OLD | NEW |
| (Empty) |
1 # Copyright (c) 2013 The Chromium Authors. All rights reserved. | |
2 # Use of this source code is governed by a BSD-style license that can be | |
3 # found in the LICENSE file. | |
4 | |
5 """Config file for Run Performance Test Bisect Tool | |
6 | |
7 This script is intended for use by anyone that wants to run a remote bisection | |
8 on a range of revisions to look for a performance regression. Modify the config | |
9 below and add the revision range, performance command, and metric. You can then | |
10 run a git try <bot>. | |
11 | |
12 Changes to this file should never be submitted. | |
13 | |
14 Args: | |
15 'command': This is the full command line to pass to the | |
16 bisect-perf-regression.py script in order to execute the test. | |
17 'good_revision': An svn or git revision where the metric hadn't regressed yet. | |
18 'bad_revision': An svn or git revision sometime after the metric had | |
19 regressed. | |
20 'metric': The name of the metric to parse out from the results of the | |
21 performance test. You can retrieve the metric by looking at the stdio of | |
22 the performance test. Look for lines of the format: | |
23 | |
24 RESULT <graph>: <trace>= <value> <units> | |
25 | |
26 The metric name is "<graph>/<trace>". | |
27 'repeat_count': The number of times to repeat the performance test. | |
28 'max_time_minutes': The script will attempt to run the performance test | |
29 "repeat_count" times, unless it exceeds "max_time_minutes". | |
30 'truncate_percent': Discard the highest/lowest % values from performance test. | |
31 | |
32 Sample config: | |
33 | |
34 config = { | |
35 'command': './tools/perf/run_measurement --browser=release blink_perf third_pa
rty/WebKit/PerformanceTests/Layout/floats_50_100.html', | |
36 'good_revision': '233015', | |
37 'bad_revision': '233115', | |
38 'metric': 'floats_50_100/floats_50_100', | |
39 'repeat_count': '20', | |
40 'max_time_minutes': '20', | |
41 'truncate_percent': '25', | |
42 } | |
43 | |
44 On Windows: | |
45 - If you're calling a python script you will need to add "python" to | |
46 the command: | |
47 | |
48 config = { | |
49 'command': 'python tools/perf/run_measurement -v --browser=release kraken', | |
50 'good_revision': '185319', | |
51 'bad_revision': '185364', | |
52 'metric': 'Total/Total', | |
53 'repeat_count': '20', | |
54 'max_time_minutes': '20', | |
55 'truncate_percent': '25', | |
56 } | |
57 | |
58 | |
59 On ChromeOS: | |
60 - Script accepts either ChromeOS versions, or unix timestamps as revisions. | |
61 - You don't need to specify --identity and --remote, they will be added to | |
62 the command using the bot's BISECT_CROS_IP and BISECT_CROS_BOARD values. | |
63 | |
64 config = { | |
65 'command': './tools/perf/run_measurement -v '\ | |
66 '--browser=cros-chrome-guest '\ | |
67 'dromaeo tools/perf/page_sets/dromaeo/jslibstylejquery.json', | |
68 'good_revision': '4086.0.0', | |
69 'bad_revision': '4087.0.0', | |
70 'metric': 'jslib/jslib', | |
71 'repeat_count': '20', | |
72 'max_time_minutes': '20', | |
73 'truncate_percent': '25', | |
74 } | |
75 | |
76 """ | |
77 | |
78 config = { | |
79 'command': '', | |
80 'good_revision': '', | |
81 'bad_revision': '', | |
82 'metric': '', | |
83 'repeat_count':'', | |
84 'max_time_minutes': '', | |
85 'truncate_percent':'', | |
86 } | |
87 | |
88 # Workaround git try issue, see crbug.com/257689 | |
OLD | NEW |