LLVM-PROFDATA(1)

📦
version 14.0.6 · 8 versions
, in ubuntu questing

🦋 bestie, meet llvm-profdata 🦋

LLVM-PROFDATA(1) from the LLVM manual. Okay so this is your profile-data bestie, and she is literally here to make profile files make sense. 💖

🌸 NAME 🌸

llvm-profdata — the profile data tool 💅

✨ SYNOPSIS ✨

llvm-profdata command [args…]

💖 DESCRIPTION 💖

The llvm-profdata tool is a small utility for working with profile data files. tiny but iconic — it handles profile data files, periodt. 💅

👑 COMMANDS 👑

  • merge
  • show
  • overlap

That’s the whole command menu, bestie. ✨

💅 MERGE 💅

🌈 SYNOPSIS 🌈

llvm-profdata merge [options] [filename…]

💫 DESCRIPTION 💫

llvm-profdata merge takes several profile data files generated by PGO instrumentation and merges them together into a single indexed profile data file. slay.

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. iykyk

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. classic help behavior, no cap.
  • --output=<output>, -o — Specify the output file name. Output cannot be - as the resulting indexed profile data can’t be written to standard output. not me trying to pipe it, bestie — it literally won’t work.
  • --weighted-input=<weight,filename> — Specify an input file name along with a weight. The profile counts of the supplied filename will be scaled (multiplied) by the supplied weight, where weight is a decimal integer >= 1. Input files specified without using this option are assigned a default weight of 1. Examples are shown below. math but make it profile glow-up ✨
  • --input-files=<path>, -f — 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>. list in, merge out.
  • --remapping-file=<path>, -r — 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. iconic little helper.

  • --instr (default) — Specify that the input profile is an instrumentation-based profile.
  • --sample — Specify that the input profile is a sample-based profile.

The format of the generated 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.
  • --gcc — Emit the profile using GCC’s gcov format (Not yet supported). not yet supported, bestie — don’t come for me.
  • --sparse[=true|false] — Do not emit function records with 0 execution count. Can only be used in conjunction with -instr. Defaults to false, since it can inhibit compiler optimization during PGO. vibe check: default false.
  • --num-threads=<N>, -j — Use N threads to perform profile merging. When N=0, llvm-profdata auto-detects an appropriate number of threads to use. This is the default. parallel slay.
  • --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’. choose your drama level.
  • --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 extbinary 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 extbinary format. compress, bestie.
  • --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 extbinary 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 extbinary format.
  • --supplement-instr-with-sample=<file> — Supplement an instrumentation profile with sample profile. The sample profile is the input of the flag. Output will be in instrumentation format (only works with -instr). instr + sample = main character crossover event.
  • --zero-counter-threshold=<float> — For the function which is cold in instr profile but hot in sample profile, if the ratio of the number of zero counters divided by the the total number of counters is above the threshold, the profile of the function will be regarded as being harmful for performance and will be dropped. if it’s harmful, it’s dropped. periodt.
  • --instr-prof-cold-threshold=<int> — User specified cold threshold for instr profile which will override the cold threshold got from profile summary.
  • --suppl-min-size-threshold=<int> — If the size of a function is smaller than the threshold, assume it can be inlined by PGO early inliner and it will not be adjusted based on sample profile.
  • --debug-info=<path> — Specify the executable or .dSYM that contains debug info for the raw profile. When -debug-info-correlate was used for instrumentation, use this option to correlate the raw profile.

📝 EXAMPLES 📝

✨ Basic Usage ✨

Merge three profiles:

llvm-profdata merge foo.profdata bar.profdata baz.profdata -output merged.profdata

⚖️ Weighted Input ⚖️

The input file foo.profdata is especially important, multiply its counts by 10:

llvm-profdata merge --weighted-input=10,foo.profdata bar.profdata baz.profdata --output merged.profdata

Exactly equivalent to the previous invocation (explicit form; useful for programmatic invocation):

llvm-profdata merge --weighted-input=10,foo.profdata --weighted-input=1,bar.profdata --weighted-input=1,baz.profdata --output merged.profdata

🫶 SHOW 🫶

🌈 SYNOPSIS 🌈

llvm-profdata show [options] [filename]

💫 DESCRIPTION 💫

llvm-profdata show takes a profile data file and displays the information about the profile counters for this file and for any of the specified function(s). show me the receipts 👀

If filename is omitted or is -, then llvm-profdata show reads its input from standard input.

