commit | c56c2429b0f6970c05d630f05acf44b6ada2f346 | [log] [tgz] |
---|---|---|
author | Hector Dearman <hjd@google.com> | Tue Jul 03 09:50:03 2018 +0100 |
committer | Hector Dearman <hjd@google.com> | Tue Jul 03 09:50:03 2018 +0100 |
tree | d65a635b815d7edec8093a280cc560697ced1647 | |
parent | ea4bcdc1e7f40770e068833fceda0ac3d4a97aa9 [diff] |
perfetto-ui: Plumb trace_processor to javascript src/trace_processor/wasm_bridge.cc is our entry point to the WASM code it exports an Initialize function and also one function for each rpc in each proto service implemented by the trace_processor (a subset of protos/perfetto/trace_processor/*). The Initialize method takes two functions which are implemented on the javascript side: - reply This is how the trace_processor replies to rpc requests. - read This is the interface the trace_processor uses to read parts of the trace file. ui/src/engine/wasm_bridge.ts is the typescript counterpart to wasm_bridge.cc. It lives in a dedicated worker thread. Initialization flow is: 1) The worker thread (wasm.ts) starts up creating WasmBridge 2) this starts up the WASM 3) the worker calls WasmBridge#initialize 4) WasmBridge#initialize (asynchronously) waits for the WASM module to start up to complete (including WASM compilation) 5) WasmBridge#initialize (asynchronously) waits the file to arrive 6) WasmBridge#initialize calls the C++ initialize, the C++ begins reading the trace via the provided read method. 7) The C++ initialization is complete we call the reply method 8) The ts WasmBridge issues any calls to the WASM which had been queued while the C++ was setting up. Change-Id: Iec86ea06124b0405d7cf16c509c37a5570406930
Perfetto is an open-source project for performance instrumentation and tracing of Linux/Android/Chrome platforms and user-space apps.
It consists of:
A portable, high efficiency, user-space tracing library
designed for tracing of multi-process systems, based on zero-alloc zero-copy zero-syscall (on fast-paths) writing of protobufs over shared memory.
OS-wide Linux/Android probes for platform debugging
Web-based frontend
A UI for inspection and analysis of traces (coming soon).
Batch processing of traces
A python / C++ (TBD) library for trace-based metrics (coming soon).
Perfetto is building the next-gen unified tracing ecosystem for:
The goal is to create an open, portable and developer friendly tracing ecosystem for app and platform performance debugging.
Designed for production
Perfetto's tracing library and daemons are designed for use in production. Privilege isolation is a key design goal:
See docs/security-model.md for more details.
Long traces
Pefetto aims at supporting hours-long / O(100GB) traces, both in terms of recording backend and UI frontend.
Interoperability
Perfetto traces (output) and configuration (input) consists of protobuf messages, in order to allow interoperability with several languages.
See docs/trace-format.md for more details.
Composability
As Perfetto is designed both for OS-level tracing and app-level tracing, its design allows to compose several instances of the Perfetto tracing library, allowing to nest multiple layers of tracing and drive then with the same frontend. This allows powerful blending of app-specific and OS-wide trace events. See docs/multi-layer-tracing.md for more details.
Portability
The only dependencies of Perfetto's tracing libraries are C++11 and Protobuf lite (plus google-test, google-benchmark, libprotobuf-full for testing).
Extensibility
Perfetto allows third parties to defined their own protobufs for:
Allowing apps to define their own strongly-typed input and output schema. See docs/trace-format.md for more details.