Trace Processor (C++)

Trace Processor is a C++ library (src/trace_processor) that ingests traces in a variety of formats and exposes an SQL interface for querying them through a consistent set of tables. It also computes trace summaries, annotates traces with human-readable descriptions, and derives new events from trace contents.

Most users interact with Trace Processor through the trace_processor shell, a command-line wrapper around the library that opens an interactive PerfettoSQL prompt. To embed Trace Processor in another C++ application, see Embedding the C++ library. Python users should use the Python API instead.

{#shell} The trace_processor shell

The trace_processor shell is a command-line binary that loads a trace and opens an interactive SQL prompt on it.

Downloading the shell

The shell is a thin Python wrapper that you download from the Perfetto website. On first use it fetches and caches the native binary for your platform (including trace_processor_shell.exe on Windows) under ~/.local/share/perfetto/prebuilts.

Running the shell

Once downloaded, run it on a trace file:

This opens an interactive SQL shell where you can query the trace. For how to write queries, see the Getting Started with PerfettoSQL guide.

TIP: the trace file can also be a ZIP or TAR archive containing several traces: they are merged onto a single timeline. See Merging traces from the command line.

For example, to see all the slices in a trace:

> SELECT ts, dur, name FROM slice LIMIT 10;
ts                   dur                  name
-------------------- -------------------- ---------------------------
     261187017446933               358594 eglSwapBuffersWithDamageKHR
     261187017518340                  357 onMessageReceived
     261187020825163                 9948 queueBuffer
     261187021345235                  642 bufferLoad
     261187121345235                  153 query
...

Or, to see the values of all counters:

> SELECT ts, value FROM counter LIMIT 10;
ts                   value
-------------------- --------------------
     261187012149954          1454.000000
     261187012399172          4232.000000
     261187012447402         14304.000000
     261187012535839         15490.000000
     261187012590890         17490.000000
     261187012590890         16590.000000
...

{#sessions} Keeping a trace warm: sessions

Parsing a large trace takes time. If you plan to run several queries against the same trace, load it once into a named background session and point each invocation at that session:

# Load the trace once into a background session.
trace_processor server unix --name mysession --daemonize trace.pftrace

# Run queries against the warm session instead of re-loading the trace.
trace_processor query --remote mysession "SELECT count(*) FROM slice"

The query, interactive, metrics and summarize subcommands all accept --remote, which talks to a session over the same TraceProcessor RPC interface the Perfetto UI uses. See Analyzing traces from the command line for a walkthrough, and the server subcommand below for the mode and flag details.

{#subcommands} Subcommand interface

Besides the interactive REPL, trace_processor takes a subcommand as its first argument for non-interactive workflows:

trace_processor <command> [flags] [positional args]

trace_processor --help prints the top-level summary below. To see the flags of one subcommand, run trace_processor <command> --help (equivalently trace_processor help <command>):

Perfetto Trace Processor.
Usage: trace_processor [command] [flags] [trace_file]

If no command is given, opens an interactive SQL shell on the trace file.

Commands:
  query         Load a trace and run a SQL query.
  interactive   Interactive SQL shell (default if no command is given).
  server        Start an RPC server.
  summarize     Compute a trace summary from specs and/or built-in metrics.
  export        Export trace data (sqlite, arrow_tar, perfetto).
  metrics       Run v1 metrics (deprecated; use 'summarize --metrics-v2').
  convert       Convert trace format.

Common flags (apply to all commands):
  -h, --help                  Show help (per-command if after a command).
  -v, --version               Print version.
      --full-sort             Force full sort ignoring windowing.
      --no-ftrace-raw         Prevent ingestion of typed ftrace into raw table.
      --add-sql-package PATH  Register SQL files from a directory as a package.
  -m, --metatrace FILE        Enable metatracing, write to FILE.

Backwards compatibility. The classic flat-flag interface (-q, -Q, --httpd, --summary, --run-metrics, -e, --stdiod) is still supported through an internal translation layer, so existing scripts keep working unchanged. Run trace_processor --help-classic for the full list of classic flags.

{#subcommand-query} query: run SQL

query loads a trace, runs one or more ;-separated SQL statements, prints the results to stdout, and exits. SQL can be passed as an argument, read from a file, or piped on stdin:

# Pass SQL as an argument.
trace_processor query trace.pftrace "SELECT ts, dur, name FROM slice LIMIT 5"

# Read SQL from a file.
trace_processor query -f queries.sql trace.pftrace

# Pipe SQL on stdin.
cat queries.sql | trace_processor query trace.pftrace

Each statement's result set is printed as CSV, and consecutive result sets are separated by a single blank line. The separator is unambiguous because every string value is quoted.

Flags:

• --remote ADDR: run against a warm session instead of loading a local trace; see sessions. ADDR is a session name, a *.sock or absolute socket path, or host:port. No trace-file argument is passed in this mode.
• -f, --query-file FILE: read SQL from FILE; pass - to read from stdin.
• -i, --interactive: drop into the interactive REPL after the queries finish.
• -W, --wide: use double-width columns when printing results.
• --perf-file FILE: write trace-load and query timings to FILE.
• --structured-query-id ID plus --summary-spec FILE (advanced): run a single structured query by ID from one or more TraceSummarySpec files, instead of the SQL sources above.

{#subcommand-interactive} interactive: REPL

interactive opens the same interactive PerfettoSQL prompt shown in the previous section. It is the default subcommand, so trace_processor trace.pftrace and trace_processor interactive trace.pftrace are equivalent. The only subcommand-specific flag is -W, --wide.

{#subcommand-server} server: HTTP, stdio, or unix RPC

server exposes trace processor over a remote-procedure-call protocol:

# HTTP server, used by ui.perfetto.dev. Listens on port 9001 by default.
trace_processor server http

# Pre-load a trace and serve it over HTTP.
trace_processor server http trace.pftrace

# stdio server: length-prefixed RPC for tooling that embeds
# trace_processor as a subprocess.
trace_processor server stdio

# Named unix-socket session: keeps the trace warm for repeated
# `query --remote <name>` calls (see the sessions section above).
trace_processor server unix --name mysession --daemonize trace.pftrace

# Stop a unix session by name or socket path.
trace_processor server kill mysession

Flags:

• --port PORT: HTTP port (default 9001).
• --ip-address IP: HTTP bind address.
• --additional-cors-origins O1,O2,...: extra CORS-allowed origins on top of the defaults (https://ui.perfetto.dev, http://localhost:10000, http://127.0.0.1:10000).
• --name NAME: session name for unix mode (default: auto-generated).
• --path PATH: explicit socket path for unix mode (mutually exclusive with --name).
• --daemonize: detach into the background (unix mode, POSIX only).
• --idle-timeout auto|DUR: reap the server after this much inactivity (e.g. 30m, 90s); auto means 30 minutes for unix and never for http, 0/never disables.
• --idle-start auto|orphaned|last-query: when the idle clock applies (default auto: owner-aware).

The trace file is optional in http and unix modes; clients can also load traces remotely. The most common client is the Perfetto UI, which auto-detects a local server and offloads trace parsing to it. See Visualising large traces for the end-user flow, or trace_processor.proto for the RPC wire schema.

{#subcommand-summarize} summarize: compute trace summaries

summarize computes a trace summary. Pass the trace file first, then any spec files; select built-in v2 metrics with --metrics-v2:

# Run every available v2 metric.
trace_processor summarize --metrics-v2 all trace.pftrace

# Run two specific metrics defined in spec.textproto.
trace_processor summarize \
  --metrics-v2 startup_metric,memory_metric \
  trace.pftrace spec.textproto

Flags:

• --metrics-v2 IDS: comma-separated metric ids, or the literal all.
• --metadata-query ID: query id used to populate the summary's metadata field.
• --format text|binary: output format for the TraceSummary proto (default text).
• --post-query FILE: run this SQL file after summarization. When set, the summary proto is not printed; the SQL output is printed instead.
• --perf-file FILE: write load/query timings to FILE.
• -i, --interactive: drop into the REPL after summarization finishes.

Spec files are detected as binary or text by extension (.pb for binary, .textproto for text), with content sniffing as a fallback.

{#subcommand-export} export: write trace data to a file

export writes the parsed trace data to a file. The format is the first positional argument, the output path is given with -o:

# Version-coupled archive, loadable by the same version of trace processor.
trace_processor export perfetto -o archive.tar trace.pftrace

# Static tables as standard Arrow files in a tar.
trace_processor export arrow_tar -o tables.tar trace.pftrace

# Static tables and views as a SQLite database.
trace_processor export sqlite -o trace.db trace.pftrace

Formats:

• perfetto: a version-coupled archive of the non-empty static tables. A fresh trace processor instance from the same version can load it back as a trace; a different version may load it, but this is not guaranteed. The only format that can be reloaded.
• arrow_tar: a tar of standard Apache Arrow files, one per statically registered table, including empty tables and implicit ID columns. Stable and forwards-compatible across versions, for external consumers (e.g. pandas, Polars, pyarrow). Cannot be loaded back into trace processor.
• sqlite: the statically registered tables plus the trace's views, as a SQLite database readable by any SQLite tool.

Flags:

• -o, --output FILE: output file path (required).

All three formats export the statically registered tables; only sqlite also includes views. Runtime tables created during the session (e.g. CREATE PERFETTO TABLE) are not exported. Exports stream to disk, so memory use stays bounded for large traces. For task-oriented recipes, see Export trace data.

{#global-flags} Global flags (apply to every subcommand)

These flags are accepted in addition to the subcommand-specific flags above and behave the same across all subcommands:

• Trace ingestion: --full-sort, --no-ftrace-raw, --analyze-trace-proto-content, --crop-track-events.
• PerfettoSQL packages: --add-sql-package PATH[@PKG], --override-sql-package PATH[@PKG], --override-stdlib PATH (requires --dev).
• Metric extensions: --metric-extension DISK_PATH@VIRTUAL_PATH.
• Auxiliary file content: --register-files-dir PATH exposes the contents of files under PATH to importers (e.g. ETM decoders).
• Development: --dev, --dev-flag KEY=VALUE, --extra-checks.
• Metatracing: -m, --metatrace FILE, --metatrace-buffer-capacity N, --metatrace-categories CATEGORIES. This produces a Perfetto trace of trace processor itself, which you can load back into the UI for performance debugging.

{#embedding} Embedding the C++ library

The public API centers on the TraceProcessor class in trace_processor.h. All high-level operations (parsing trace bytes, executing SQL queries, computing summaries) are member functions of this class.

Create an instance with CreateInstance:

#include "perfetto/trace_processor/trace_processor.h"

using namespace perfetto::trace_processor;

Config config;
std::unique_ptr<TraceProcessor> tp = TraceProcessor::CreateInstance(config);

Loading a trace

To ingest a trace, call Parse repeatedly with chunks of trace bytes, then NotifyEndOfFile once the whole trace has been pushed:

while (/* more data available */) {
  TraceBlobView blob = /* ... */;
  base::Status status = tp->Parse(std::move(blob));
  if (!status.ok()) { /* handle error */ }
}
base::Status status = tp->NotifyEndOfFile();

Because reading a trace from the filesystem is a common case, a helper ReadTrace is provided in read_trace.h:

#include "perfetto/trace_processor/read_trace.h"

base::Status status = ReadTrace(tp.get(), "/path/to/trace.pftrace");

ReadTrace reads the file from disk, calls Parse with the contents, and calls NotifyEndOfFile for you.

Executing queries

Run queries with ExecuteQuery, which returns an Iterator that streams rows back to the caller:

auto it = tp->ExecuteQuery("SELECT ts, name FROM slice LIMIT 10");
while (it.Next()) {
  int64_t ts = it.Get(0).AsLong();
  std::string name = it.Get(1).AsString();
  // ...
}
if (!it.Status().ok()) {
  // Query produced an error.
}

Two important rules when using the iterator:

• Always call Next before accessing values. The iterator is positioned before the first row when returned, so Get cannot be called until Next has returned true.
• Always check Status after iteration finishes. A query may fail partway through; Next returning false only means iteration stopped, not that it succeeded. Inspect Status() to distinguish EOF from an error.

See the comments in iterator.h for the full iterator API.

Other functionality

The TraceProcessor class also exposes:

• Trace summarization (Summarize): computes structured summaries of a trace. See Trace Summarization for a user-facing description.
• Custom SQL packages (RegisterSqlPackage): registers PerfettoSQL files under a package name so queries can INCLUDE them.
• Out-of-band file content (RegisterFileContent): passes auxiliary data to importers, e.g. binaries used to decode ETM traces.
• Metatracing (EnableMetatrace / DisableAndReadMetatrace): traces Trace Processor itself for performance debugging.

Refer to the comments in trace_processor.h for the complete API surface.