⚙️ OPTIONS ⚙️

  • --all-functions — Print details for every function. every single one.
  • --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. substring matching, bestie.
  • --help — Print a summary of command line options.
  • --output=<output>, -o — Specify the output file name. If output is - or it isn’t specified, then the output is sent to standard output. stdout is always an option here.
  • --instr (default) — 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. text mode vs. annotated text mode, iykyk.
  • --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. hot girl hot blocks 💅
  • --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. cutoff says 0, so everyone starts invited.
  • --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. context sensitive counts only, no cap.
  • --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 extbinary 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 extbinary format.

✨ OVERLAP ✨

🌸 SYNOPSIS 🌸

llvm-profdata overlap [options] [base profile file] [test profile file]

💖 DESCRIPTION 💖

llvm-profdata overlap takes two profile data files and shows you the overlap of counter distribution — both between the whole files and between any of the specific functions you point it at. It's giving vibe check, but for profiles. 💅

okay so bestie, here's how overlap is actually defined, because this part is the whole plot:

Say the base profile file has these counts:

{c1_1, c1_2, …, c1_n, c1_u_1, c2_u_2, …, c2_u_s}

and the test profile file has:

{c2_1, c2_2, …, c2_n, c2_v_1, c2_v_2, …, c2_v_t}

Here c{1|2}_i (for i = 1 .. n) are the matched counters, while c1_u_i (for i = 1 .. s) and c2_v_i (for i = 1 .. v) are the unmatched counters — meaning the counters that only exist in the base profile file and only exist in the test profile file, respectively. 🙈

Now, the math, and I need you to stay with me because this is lowkey the main character moment:

sum_1 = c1_1 + c1_2 +  … + c1_n +  c1_u_1 + c2_u_2 + … + c2_u_s
sum_2 = c2_1 + c2_2 + … + c2_n + c2_v_1 + c2_v_2 + … + c2_v_t
overlap = min(c1_1/sum_1, c2_1/sum_2) + min(c1_2/sum_1, c2_2/sum_2) + … + min(c1_n/sum_1, c2_n/sum_2)

The result — the overlap distribution — is a percentage number, ranging from 0.0% to 100.0%. Where 0.0% means there is no overlap at all (💀) and 100.0% means a perfect overlap (👑 periodt).

Here's an example, so it's not just vibes: if the base profile file has counts of {400, 600}, and the test profile file has matched counts of {60000, 40000}, the overlap is 80%. Slay. ✨

🎀 OPTIONS 🎀

  • --function=<string> — prints the details for a function if that function's name contains the given string. Basically a little name-based filter, iykyk. 🔍
  • --help — prints a summary of the command line options. The classic "help me help you" button. 🫶
  • --output=<output>, -o — picks the output file name for you. If output is -, or if it isn't specified at all, the output gets sent to standard output instead. 🌈
  • --value-cutoff=<n> — shows only those functions whose max count values are greater or equal to n. By default, the value-cutoff is set to the max of unsigned long long, i.e. everybody's invited. 🎉
  • --cs — only shows overlap for the context sensitive profile counts. The default is to show non-context sensitive profile counts. Pick your lane. 💜

💀 EXIT STATUS 💀

llvm-profdata returns 1 if:

  • the command is omitted or is invalid,
  • it cannot read the input files, or
  • there is a mismatch between their data. 😭

👑 AUTHOR 👑

Maintained by the LLVM Team (https://llvm.org/).

2003-2025, LLVM Project

and that's the whole page, bestie — this was LLVM-PROFDATA(1), version 14, dated 2025-05-22. go forth and overlap responsibly 💖✨🎀

📦 where this page comes from