No, It's (Still) Never Too Late to Upstream Your Legacy Linux Based Platforms
summary
Argument: pushing legacy SoC and board patches into mainline Linux pays off. Example: a recent platform takes six months to prepare and achieves 90% of patches merged into 6.8, with remaining clock/interconnect pieces merging next cycle. Workflows use U-Boot, device tree bindings and headers, Linux-next, and a CI test suite and clean DTS/bindings; simple drivers can merge in weeks, complex platforms take many kernel releases or a year or more.
AI-written summary · about
chapters (9)
- 0:00Upstreaming legacy Linux platforms
- 2:11Why upstreaming matters
- 9:43Key benefits of upstreaming
- 13:58Efficient upstream workflow
- 17:41Time and effort for upstreaming
- 23:30Delays and challenges in upstreaming
- 29:25Upstreaming very old Linux ports
- 31:57Business and technical advantages
- 34:12Q&A
related talks
36mNo, It's (Still) Never Too Late to Upstream Your Legacy Linux Based Platforms
36mNo, It's Never Too Late to Upstream Your Legacy Linux Based Platform
58mUpstreaming 101
45mBOF: Corporate Use of Embedded Linux
44mBackporting is so 1993
39mThe Life of a Kernel Bugfix
48mStep by Step, What Should We Do for the Kernel Ecosystem?
51mPanel Discussion: Industrial Linux Beyond 2022