3 sleepgraph \- Suspend/Resume timing analysis
10 \fBsleepgraph \fP is designed to assist kernel and OS developers
11 in optimizing their linux stack's suspend/resume time. Using a kernel
12 image built with a few extra options enabled, the tool will execute a
13 suspend and capture dmesg and ftrace data until resume is complete.
14 This data is transformed into a device timeline and an optional
15 callgraph to give a detailed view of which devices/subsystems are
16 taking the most time in suspend/resume.
18 If no specific command is given, the default behavior is to initiate
21 Generates output files in subdirectory: suspend-yymmdd-HHMMSS
22 html timeline : <hostname>_<mode>.html
23 raw dmesg file : <hostname>_<mode>_dmesg.txt
24 raw ftrace file : <hostname>_<mode>_ftrace.txt
31 Print the current tool version.
34 Print extra information during execution and analysis.
37 Pull arguments and config options from a file.
40 Mode to initiate for suspend e.g. standby, freeze, mem (default: mem).
43 Overrides the output subdirectory name when running a new test.
44 Use {date}, {time}, {hostname} for current values.
46 e.g. suspend-{hostname}-{date}-{time}
48 \fB-rtcwake \fIt\fR | off
49 Use rtcwake to autoresume after \fIt\fR seconds (default: 15). Set t to "off" to
50 disable rtcwake and require a user keypress to resume.
53 Add the dmesg and ftrace logs to the html output. They will be viewable by
54 clicking buttons in the timeline.
57 By default, if turbostat is found and the requested mode is freeze, sleepgraph
58 will execute the suspend via turbostat and collect data in the timeline log.
59 This option disables the use of turbostat.
62 Export a results table to a text file for parsing.
65 Sync the filesystems before starting the test. This reduces the size of
66 the sys_sync call which happens in the suspend_prepare phase.
68 \fB-rs \fIenable/disable\fR
69 During test, enable/disable runtime suspend for all devices. The test is delayed
70 by 5 seconds to allow runtime suspend changes to occur. The settings are restored
71 after the test is complete.
73 \fB-display \fIon/off/standby/suspend\fR
74 Switch the display to the requested mode for the test using the xset command.
75 This helps maintain the consistency of test data for better comparison.
78 If a wifi connection is available, check that it reconnects after resume. Include
79 the reconnect time in the total resume time calculation and treat wifi timeouts
83 Trace through the wifi reconnect time and include it in the timeline.
86 Add timestamp to each printed output line, accurate to the millisecond.
91 Gzip the trace and dmesg logs to save space. The tool can also read in gzipped
95 Run the timeline over a custom suspend command, e.g. pm-suspend. By default
96 the tool forces suspend via /sys/power/state so this allows testing over
97 an OS's official suspend method. The output file will change to
98 hostname_command.html and will autodetect which suspend mode was triggered.
100 \fB-filter \fI"d1,d2,..."\fR
101 Filter out all but these device callbacks. These strings can be device names
102 or module names. e.g. 0000:00:02.0, ata5, i915, usb, etc.
105 Discard all device callbacks shorter than \fIt\fR milliseconds (default: 0.0).
106 This reduces the html file size as there can be many tiny callbacks which are barely
107 visible. The value is a float: e.g. 0.001 represents 1 us.
110 Add usermode process info into the timeline (default: disabled).
113 Add kernel source calls and threads to the timeline (default: disabled).
116 Run two suspend/resumes back to back (default: disabled).
119 Include \fIt\fR ms delay between multiple test runs (default: 0 ms).
122 Include \fIt\fR ms delay before 1st suspend (default: 0 ms).
124 \fB-postdelay \fIt\fR
125 Include \fIt\fR ms delay after last resume (default: 0 ms).
128 Used for endurance testing. If \fIn\fR is entirely numeric, it's treated as a count:
129 Execute \fIn\fR consecutive tests at \fId\fR second intervals.
130 If \fIn\fR is an integer followed by a "d", "h", or "m", it's treated as a duration:
131 Execute tests continuously over \fIn\fR days, hours, or minutes at \fId\fR second intervals.
132 The outputs will be created in a new subdirectory, for count: suspend-{date}-{time}-xN,
133 for duration: suspend-{date}-{time}-Nm. When the multitest run is done, the \fI-summary\fR
134 command is called automatically to create summary html files for all the data (unless you
135 use \fI-skiphtml\fR). \fI-skiphtml\fR will speed up the testing by not creating timelines
136 or summary html files. You can then run the tool again at a later time with \fI-summary\fR
137 and \fI-genhtml\fR to create the timelines.
140 Abort a -multi run after \fIn\fR consecutive fails. 0 means never abort (default = 0).
143 Run the test and capture the trace logs, but skip the timeline generation.
144 You can generate the html timelines later with \fI-dmesg\fR & \fI-ftrace\fR, or
145 by running \fI-summary\fR and \fI-genhtml\fR.
150 Use ftrace to create device callgraphs (default: disabled). This can produce
151 very large outputs, i.e. 10MB - 100MB.
154 Use ftrace on the top level call: "suspend_devices_and_enter" only (default: disabled).
155 This option implies -f and creates a single callgraph covering all of suspend/resume.
157 \fB-maxdepth \fIlevel\fR
158 limit the callgraph trace depth to \fIlevel\fR (default: 0=all). This is
159 the best way to limit the output size when using callgraphs via -f.
162 pre-expand the callgraph data in the html output (default: disabled)
165 Add functions to be graphed in the timeline from a list in a text file
168 Discard all callgraphs shorter than \fIt\fR milliseconds (default: 0.0).
169 This reduces the html file size as there can be many tiny callgraphs
170 which are barely visible in the timeline.
171 The value is a float: e.g. 0.001 represents 1 us.
173 \fB-cgfilter \fI"func1,func2,..."\fR
174 Reduce callgraph output in the timeline by limiting it certain devices. The
175 argument can be a single device name or a comma delimited list.
178 \fB-cgskip \fIfile\fR
179 Reduce callgraph timeline size by skipping over uninteresting functions
180 in the trace, e.g. printk or console_unlock. The functions listed
181 in this file will show up as empty leaves in the callgraph with only the start/end
182 times displayed. cgskip.txt is used automatically if found in the path, so
183 use "off" to disable completely (default: cgskip.txt)
186 Only show callgraph data for phase \fIp\fR (e.g. suspend_late).
189 In an x2 run, only show callgraph data for test \fIn\fR (e.g. 0 or 1).
192 Number of significant digits in timestamps (0:S, [3:ms], 6:us).
195 Set trace buffer size to N kilo-bytes (default: all of free memory up to 3GB)
199 \fB-summary \fIindir\fR
200 Create a set of summary pages for all tests in \fIindir\fR recursively.
201 Creates summary.html, summary-issues.html, and summary-devices.html in the current folder.
202 summary.html is a table of tests with relevant info sorted by kernel/host/mode,
203 and links to the test html files. It identifies the minimum, maximum, and median
204 suspend and resume times for you with highlights and links in the header.
205 summary-issues.html is a list of kernel issues found in dmesg from all the tests.
206 summary-devices.html is a list of devices and times from all the tests.
208 Use \fI-genhtml\fR to regenerate any tests with missing html.
211 Used with \fI-summary\fR to regenerate any missing html timelines from their
212 dmesg and ftrace logs. This will require a significant amount of time if there
213 are thousands of tests.
216 List available suspend modes.
219 Test to see if the system is able to run this tool. Use this along
220 with any options you intend to use to see if they will work.
223 Print out the contents of the ACPI Firmware Performance Data Table.
226 Print out wifi status and connection details.
228 \fB-xon/-xoff/-xstandby/-xsuspend\fR
229 Test xset by attempting to switch the display to the given mode. This
230 is the same command which will be issued by \fB-display \fImode\fR.
