xref: /linux/tools/perf/Documentation/perf-diff.txt (revision 9f3926e08c26607a0dd5b1bc8a8aa1d03f72fcdc)
1perf-diff(1)
2============
3
4NAME
5----
6perf-diff - Read perf.data files and display the differential profile
7
8SYNOPSIS
9--------
10[verse]
11'perf diff' [baseline file] [data file1] [[data file2] ... ]
12
13DESCRIPTION
14-----------
15This command displays the performance difference amongst two or more perf.data
16files captured via perf record.
17
18If no parameters are passed it will assume perf.data.old and perf.data.
19
20The differential profile is displayed only for events matching both
21specified perf.data files.
22
23If no parameters are passed the samples will be sorted by dso and symbol.
24As the perf.data files could come from different binaries, the symbols addresses
25could vary. So perf diff is based on the comparison of the files and
26symbols name.
27
28OPTIONS
29-------
30-D::
31--dump-raw-trace::
32        Dump raw trace in ASCII.
33
34--kallsyms=<file>::
35        kallsyms pathname
36
37-m::
38--modules::
39        Load module symbols. WARNING: use only with -k and LIVE kernel
40
41-d::
42--dsos=::
43	Only consider symbols in these dsos. CSV that understands
44	file://filename entries.  This option will affect the percentage
45	of the Baseline/Delta column.  See --percentage for more info.
46
47-C::
48--comms=::
49	Only consider symbols in these comms. CSV that understands
50	file://filename entries.  This option will affect the percentage
51	of the Baseline/Delta column.  See --percentage for more info.
52
53-S::
54--symbols=::
55	Only consider these symbols. CSV that understands
56	file://filename entries.  This option will affect the percentage
57	of the Baseline/Delta column.  See --percentage for more info.
58
59-s::
60--sort=::
61	Sort by key(s): pid, comm, dso, symbol, cpu, parent, srcline.
62	Please see description of --sort in the perf-report man page.
63
64-t::
65--field-separator=::
66
67	Use a special separator character and don't pad with spaces, replacing
68	all occurrences of this separator in symbol names (and other output)
69	with a '.' character, that thus it's the only non valid separator.
70
71-v::
72--verbose::
73	Be verbose, for instance, show the raw counts in addition to the
74	diff.
75
76-q::
77--quiet::
78	Do not show any message.  (Suppress -v)
79
80-f::
81--force::
82        Don't do ownership validation.
83
84--symfs=<directory>::
85        Look for files with symbols relative to this directory.
86
87-b::
88--baseline-only::
89        Show only items with match in baseline.
90
91-c::
92--compute::
93        Differential computation selection - delta, ratio, wdiff, delta-abs
94        (default is delta-abs).  Default can be changed using diff.compute
95        config option.  See COMPARISON METHODS section for more info.
96
97-p::
98--period::
99        Show period values for both compared hist entries.
100
101-F::
102--formula::
103        Show formula for given computation.
104
105-o::
106--order::
107       Specify compute sorting column number.  0 means sorting by baseline
108       overhead and 1 (default) means sorting by computed value of column 1
109       (data from the first file other base baseline).  Values more than 1
110       can be used only if enough data files are provided.
111       The default value can be set using the diff.order config option.
112
113--percentage::
114	Determine how to display the overhead percentage of filtered entries.
115	Filters can be applied by --comms, --dsos and/or --symbols options.
116
117	"relative" means it's relative to filtered entries only so that the
118	sum of shown entries will be always 100%.  "absolute" means it retains
119	the original value before and after the filter is applied.
120
121--time::
122	Analyze samples within given time window. It supports time
123	percent with multiple time ranges. Time string is 'a%/n,b%/m,...'
124	or 'a%-b%,c%-%d,...'.
125
126	For example:
127
128	Select the second 10% time slice to diff:
129
130	  perf diff --time 10%/2
131
132	Select from 0% to 10% time slice to diff:
133
134	  perf diff --time 0%-10%
135
136	Select the first and the second 10% time slices to diff:
137
138	  perf diff --time 10%/1,10%/2
139
140	Select from 0% to 10% and 30% to 40% slices to diff:
141
142	  perf diff --time 0%-10%,30%-40%
143
144	It also supports analyzing samples within a given time window
145	<start>,<stop>. Times have the format seconds.nanoseconds. If 'start'
146	is not given (i.e. time string is ',x.y') then analysis starts at
147	the beginning of the file. If stop time is not given (i.e. time
148	string is 'x.y,') then analysis goes to the end of the file.
