You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
There are a few aspects to the forkst maintenance to be discussed. One is the branching model we use right now. The other is how to approach to the fork maintenance (and keeping it up to date) in general.
Keeping the per-platform branches is becoming more and more effort as the number of platform increases.
Features and fixes are out-of-sync between the branches. Often some extra work is needed to apply the same set of fixes on multiple branches. It it easier to miss some stuff this way.
Merging changes from upstream requires more and more effort. There are a lot of changes and it is easy to miss something.
Maybe using patch files would be easier to track our custom changes and send them upstream.
Describe the solution you'd like
TBD
Where is the value to a user, and who might that user be?
Dasharo developers can spend less time rebasing coreboot/edk2/etc repositories to latest versions.
Dasharo developers can have easier time extracting patches from Dasharo forks for sending them upstream.
Describe alternatives you've considered
There are a lot of approaches out there used by different projects:
To some extent fork maintenance is related with versioning. Because of that I recommend reading Debian and Fedora versioning schemes to understand what their maintainers learned over years. Interestingly a lot of corner cases were covered.
This problem recently occurred in #314 and while preparing PC Engines release 4.19.0.1.
Versioning problem is little bit related with fork maintenance. If versioning topic do not belong here we probably should create another RFC issue.
I'm getting through the PC Engines v4.19.0.1 rebase process, which is excruciating. During the discussion with Andrew Cooper on the matrix channel, he mentioned that they track patchqueue in dedicated Xen repo. Repo in itself uses guilt.
The problem you're addressing (if any)
There are a few aspects to the forkst maintenance to be discussed. One is the branching model we use right now. The other is how to approach to the fork maintenance (and keeping it up to date) in general.
Maybe using patch files would be easier to track our custom changes and send them upstream.
Describe the solution you'd like
TBD
Where is the value to a user, and who might that user be?
Dasharo developers can spend less time rebasing coreboot/edk2/etc repositories to latest versions.
Dasharo developers can have easier time extracting patches from Dasharo forks for sending them upstream.
Describe alternatives you've considered
There are a lot of approaches out there used by different projects:
Additional context
The text was updated successfully, but these errors were encountered: