v8/tools/heap-stats
Michael Lippautz d6daf8b0af [object-stats] Record details for BytecodeArray and Code
FixedArrays hanging off recursively of the constant pool without any
real objects in between can be considered as meta data. They are shared
with optimized code (embedder pointers).

Bug: v8:7266
Change-Id: I4006675e17e8eea3bdc8565254d80e2ffece0ad0
Reviewed-on: https://chromium-review.googlesource.com/883361
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50837}
2018-01-24 14:32:22 +00:00
..
categories.js [object-stats] Record details for BytecodeArray and Code 2018-01-24 14:32:22 +00:00
details-selection.html [object-stats] Visualizer: Show percentages in details selection 2018-01-22 16:30:05 +00:00
details-selection.js [object-stats] Visualizer: Reset on Isolate change 2018-01-24 09:17:23 +00:00
global-timeline.html [tools] V8 heap stats visualizer 2018-01-15 15:27:17 +00:00
global-timeline.js [object-stats] Add histogram 2018-01-18 11:57:35 +00:00
histogram-viewer.html [object-stats] Add overall counter to viewer 2018-01-22 09:28:33 +00:00
histogram-viewer.js [object-stats] Add overall counter to viewer 2018-01-22 09:28:33 +00:00
index.html [object-stats] Unpack gzipped trace files on the fly 2018-01-23 15:01:02 +00:00
README.md [object-stats] Unpack gzipped trace files on the fly 2018-01-23 15:01:02 +00:00
trace-file-reader.html [tools] V8 heap stats visualizer 2018-01-15 15:27:17 +00:00
trace-file-reader.js [object-stats] Unpack gzipped trace files on the fly 2018-01-23 15:01:02 +00:00

Heap Stats

Heap stats is a HTML-based tool for visualizing V8-internal object statistics. For example, the tool can be used to visualize how much heap memory is used for maintaining internal state versus actually allocated by the user.

The tool consumes log files produced by d8 (or Chromium) by passing --trace-gc-object-stats or a trace captured using Chrome's tracing infrastructure. Chrome trace files can either be processed as gzip or raw text files.

Hosting requires a web server, e.g.:

cd tools/heap-stats
python -m SimpleHTTPServer 8000