9.2. c29profdata - Profile Data Tool¶

The c29profdata tool can be used to work with profile data files.

9.2.1. Usage¶

c29profdata command [options] <filenames>

c29profdata - Command to invoke the profile data tool.
command - One of the available c29profdata modes of operation: merge or show
options - one or more options arguments appropriate for the specified command mode
<filenames> - one or more input profile data files

9.2.2. Commands¶

9.2.2.1. merge¶

The c29profdata merge command takes several profile data files generated by c29clang instrumentation options and merges them together into a single indexed profile data file.

By default profile data is merged without modification. This means that the relative importance of each input file is proportional to the number of samples or counts it contains. In general, the input from a longer training run will be interpreted as relatively more important than a shorter run. Depending on the nature of the training runs it may be useful to adjust the weight given to each input file by using the -weighted-input option.

Profiles passed in via -weighted-input, -input-files, or via positional arguments are processed once for each time they are seen.

Options

-help¶: Print a summary of command line options.

-output=<filename>, -o=<filename>¶: Specify the output <filename>.

-weighted-input=<weight>,<filename>¶: Specify an input <filename> along with a <weight>. The profile counts of the supplied <filename> will be scaled (multiplied) by the supplied <weight>, where <weight> is an integer >= 1. Input files specified with using this option are assigned a default <weight> of 1.

-input-files=<path>, -f=<path>¶: Specify a file which contains a list of files to merge. The entries in this file are newline-separated. Lines starting with ‘#’ are skipped. Entries may be of the form <filename> or <weight>,<filename>.

-remapping-file=<path>, -r=<path>¶

Specify a file which contains a remapping from symbol names in the input profile to the symbol names that should be used in the output profile. The file should consist of lines of the form <input-symbol> <output-symbol>. Blank lines and lines starting with ‘#’ are skipped.

The llvm-cxxmap tool can be used to generate the symbol remapping file.

-instr¶: Specify that the input profile is an instrumentation-based profile (default).

-sample¶

Specify that the input profile is a sample-based profile.

The format of the output file can be generated in one of three ways:

-binary (default)¶: Emit the profile using a binary encoding. For instrumentation-based profile the output format is the indexed binary format.

-extbinary¶: Emit the profile using an extensible binary encoding. This option can only be used with sample-based profile. The extensible binary encoding can be more compact with compression enabled and can be loaded faster than the default binary encoding.

-text¶: Emit the profile in text mode. This option can also be used with both sample-based and instrumentation-based profile. When this option is used the profile will be dumped in the text format that is parsable by the profile reader.

-sparse=[true|false]¶: Do not emit function records with 0 execution count. This can only be used in conjunction with the -instr option. Defaults to false, since it can inhibit compiler optimization during profile guided optimization.

-num-threads=<N>, -j=<N>¶: Use <N> threads to perform profile merging. When <N>=0, c29profdata auto-detects an appropriate number of threads to use. This is the default.

-failure-mode=[any|all]¶

Set the failure mode. There are two options:

any causes the merge command to fail if any profiles are invalid, and
all causes the merge command to fail only if all profiles are invalid.

If all is set, information from any invalid profiles is excluded from the final merged product. The default failure mode is any.

-prof-sym-list=<path>¶: Specify a file which contains a list of symbols to generate profile symbol list in the profile. This option can only be used with sample-based profile in extensible binary format. The entries in this file are newline-separated.

-compress-all-sections=[true|false]¶: Compress all sections when writing the profile. This option can only be used with sample-based profile in extensible binary format.

-use-md5=[true|false]¶: Use MD5 to represent string in name table when writing the profile. This option can only be used with sample-based profile in extensible binary format.

-gen-partial-profile=[true|false]¶: Mark the profile to be a partial profile which only provides partial profile coverage for the optimized target. This option can only be used with sample-based profile in extensible binary format.

-supplement-instr-with-sample=<path to sample profile>¶: Supplement an instrumentation profile with sample profile. The sample profile is the input of the flag. Output will be in instrumentation format (this only works in combination with the -instr option).

9.2.2.2. show¶

The c29profdata show command takes a profile data file and displays the information about the profile counters for the specified input file and for any of the specified functions.

If the input file is omitted or is ‘-’, then c29profdata show reads its input from standard input.

Options

-all-functions¶: Print details for every function.

-counts¶: Print the counter values for the displayed functions.

-function=<string>¶: Print details for a function if the function’s name contains the given <string>.

-help¶: Print a summary of command line options.

-output=<filename>, -o=<filename>¶: Specify the output <filename>. If <filename> is ‘-’ or it is not specified, then the output is sent to standard output.

-instr¶: Specify that the input profile is an instrumentation-based profile.

-text¶: Instruct the profile dumper to show profile counts in the text format of the instrumentation-based profile data representation. By default, the profile information is dumped in a more human readable form (also in text) with annotations.

-topn=<n>¶: Instruct the profile dumper to show the top <n> functions with the hottest basic blocks in the summary section. By default, the topn functions are not dumped.

-sample¶: Specify that the input profile is a sample-based profile.

-memop-sizes¶: Show the profiled sizes of the memory intrinsic calls for shown functions.

-value-cutoff=<n>¶: Show only those functions whose max count values are greater or equal to <n>. By default, the value-cutoff is set to 0.

-list-below-cutoff¶: Only output names of functions whose max count value are below the cutoff value.

-showcs¶: Only show context sensitive profile counts. The default is to filter all context sensitive profile counts.

-show-prof-sym-list=[true|false]¶: Show profile symbol list if it exists in the profile. This option is only meaningful for sample-based profile in extensible binary format.

-show-sec-info-only=[true|false]¶: Show basic information about each section in the profile. This option is only meaningful for sample-based profile in extensible binary format.

9.2.3. Exit Status¶

c29profdata returns 1 if the command is omitted or is invalid, if it cannot read input files, or if there is a mismatch between their data.