Update on Flutter in AGL and IVI Expert Group
summary
AGL now runs Flutter apps via a reproducible build and dev flow: a Python Flutter workspace automation script pulls specific Flutter SDK and engine builds, an AGL QEMU image and custom device configs in about two minutes, then runs Flutter Auto with a Vulkan backend on host or QEMU. Debugging works through VS Code; profiling/release remote workflows are not yet supported. Platform views can halve frame rate, platform channels add ~10 milliseconds per message, while FFI is zero‑latency.
AI-written summary · about
chapters (15)
- 0:00Flutter in AGL and IVI
- 1:30Flutter components and benefits
- 2:57Workspace automation demo
- 9:26Workspace automation explained
- 14:02Debugging Flutter Auto on QEMU
- 21:23Custom device config
- 34:38Debugging scenarios and limits
- 37:28Linux GTK platform overview
- 41:41Platform channels and FFI
- 44:52Embedded build environment
- 51:51Supported devices and performance
- 52:34Flutter Auto home screen
- 58:14Flutter Auto CLI options
- 1:00:45Profiling and troubleshooting
- 1:01:49Q&A
related talks
44mFlutter for the Discerning Android Developer
31mFlutter-workspace Tool by Example - Setup for AGL Terrific Trout
28mAMA with Brian Egan and Filip Hracek
23mFlutter For The Discerning Android Developer
49mFirst Flutter, then Fly: Build Reactive & Performant Cross-Platform Apps
24mKeynote: Flutter
49mWhat's Awesome about Flutter
43mIntroduction to Flutter