1 llvm-profdata - Profile data tool
2 =================================
4 .. program:: llvm-profdata
9 :program:`llvm-profdata` *command* [*args...*]
14 The :program:`llvm-profdata` tool is a small utility for working with profile
20 * :ref:`merge <profdata-merge>`
21 * :ref:`show <profdata-show>`
22 * :ref:`overlap <profdata-overlap>`
24 .. program:: llvm-profdata merge
34 :program:`llvm-profdata merge` [*options*] [*filename...*]
39 :program:`llvm-profdata merge` takes several profile data files
40 generated by PGO instrumentation and merges them together into a single
41 indexed profile data file.
43 By default profile data is merged without modification. This means that the
44 relative importance of each input file is proportional to the number of samples
45 or counts it contains. In general, the input from a longer training run will be
46 interpreted as relatively more important than a shorter run. Depending on the
47 nature of the training runs it may be useful to adjust the weight given to each
48 input file by using the ``-weighted-input`` option.
50 Profiles passed in via ``-weighted-input``, ``-input-files``, or via positional
51 arguments are processed once for each time they are seen.
59 Print a summary of command line options.
61 .. option:: -output=output, -o=output
63 Specify the output file name. *Output* cannot be ``-`` as the resulting
64 indexed profile data can't be written to standard output.
66 .. option:: -weighted-input=weight,filename
68 Specify an input file name along with a weight. The profile counts of the
69 supplied ``filename`` will be scaled (multiplied) by the supplied
70 ``weight``, where where ``weight`` is a decimal integer >= 1.
71 Input files specified without using this option are assigned a default
72 weight of 1. Examples are shown below.
74 .. option:: -input-files=path, -f=path
76 Specify a file which contains a list of files to merge. The entries in this
77 file are newline-separated. Lines starting with '#' are skipped. Entries may
78 be of the form <filename> or <weight>,<filename>.
80 .. option:: -remapping-file=path, -r=path
82 Specify a file which contains a remapping from symbol names in the input
83 profile to the symbol names that should be used in the output profile. The
84 file should consist of lines of the form ``<input-symbol> <output-symbol>``.
85 Blank lines and lines starting with ``#`` are skipped.
87 The :doc:`llvm-cxxmap <llvm-cxxmap>` tool can be used to generate the symbol
90 .. option:: -instr (default)
92 Specify that the input profile is an instrumentation-based profile.
96 Specify that the input profile is a sample-based profile.
98 The format of the generated file can be generated in one of three ways:
100 .. option:: -binary (default)
102 Emit the profile using a binary encoding. For instrumentation-based profile
103 the output format is the indexed binary format.
107 Emit the profile in text mode. This option can also be used with both
108 sample-based and instrumentation-based profile. When this option is used
109 the profile will be dumped in the text format that is parsable by the profile
114 Emit the profile using GCC's gcov format (Not yet supported).
116 .. option:: -sparse[=true|false]
118 Do not emit function records with 0 execution count. Can only be used in
119 conjunction with -instr. Defaults to false, since it can inhibit compiler
120 optimization during PGO.
122 .. option:: -num-threads=N, -j=N
124 Use N threads to perform profile merging. When N=0, llvm-profdata auto-detects
125 an appropriate number of threads to use. This is the default.
131 Merge three profiles:
135 llvm-profdata merge foo.profdata bar.profdata baz.profdata -output merged.profdata
139 The input file `foo.profdata` is especially important, multiply its counts by 10:
143 llvm-profdata merge -weighted-input=10,foo.profdata bar.profdata baz.profdata -output merged.profdata
145 Exactly equivalent to the previous invocation (explicit form; useful for programmatic invocation):
149 llvm-profdata merge -weighted-input=10,foo.profdata -weighted-input=1,bar.profdata -weighted-input=1,baz.profdata -output merged.profdata
151 .. program:: llvm-profdata show
161 :program:`llvm-profdata show` [*options*] [*filename*]
166 :program:`llvm-profdata show` takes a profile data file and displays the
167 information about the profile counters for this file and
168 for any of the specified function(s).