149	Multiple ranges can be separated by spaces, which requires the argument
150	to be quoted e.g. --time "1234.567,1234.789 1235,"
151	Time string is'a1.b1,c1.d1:a2.b2,c2.d2'. Use ':' to separate timestamps
152	for different perf.data files.
153
154	For example, we get the timestamp information from 'perf script'.
155
156	  perf script -i perf.data.old
157	    mgen 13940 [000]  3946.361400: ...
158
159	  perf script -i perf.data
160	    mgen 13940 [000]  3971.150589 ...
161
162	  perf diff --time 3946.361400,:3971.150589,
163
164	It analyzes the perf.data.old from the timestamp 3946.361400 to
165	the end of perf.data.old and analyzes the perf.data from the
166	timestamp 3971.150589 to the end of perf.data.
167
168--cpu:: Only diff samples for the list of CPUs provided. Multiple CPUs can
169	be provided as a comma-separated list with no space: 0,1. Ranges of
170	CPUs are specified with -: 0-2. Default is to report samples on all
171	CPUs.
172
173--pid=::
174	Only diff samples for given process ID (comma separated list).
175
176--tid=::
177	Only diff samples for given thread ID (comma separated list).
178
179COMPARISON
180----------
181The comparison is governed by the baseline file. The baseline perf.data
182file is iterated for samples. All other perf.data files specified on
183the command line are searched for the baseline sample pair. If the pair
184is found, specified computation is made and result is displayed.
185
186All samples from non-baseline perf.data files, that do not match any
187baseline entry, are displayed with empty space within baseline column
188and possible computation results (delta) in their related column.
189
190Example files samples:
191- file A with samples f1, f2, f3, f4,    f6
192- file B with samples     f2,     f4, f5
193- file C with samples f1, f2,         f5
194
195Example output:
196  x - computation takes place for pair
197  b - baseline sample percentage
198
199- perf diff A B C
200
201  baseline/A compute/B compute/C  samples
202  ---------------------------------------
203  b                    x          f1
204  b          x         x          f2
205  b                               f3
206  b          x                    f4
207  b                               f6
208             x         x          f5
209
210- perf diff B A C
211
212  baseline/B compute/A compute/C  samples
213  ---------------------------------------
214  b          x         x          f2
215  b          x                    f4
216  b                    x          f5
217             x         x          f1
218             x                    f3
219             x                    f6
220
221- perf diff C B A
222
223  baseline/C compute/B compute/A  samples
224  ---------------------------------------
225  b                    x          f1
226  b          x         x          f2
227  b          x                    f5
228                       x          f3
229             x         x          f4
230                       x          f6
231
232COMPARISON METHODS
233------------------
234delta
235~~~~~
236If specified the 'Delta' column is displayed with value 'd' computed as:
237
238  d = A->period_percent - B->period_percent
239
240with:
241  - A/B being matching hist entry from data/baseline file specified
242    (or perf.data/perf.data.old) respectively.
243
244  - period_percent being the % of the hist entry period value within
245    single data file
246
247  - with filtering by -C, -d and/or -S, period_percent might be changed
248    relative to how entries are filtered.  Use --percentage=absolute to
249    prevent such fluctuation.
250
251delta-abs
252~~~~~~~~~
253Same as 'delta` method, but sort the result with the absolute values.
254
255ratio
256~~~~~
257If specified the 'Ratio' column is displayed with value 'r' computed as:
258
259  r = A->period / B->period
260
261with:
262  - A/B being matching hist entry from data/baseline file specified
263    (or perf.data/perf.data.old) respectively.
264
265  - period being the hist entry period value
266
267wdiff:WEIGHT-B,WEIGHT-A
268~~~~~~~~~~~~~~~~~~~~~~~
269If specified the 'Weighted diff' column is displayed with value 'd' computed as:
270
271   d = B->period * WEIGHT-A - A->period * WEIGHT-B
272
273  - A/B being matching hist entry from data/baseline file specified
274    (or perf.data/perf.data.old) respectively.
275
276  - period being the hist entry period value
277
278  - WEIGHT-A/WEIGHT-B being user supplied weights in the the '-c' option
279    behind ':' separator like '-c wdiff:1,2'.
280    - WEIGHT-A being the weight of the data file
281    - WEIGHT-B being the weight of the baseline data file
282
283SEE ALSO
284--------
285linkperf:perf-record[1], linkperf:perf-report[1]
286