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, cycles,
94        delta-abs (default is delta-abs).  Default can be changed using
95        diff.compute config option.  See COMPARISON METHODS section for
96        more info.
97
98--cycles-hist::
99	Report a histogram and the standard deviation for cycles data.
100	It can help us to judge if the reported cycles data is noisy or
101	not. This option should be used with '-c cycles'.
102
103-p::
104--period::
105        Show period values for both compared hist entries.
106
107-F::
108--formula::
109        Show formula for given computation.
110
111-o::
112--order::
113       Specify compute sorting column number.  0 means sorting by baseline
114       overhead and 1 (default) means sorting by computed value of column 1
115       (data from the first file other base baseline).  Values more than 1
116       can be used only if enough data files are provided.
117       The default value can be set using the diff.order config option.
118
119--percentage::
120	Determine how to display the overhead percentage of filtered entries.
121	Filters can be applied by --comms, --dsos and/or --symbols options.
122
123	"relative" means it's relative to filtered entries only so that the
124	sum of shown entries will be always 100%.  "absolute" means it retains
125	the original value before and after the filter is applied.
126
127--time::
128	Analyze samples within given time window. It supports time
129	percent with multiple time ranges. Time string is 'a%/n,b%/m,...'
130	or 'a%-b%,c%-%d,...'.
131
132	For example:
133
134	Select the second 10% time slice to diff:
135
136	  perf diff --time 10%/2
137
138	Select from 0% to 10% time slice to diff:
139
140	  perf diff --time 0%-10%
141
142	Select the first and the second 10% time slices to diff:
143
144	  perf diff --time 10%/1,10%/2
145
146	Select from 0% to 10% and 30% to 40% slices to diff:
147
148	  perf diff --time 0%-10%,30%-40%
149
150	It also supports analyzing samples within a given time window
151	<start>,<stop>. Times have the format seconds.nanoseconds. If 'start'
152	is not given (i.e. time string is ',x.y') then analysis starts at
153	the beginning of the file. If stop time is not given (i.e. time
154	string is 'x.y,') then analysis goes to the end of the file.
155	Multiple ranges can be separated by spaces, which requires the argument
156	to be quoted e.g. --time "1234.567,1234.789 1235,"
157	Time string is'a1.b1,c1.d1:a2.b2,c2.d2'. Use ':' to separate timestamps
158	for different perf.data files.
159
160	For example, we get the timestamp information from 'perf script'.
161
162	  perf script -i perf.data.old
163	    mgen 13940 [000]  3946.361400: ...
164
165	  perf script -i perf.data
166	    mgen 13940 [000]  3971.150589 ...
167
168	  perf diff --time 3946.361400,:3971.150589,
169
170	It analyzes the perf.data.old from the timestamp 3946.361400 to
171	the end of perf.data.old and analyzes the perf.data from the
172	timestamp 3971.150589 to the end of perf.data.
173
174--cpu:: Only diff samples for the list of CPUs provided. Multiple CPUs can
175	be provided as a comma-separated list with no space: 0,1. Ranges of
176	CPUs are specified with -: 0-2. Default is to report samples on all
177	CPUs.
178
179--pid=::
180	Only diff samples for given process ID (comma separated list).
181
182--tid=::
183	Only diff samples for given thread ID (comma separated list).
184
185--stream::
186	Enable hot streams comparison. Stream can be a callchain which is
187	aggregated by the branch records from samples.
188
189COMPARISON
190----------
191The comparison is governed by the baseline file. The baseline perf.data
192file is iterated for samples. All other perf.data files specified on
193the command line are searched for the baseline sample pair. If the pair
194is found, specified computation is made and result is displayed.
195
196All samples from non-baseline perf.data files, that do not match any
197baseline entry, are displayed with empty space within baseline column
198and possible computation results (delta) in their related column.
199
200Example files samples:
201- file A with samples f1, f2, f3, f4,    f6
202- file B with samples     f2,     f4, f5
203- file C with samples f1, f2,         f5
204
205Example output:
206  x - computation takes place for pair
207  b - baseline sample percentage
208
209- perf diff A B C
210
211  baseline/A compute/B compute/C  samples
212  ---------------------------------------
213  b                    x          f1
214  b          x         x          f2
215  b                               f3
216  b          x                    f4
217  b                               f6
218             x         x          f5
219
220- perf diff B A C
221
222  baseline/B compute/A compute/C  samples
223  ---------------------------------------
224  b          x         x          f2
225  b          x                    f4
226  b                    x          f5
227             x         x          f1
228             x                    f3
229             x                    f6
230
231- perf diff C B A
232
233  baseline/C compute/B compute/A  samples
234  ---------------------------------------
235  b                    x          f1
236  b          x         x          f2
237  b          x                    f5
238                       x          f3
239             x         x          f4
240                       x          f6
241
242COMPARISON METHODS
243------------------
244delta
245~~~~~
246If specified the 'Delta' column is displayed with value 'd' computed as:
247
248  d = A->period_percent - B->period_percent
249
250with:
251  - A/B being matching hist entry from data/baseline file specified
252    (or perf.data/perf.data.old) respectively.
253
254  - period_percent being the % of the hist entry period value within
255    single data file
256
257  - with filtering by -C, -d and/or -S, period_percent might be changed
258    relative to how entries are filtered.  Use --percentage=absolute to
259    prevent such fluctuation.
260
261delta-abs
262~~~~~~~~~
263Same as 'delta` method, but sort the result with the absolute values.
264
265ratio
266~~~~~
267If specified the 'Ratio' column is displayed with value 'r' computed as:
268
269  r = A->period / B->period
270
271with:
272  - A/B being matching hist entry from data/baseline file specified
273    (or perf.data/perf.data.old) respectively.
274
275  - period being the hist entry period value
276
277wdiff:WEIGHT-B,WEIGHT-A
278~~~~~~~~~~~~~~~~~~~~~~~
279If specified the 'Weighted diff' column is displayed with value 'd' computed as:
280
281   d = B->period * WEIGHT-A - A->period * WEIGHT-B
282
283  - A/B being matching hist entry from data/baseline file specified
284    (or perf.data/perf.data.old) respectively.
285
286  - period being the hist entry period value
287
288  - WEIGHT-A/WEIGHT-B being user supplied weights in the the '-c' option
289    behind ':' separator like '-c wdiff:1,2'.
290    - WEIGHT-A being the weight of the data file
291    - WEIGHT-B being the weight of the baseline data file
292
293cycles
294~~~~~~
295If specified the '[Program Block Range] Cycles Diff' column is displayed.
296It displays the cycles difference of same program basic block amongst
297two perf.data. The program basic block is the code between two branches.
298
299'[Program Block Range]' indicates the range of a program basic block.
300Source line is reported if it can be found otherwise uses symbol+offset
301instead.
302
303SEE ALSO
304--------
305linkperf:perf-record[1], linkperf:perf-report[1]
306