233 Get the current DPMS display mode.
236 Print out system info extracted from BIOS. Reads /dev/mem directly instead of going through dmidecode.
239 Print out the pm settings of all devices which support runtime suspend.
242 Print out all the platform data collected from the system that makes it into the logs.
245 Print the list of ftrace functions currently being captured. Functions
246 that are not available as symbols in the current kernel are shown in red.
247 By default, the tool traces a list of important suspend/resume functions
248 in order to better fill out the timeline. If the user has added their own
249 with -fadd they will also be checked.
252 Print all ftrace functions capable of being captured. These are all the
253 possible values you can add to trace via the -fadd argument.
256 \fB-ftrace \fIfile\fR
257 Create HTML output from an existing ftrace file.
260 Create HTML output from an existing dmesg file.
263 .SS "simple commands"
264 Check which suspend modes are currently supported.
266 \f(CW$ sleepgraph -modes\fR
268 Read the Firmware Performance Data Table (FPDT)
270 \f(CW$ sudo sleepgraph -fpdt\fR
272 Print out the current USB power topology
274 \f(CW$ sleepgraph -usbtopo
276 Verify that you can run a command with a set of arguments
278 \f(CW$ sudo sleepgraph -f -rtcwake 30 -status
280 Generate a summary of all timelines in a particular folder.
282 \f(CW$ sleepgraph -summary ~/workspace/myresults/\fR
285 .SS "capturing basic timelines"
286 Execute a mem suspend with a 15 second wakeup. Include the logs in the html.
288 \f(CW$ sudo sleepgraph -rtcwake 15 -addlogs\fR
290 Execute a standby with a 15 second wakeup. Change the output folder name.
292 \f(CW$ sudo sleepgraph -m standby -rtcwake 15 -o "standby-{host}-{date}-{time}"\fR
294 Execute a freeze with no wakeup (require keypress). Change output folder name.
296 \f(CW$ sudo sleepgraph -m freeze -rtcwake off -o "freeze-{hostname}-{date}-{time}"\fR
299 .SS "capturing advanced timelines"
300 Execute a suspend & include dev mode source calls, limit callbacks to 5ms or larger.
302 \f(CW$ sudo sleepgraph -m mem -rtcwake 15 -dev -mindev 5\fR
304 Run two suspends back to back, include a 500ms delay before, after, and in between runs.
306 \f(CW$ sudo sleepgraph -m mem -rtcwake 15 -x2 -predelay 500 -x2delay 500 -postdelay 500\fR
308 Execute a suspend using a custom command.
310 \f(CW$ sudo sleepgraph -cmd "echo mem > /sys/power/state" -rtcwake 15\fR
313 .SS "endurance testing using -multi"
315 Do a batch run of 10 freezes with 30 seconds delay between runs.
317 \f(CW$ sudo sleepgraph -m freeze -rtcwake 15 -multi 10 30\fR
319 Do a batch run of freezes for 24 hours.
321 \f(CW$ sudo sleepgraph -m freeze -rtcwake 15 -multi 24h 0\fR
323 .SS "adding callgraph data"
324 Add device callgraphs. Limit the trace depth and only show callgraphs 10ms or larger.
326 \f(CW$ sudo sleepgraph -m mem -rtcwake 15 -f -maxdepth 5 -mincg 10\fR
328 Capture a full callgraph across all suspend, then filter the html by a single phase.
330 \f(CW$ sudo sleepgraph -m mem -rtcwake 15 -f\fR
332 \f(CW$ sleepgraph -dmesg host_mem_dmesg.txt -ftrace host_mem_ftrace.txt -f -cgphase resume
335 .SS "rebuild timeline from logs"
337 Rebuild the html from a previous run's logs, using the same options.
339 \f(CW$ sleepgraph -dmesg dmesg.txt -ftrace ftrace.txt -callgraph\fR
341 Rebuild the html with different options.
343 \f(CW$ sleepgraph -dmesg dmesg.txt -ftrace ftrace.txt -addlogs -srgap\fR
350 Written by Todd Brandt <todd.e.brandt@linux.intel.com>