --es in_preset          ("generic", "camcorder", "voicerec", "voicecomm", "unprocessed", "performance"}
110
111There are several optional parameters for just the "data_paths" test. Note the  Note the use of "-ez" for the boolean parameters.
112
113    --ez use_input_presets  {"true", 1, "false", 0}  // Whether to test various input presets.
114    --ez use_all_sample_rates {"true", 1, "false", 0}  // Whether to test all sample rates. Note use of "-ez". Default is false
115    --ei single_test_index  {testId}  // Index for testing one specific test
116
117These parameters are used with the "data_paths" test starting with v2.5.11.
118
119    --ez use_input_channel_masks {"true", 1, "false", 0}  // Whether to test the reported input channel MASKS. Default is false.
120    --ez use_all_channel_counts {"true", 1, "false", 0}   // Whether to test all the supported channel COUNTS. Default is true.
121    --ei output_channel_masks_level {0, 1, 2}    // Whether to test NONE=0, SOME=1, or ALL=2 channel masks. Default is false.
122
123These parameters were used with the "data_paths" test prior to v2.5.11.
124
125    --ez use_input_devices  {"true", 1, "false", 0}  // Whether to test various input devices.
126    --ez use_output_devices {"true", 1, "false", 0}  // Whether to test various output devices.
127    --ez use_all_output_channel_masks {"true", 1, "false", 0}  // Whether to test all output channel masks. Default is false
128
129There are some optional parameters for just the "output" test:
130
131    --es signal_type        {sine, sawtooth, freq_sweep, pitch_sweep, white_noise} // type of sound to play, default is sine
132
133For example, a complete command for a "latency" test might be:
134
135    adb shell am start -n com.mobileer.oboetester/.MainActivity \
136        --es test latency \
137        --ei buffer_bursts 2 \
138        --ef volume 0.8 \
139        --es volume_type music \
140        --ei buffer_bursts 2 \
141        --ei out_channels 1 \
142        --es out_usage game \
143        --es file latency20230608.txt
144
145or for a "glitch" test:
146
147    adb shell am start -n com.mobileer.oboetester/.MainActivity \
148        --es test glitch \
149        --es in_perf lowlat \
150        --es out_perf lowlat \
151        --es in_sharing exclusive \
152        --es out_sharing exclusive \
153        --ei buffer_bursts 2 \
154        --ei sample_rate 48000 \
155        --ef tolerance 0.123 \
156        --ei in_channels 2 \
157        --es file glitch20230608.txt
158
159or for a "data_paths" test:
160
161    adb shell am start -n com.mobileer.oboetester/.MainActivity \
162        --es test data_paths \
163        --ez use_input_presets true \
164        --ez use_input_devices false \
165        --ez use_output_devices true \
166        --es file datapaths20230608.txt
167
168## Interpreting Test Results
169
170Test results are simple files with "name = value" pairs.
171After running the test you can determine where the results file was written by entering:
172
173    adb logcat | grep EXTFILE
174
175The test results can be obtained using adb pull. For example:
176
177    adb pull /storage/emulated/0/Android/data/com.mobileer.oboetester/files/glitch20230608.txt .
178
179The beginning of the report is common to latency and glitch tests:
180
181```
182build.fingerprint = google/cheetah/cheetah:14/MASTER/eng.philbu.20230518.172104:userdebug/dev-keys
183test.version = 2.5.1
184test.version.code = 72
185time.millis = 1686156503523
186in.channels = 2
187in.perf = ll
188in.preset = voicerec
189in.sharing = ex
190in.api = aaudio
191in.rate = 48000
192in.device = 22
193in.mmap = yes
194in.rate.conversion.quality = 0
195in.hardware.channels = 2
196in.hardware.sampleRate = 48000
197in.hardware.format = i32
198in.burst.frames = 96
199in.xruns = 0
200out.channels = 1
201out.perf = ll
202out.usage = game
203out.contentType = music
204out.sharing = ex
205out.api = aaudio
206out.rate = 48000
207out.device = 3
208out.mmap = yes
209out.rate.conversion.quality = 0
210out.hardware.channels = 2
211out.hardware.sampleRate = 48000
212out.hardware.format = float
213out.burst.frames = 96
214out.buffer.size.frames = 192
215out.buffer.capacity.frames = 1920
216out.xruns = 0
217```
218
219### Latency Report
220
221Each test also adds specific value. For "latency". If the test fails then some values will be unavailable.
222
223Here is a report from a good test. The '#' comments were added for this document and are not in the report.
224
225```
226confidence = 0.892          # quality of the latency result between 0.0 and 1.0, higher is better
227result.text = OK            # text equivalent of the result
228latency.msec = 23.27        # round trip latency in milliseconds
229latency.frames = 1117       # round trip latency in frames
230latency.empty.msec = 19.27  # round trip latency if the top output buffer was empty
231latency.empty.frames = 925  # same but translated to frames
232rms.signal = 0.03142        # Root Mean Square of the signal, if it can be detected
233rms.noise = 0.00262         # Root Mean Square of the background noise before the signal is detected
234correlation = 0.975         # raw normalized cross-correlation peak
235timestamp.latency.msec = 10.35 # latency based on timestamps
236timestamp.latency.mad = 0.05   # Mean absolute deviation
237timestamp.latency.count = 12   # number of measurements
238reset.count = 1             # number of times the full duplex stream input underflowed and had to resynchronize
239result = 0                  # 0 or a negative error
240```
241
242Here is a report from a test that failed because the output was muted. Note the latency.msec is
243missing because it could not be measured.
244
245    rms.signal = 0.00000
246    rms.noise = 0.00048
247    reset.count = 3
248    result = -96
249    result.text = ERROR_CONFIDENCE
250    confidence =  0.009
251
252### Glitch Report
253
254Here is a report from a good test. The '#' comments were added for this document and are not in the report.
255
256    tolerance = 0.123
257    state = LOCKED
258    unlocked.frames = 2528   # frames spent trying to lock onto the signal
259    locked.frames = 384084   # frames spent locked onto a good signal with no glitches
260    glitch.frames = 0        # frames spent glitching or recovering from a glitch
261    reset.count = 208        # number of times the full duplex stream input underflowed and had to resynchronize
262    peak.amplitude = 0.057714  # peak amplitude of the input signal, between 0.0 and 1.0
263    signal.noise.ratio.db =  96.3
264    time.total =     9.96 seconds  # close to your specified duration
265    time.no.glitches =     9.96    # time we have been running with no glitches
266    max.time.no.glitches =     9.96 # max time with no glitches
267    glitch.count = 0               # number of glitch events, actual number may be higher if close together
268
269Here is a report from a test that failed because the output was muted. Note the glitch.count is
270missing because it could not be measured.
271
272    state = WAITING_FOR_SIGNAL
273    unlocked.frames = 0
274    locked.frames = 0
275    glitch.frames = 0
276    reset.count = 1
277    time.total =     9.95 seconds
278
279### Data Paths Report
280
281The report first goes through the info about the specific device before going through input preset tests,
282input devices tests, and output tests.
283Each will show the specific configuration of a test before showing whether it passed or failed.
284At the end of the report, an analysis of the failed tests will be given
285followed by the number of passed, failed, and skipped tests.
286