170 If *filename* is omitted or is ``-``, then **llvm-profdata show** reads its
171 input from standard input.
176 .. option:: -all-functions
178 Print details for every function.
182 Print the counter values for the displayed functions.
184 .. option:: -function=string
186 Print details for a function if the function's name contains the given string.
190 Print a summary of command line options.
192 .. option:: -output=output, -o=output
194 Specify the output file name. If *output* is ``-`` or it isn't specified,
195 then the output is sent to standard output.
197 .. option:: -instr (default)
199 Specify that the input profile is an instrumentation-based profile.
203 Instruct the profile dumper to show profile counts in the text format of the
204 instrumentation-based profile data representation. By default, the profile
205 information is dumped in a more human readable form (also in text) with
210 Instruct the profile dumper to show the top ``n`` functions with the
211 hottest basic blocks in the summary section. By default, the topn functions
216 Specify that the input profile is a sample-based profile.
218 .. option:: -memop-sizes
220 Show the profiled sizes of the memory intrinsic calls for shown functions.
222 .. option:: -value-cutoff=n
224 Show only those functions whose max count values are greater or equal to ``n``.
225 By default, the value-cutoff is set to 0.
227 .. option:: -list-below-cutoff
229 Only output names of functions whose max count value are below the cutoff
234 Only show context sensitive profile counts. The default is to filter all
235 context sensitive profile counts.
237 .. program:: llvm-profdata overlap
239 .. _profdata-overlap:
247 :program:`llvm-profdata overlap` [*options*] [*base profile file*] [*test profile file*]
252 :program:`llvm-profdata overlap` takes two profile data files and displays the
253 *overlap* of counter distribution between the whole files and between any of the
256 In this command, *overlap* is defined as follows:
257 Suppose *base profile file* has the following counts:
258 {c1_1, c1_2, ..., c1_n, c1_u_1, c2_u_2, ..., c2_u_s},
259 and *test profile file* has
260 {c2_1, c2_2, ..., c2_n, c2_v_1, c2_v_2, ..., c2_v_t}.
261 Here c{1|2}_i (i = 1 .. n) are matched counters and c1_u_i (i = 1 .. s) and
262 c2_v_i (i = 1 .. v) are unmatched counters (or counters only existing in)
263 *base profile file* and *test profile file*, respectively.
264 Let sum_1 = c1_1 + c1_2 + ... + c1_n + c1_u_1 + c2_u_2 + ... + c2_u_s, and
265 sum_2 = c2_1 + c2_2 + ... + c2_n + c2_v_1 + c2_v_2 + ... + c2_v_t.
266 *overlap* = min(c1_1/sum_1, c2_1/sum_2) + min(c1_2/sum_1, c2_2/sum_2) + ...
267 + min(c1_n/sum_1, c2_n/sum_2).
269 The result overlap distribution is a percentage number, ranging from 0.0% to
270 100.0%, where 0.0% means there is no overlap and 100.0% means a perfect
273 Here is an example, if *base profile file* has counts of {400, 600}, and
274 *test profile file* has matched counts of {60000, 40000}. The *overlap* is 80%.
279 .. option:: -function=string
281 Print details for a function if the function's name contains the given string.
285 Print a summary of command line options.
287 .. option:: -o=output or -o output
289 Specify the output file name. If *output* is ``-`` or it isn't specified,
290 then the output is sent to standard output.
292 .. option:: -value-cutoff=n
294 Show only those functions whose max count values are greater or equal to ``n``.
295 By default, the value-cutoff is set to max of unsigned long long.
299 Only show overlap for the context sensitive profile counts. The default is to show
300 non-context sensitive profile counts.
305 :program:`llvm-profdata` returns 1 if the command is omitted or is invalid,
306 if it cannot read input files, or if there is a mismatch between their data.