| commit | 83929789340de23f0198cfd33b7865256a67172a | [log] [tgz] |
|---|---|---|
| author | Stephen Nusko <nuskos@google.com> | Wed Jul 29 12:31:32 2020 +0100 |
| committer | Stephen Nusko <nuskos@google.com> | Wed Jul 29 12:31:32 2020 +0100 |
| tree | ae8c6d1b1f140afe14cfafe042e7b82c5f84cda0 | |
| parent | 16c50deb5eb4c163a18666986ee9e28bf735e1a9 [diff] |
Add new metric for queuing delay scroll jank cause. This metric finds slices that could have blocked the next step in the critical flow of a scroll from executing. It does this by looking at the period of time between steps collecting all slices and figuring out the overlap. This allows us to count how often a certain task appears between janky and non-janky scrolls. In the process of porting our current version of this metric and adding the diff test I found 2 bugs related to including the current slice stack which means we were inflating numbers for events that happen as a part of the critical flow. Bug: 159606819 Change-Id: I0cf7d01b542e36dca49987b80729bd407877f4a7
Perfetto is a production-grade open-source stack for performance instrumentation and trace analysis. It offers services and libraries and for recording system-level and app-level traces, native + java heap profiling, a library for analyzing traces using SQL and a web-based UI to visualize and explore multi-GB traces.
See https://perfetto.dev/docs or the /docs/ directory for documentation.