• Empower communities who depend upon their mobile devices to access the internet on ARMv6 phones - remember that in emerging markets, more users access the Internet via their mobile browser than a desktop
• Introduce support for a new platform in order to increase market installs & share - this also serves as a further differentiator against some of our competition by supporting a varied and large breadth of devices
• Current goal is to support a conservative number of ARMv6 devices in Fx16
It is expected that performance will not be on par as on ARMv7 devices, however overall browsing performance must be usable from a UX perspective for top mobile sites in our supported locales
Feature page: https://wiki.mozilla.org/ARMv6_support_-_phase_1
Getting down to 600 MHz, 256 MB RAM, 240x240 resolution doubles our addressable market from 40% to 81% of all Android devices.
- Competitive Analysis <== we need this
- Start-up time comparison between top 3 ARMv6 devices to top 3 ARMv7 devices to stock browser on the top 3 ARMv6 devices to Opera Mobile
- Performance (scrolling, panning/zooming, checkerboarding) comparison between top 3 ARMv6 devices to top 3 ARMv7 devices to stock browser on the top 3 ARMv6 devices to Opera Mobile
- Top browser feature comparison
- Install size of stock browser and Opera Mobile
Overall Value Proposition <== we need this
- Stability: Less than 5 crashes per ADU
- Performance: On par with stock browser on top 3 ARMv6 devices
- UX: Installs and allows typical browsing to top mobile sites
- Usability: Installs and allows typical browsing to top mobile sites
- Feature Parity:
- Minimum System Requirements: ARMv6, 800 MHz, 512 MB RAM, OpenGL support
- Install size:
• Where are we today?
• What's next?
- No expected impact on Sync from ARMv6. There are potential impacts due to resource-constrained devices (e.g., OOM, slow syncs, poor system responsiveness during syncing), but we hit this today with older ARMv7 devices. C'est la vie.
- Prepare new set of graphic assets for low density displays (ETA mid-september)
- bug 716403 Enable scrolling title bar, to maximize the amount of viewable web content on screen
- Revisit tabs UI to optimize for small screen heights (could include opening the tabs tray wider, or scrolling thumbnails horizontally, etc)
- Checking behaviours on devices with hardware keyboards -- does the interaction feel right?
- Awesomebar / search suggest
- Find in Page
- Form input / navigation
- What other data do we need in order to understand what the UX requirements are or do we know this information already?
- It would help us to see a list of any other UX regressions that have been found to date that have not been mentioned above.
Where are we today?
• We have a good baseline in terms of variety testing and an initial bug hunt. However, need to clarity on the quality goals
• Initial Compat Testing
- Armv6 is currently being reported mixed with Armv7. Currently the easiest way to separate them is to look via separate devices through Kairo's report:
- bug 777372 has been done on the soccoro end
- there still needs to be additional work to get the reports separated.
• Building MTD
• Outreach via other developer sites
• What else?
- Where are we today?
- working on getting builds/tools/tests to run on a beagleboard (armv6 hardware)
- working with releng to run armb6 builds on existing tegras (non arvm6 hardware)
- beagleboards are p2 behind existing automation stability and pandaboards online and testing
- What's next?
- resolve sutagent issues on the beagle board
- validate all tests
- build burnin script (similar to panda burnin script)
- document instructions for building image/flashing device
- probably a bunch of other smaller tasks
ARMv6 Reported Issues & Tracking Bugs
69 Total; 4 Open (5.8%); 63 Resolved (91.3%); 2 Verified (2.9%);