Now that we have updater support for RW_LEGACY, ship SeaBIOS updates to all Rambi and Auron derivatives |
|||||
Issue description[continuation from https://groups.google.com/a/google.com/forum/#!topic/chromeos-firmware/ZsJ5aoei-7I ] Since we've done all the work to allow updating the legacy firmware section, we should actually use it. We know that all Rambi (BayTrail) and Auron (non-Samus Broadwell) based Chromebooks shipped without functional SeaBIOS, even though we now have updates available to fix that. We just need to add the hash of the current legacy blob as LEGACY_UPGRADE_WHITELIST to the respective firmware ebuild and update the updater image with a newer version. Since this is a completely different flash section that's not touched by the normal mode boot flow, maybe we can convince the TPMs and QA to allow this update without doing a whole firmware test run again? We could just replace the current stable BCS image with one that has the new legacy section dd'ed on, without uprevving any of the other components. (The current updater aborts early from autoupdate before checking need_update_legacy, but it would be easy to change that to make it only update the legacy section but touch nothing else in that case.)
,
Mar 4 2016
If we will have lots of boards needing legacy update, I'd recommend to: 1. Change the LEGACY_UPGRADE_WHITELIST into a ECLASS variable in cros-firmware.eclass 2. Change firmware updater pack script to take that variable 3. So we can easily define that in ebuild files instead of writing customization shell scripts. And in updater we can do "if LEGACY_UPGRADE_WHITELIST is not empty, try legacy update first", to solve jwerner's request (push legacy without changing main FW ID)
,
Mar 4 2016
,
Mar 28 2016
,
Aug 23 2016
,
Aug 22 2017
We have the mechanism now, but for Auron/Rambi (where this was somewhat interesting compared to its contemporaries) that ship has sailed a long time ago: whoever is interested in running seabios on those probably moved to builds from johnlewis.ie already |
|||||
►
Sign in to add a comment |
|||||
Comment 1 by reinauer@chromium.org
, Mar 4 2016