No, It's Never Too Late to Upstream Your Legacy Linux Based Platform
summary
Argues that it is never too late to upstream a legacy Linux board and gives concrete workflows: collect out-of-tree patches, extract vendor blobs from obscure GitHub repos or tarballs, and submit staged patches to mainline using git and patchwork; claims push changes to mainline as the goal and shows that fast upstreaming reduces long-term maintenance by cutting duplicate backports. Details time and copyright traps, lists typical sources (Qualcomm, Atmel, Freebox), and offers pragmatic tips so ports become maintainable rather than brittle.
AI-written summary · about
chapters (8)
- 0:00Upstreaming legacy Linux platforms
- 5:35Why legacy platforms resist upstreaming
- 10:39Benefits of upstreaming legacy code
- 16:30Simplifying upstream contributions
- 22:43Overcoming upstream barriers
- 28:21Real-world upstreaming examples
- 33:01Sustainable upstream strategies
- 35:25Q&A
related talks
36mNo, It's (Still) Never Too Late to Upstream Your Legacy Linux Based Platforms- 36mNo, It's (Still) Never Too Late to Upstream Your Legacy Linux Based Platforms
44mBackporting is so 1993
58mUpstreaming 101
44mBackporting is so 1993
10mUpstreaming a linux kernel driver.
53mUpstreaming 201
45mBOF: Corporate Use of Embedded Linux