Address performance degradation of Total Blocking Time (TBT) in Project Merge Request Discussions page to ~10s
The issue has been updated to track degradation of MR discussions seen in #331172 (comment 1252417974)
The main issue for performance improvement is moved to #395192
The Project Merge Request Discussions page's browser Total Blocking Time (TBT) performance is unperformant:
NAME | FCP (ms) | LCP (ms) | TBT (ms) | SI (ms) | LVC (ms) | TFR SIZE (kb) | SCORE | RESULT
--------------------------------------|----------|------------------|------------------|---------|----------|---------------|-------|-------
web_project_merge_request_discussions | 576 | ✓ 716 (<2500) | ✓ 5136 (<8000) | 1336 | 11100 | 1032.8 | 65 | Passed
The Project Merge Request Discussions page being tested can be seen here on Staging. This page is showing overview tab on MR page, below is the screenshot of page's historical test TBT results.
Test Details
Testing was done on our 10k Reference Architecture environment with our lab condition SiteSpeed pipeline (named GitLab Browser Performance Tool internally). The project being tested is a copy of gitlabhq (tarball can be found here). GitLab Browser Performance Tool tests information is listed at Current test details page.
The latest GitLab Browser Performance pipeline results can always be found here. Through this page the full SiteSpeed report can be found via the SiteSpeed Report
link on that page.
As per our performance targets this page's TBT metric is above the target of 2000 ms which is severity2. Task is to improve the page's browser performance into next tier.