LLVM-DWARFDUMP(1)
π LLVM-DWARFDUMP(1) β bestie's guide to the LLVM manual β¨
okay so this is llvm-dwarfdump, a page from the LLVM manual, and it's literally the tool you call when you want to see what's actually inside your DWARF debug info. let's get into it π«Ά
πΈ NAME πΈ
llvm-dwarfdump β dump and verify DWARF debug information. that's the whole vibe. π
β¨ SYNOPSIS β¨
llvm-dwarfdump [options] [filename ...]
π¦ DESCRIPTION π¦
okay so llvm-dwarfdump parses the DWARF sections inside object files, archives, and .dSYM bundles, and then prints the contents in human-readable form. it's giving translator, honestly π
here's the important part though: only the .debug_info section gets printed unless you specifically ask for one of the section-specific options, or you hit it with --all. so don't come for it when you only see .debug_info β that's the default, bestie.
and if you don't give it any input file at all? it just uses a.out instead. iconic default behavior β¨ and if you pass - as the input file, llvm-dwarfdump reads the input from its standard input stream. very cat-core of it.
π OPTIONS π
-a,--allβ dump all supported DWARF sections. all of them. every single one. periodt.--arch=<arch>β dump DWARF debug information for the specified CPU architecture. you can name the architecture by name or by number, and you can pass this option multiple times β once for each architecture you want. by default, all CPU architectures will be printed.-c,--show-childrenβ show a debug info entry's children when you're selectively printing with the=<offset>argument of--debug-info, or with options like--findor--name. the kids are coming too π--colorβ use colors in output. slay π--error-display=<value>β sets how much detail and summary you get when verifying. implies--verify. the supported values are:quietβ only display whether errors occurred.summaryβ display only a summary of the errors found.detailsβ display each error in detail but no summary.fullβ display each error as well as a summary. [default]
-f <name>,--find=<name>β search for the exact text<name>in the accelerator tables and print the matching debug information entries. when there are no accelerator tables, or the name of the DIE you're looking for isn't in the accelerator tables, try the slower but more complete--nameoption. she's thorough but she's not fast, iykyk π-F,--show-formβ show DWARF form types after the DWARF attribute types. little bonus info, we love β¨-h,--helpβ show help and usage for this command.--help-listβ show help and usage for this command without grouping the options into categories. ungrouped. unbothered. π-i,--ignore-caseβ ignore case distinctions when using--name.-n <name>,--name=<name>β find and print all debug info entries whose name (theDW_AT_nameattribute) is<name>. yes, all of them.--lookup=<address>β look up<address>in the debug information and print out the file, function, block, and line table details. the full backstory fr fr-o <path>β redirect output to a file specified by<path>, where-is the standard output stream.-p,--show-parentsβ show a debug info entry's parents when selectively printing with the=<offset>argument of--debug-info, or options like--findor--name. family tree reveal π§--parent-recurse-depth=<N>β when displaying debug info entry parents, only show them to a maximum depth of<N>. we're not going all the way back to the ancestors.--quietβ use with--verifyto not emit to STDOUT. shhh π€«-r <N>,--recurse-depth=<N>β when displaying debug info entries, only show children to a maximum depth of<N>.--show-section-sizesβ show the sizes of all debug sections, expressed in bytes.--show-sourcesβ print all source files mentioned in the debug information. absolute paths are given whenever possible.--statisticsβ collect debug info quality metrics and print the results as machine-readable single-line JSON output. the output format is described below in the section FORMAT OF STATISTICS OUTPUT. π--summarize-typesβ abbreviate the description of type unit entries. less reading, same energy.-x,--regexβ treat any<name>strings as regular expressions when searching with--name. and if--ignore-caseis also specified, the regular expression becomes case-insensitive. regex girlies rise π-u,--uuidβ show the UUID for each architecture.--diffβ dump the output in a format that is more friendly for comparing DWARF output from two different files.-v,--verboseβ display verbose information when dumping. this can help to debug DWARF issues. when things are weird, go verbose, ngl.--verifyβ verify the structure of the DWARF information by verifying the compile unit chains, DIE relationships graph, address ranges, and more. the vibe check of debug info π--verify-json=<path>β output JSON-formatted error summary to the file specified by<path>. implies--verify. the output format is described below in the section FORMAT OF VERIFY JSON OUTPUT.--versionβ display the version of the tool.--debug-abbrev,--debug-addr,--debug-aranges,--debug-cu-index,--debug-frame [=<offset>],--debug-gnu-pubnames,--debug-gnu-pubtypes,--debug-info [=<offset>],--debug-line [=<offset>],--debug-line-str,--debug-loc [=<offset>],--debug-loclists [=<offset>],--debug-macro,--debug-names,--debug-pubnames,--debug-pubtypes,--debug-ranges,--debug-rnglists,--debug-str,--debug-str-offsets,--debug-tu-index,--debug-types [=<offset>],--eh-frame [=<offset>],--gdb-index,--apple-names,--apple-types,--apple-namespaces,--apple-objcβ dump the specified DWARF section by name. only the.debug_infosection is shown by default. some entries support adding an=<offset>as a way to provide an optional offset of the exact entry to dump within the respective section. when an offset is provided, only the entry at that offset will be dumped; otherwise the entire section will be dumped.- heads up: the
--debug-macrooption prints both the.debug_macroand the.debug_macinfosections. two for one π - also,
--debug-frameand--eh-frameare aliases β in cases where both sections are present, one command outputs both.
- heads up: the
@<FILE>β read command-line options from<FILE>. cute lil options file π
π§ FORMAT OF STATISTICS OUTPUT π§
the --statistics option generates single-line JSON output representing quality metrics of the processed debug info. these metrics are useful to compare changes between two compilers β particularly for judging the effect that a change to the compiler has on the debug info quality. like a report card for your compiler, honestly π
the output is formatted as key-value pairs. the first pair contains a version number. the following naming scheme is used for the keys:
- variables ==> local variables and parameters
- local vars ==> local variables
- params ==> formal parameters
for aggregated values, the following keys are used:
sum_of_all_variables(...)==> the sum applied to all variables#bytes==> the number of bytes#variables - entry values ...==> the number of variables excluding the entry values etc.
π₯ FORMAT OF VERIFY JSON OUTPUT π₯
the format of the JSON output created by the --verify-json is:
{
"error-categories": {
"<first category description>": {"count": 1234},
"<next category description>": {"count": 4321}
},
"error-count": 5555
}
the following is generated if there are no errors reported:
{
"error-categories": {},
"error-count": 0
}
no errors? no categories. zero error-count. we love a clean run β¨
β EXIT STATUS β
llvm-dwarfdump returns 0 if the input files were parsed and dumped successfully. otherwise, it returns 1. simple as that, bestie.
π SEE ALSO π
dsymutil(1) β <dsymutil.html>
π Author π
maintained by the LLVM Team (https://llvm.org/). they're the main characters of this story, fr fr π
π Copyright π
2003-2026, LLVM Project
and that's the whole page, no cap π this was LLVM-DWARFDUMP(1) from the LLVM manual, dated 2026-05-15 β go dump some DWARF, you absolute icon π¦π
π¦ where this page comes from
- fedora 44: llvm21 21.1.8-5.fc44 Β· Apache-2.0 WITH LLVM-exception OR NCSA
- fedora 44: llvm21 21.1.8-5.fc44 Β· Apache-2.0 WITH LLVM-exception OR NCSA
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms