Post #1890225

2026-04-25 10:15 UTC

@kernellogger Heh .... ok ISDN and the 3c509 stuff is sad.. truly.. But at ax25 and amateur/ham-radio stuff, it may worth to wait a bit for their removal, as HAMs just recently start to discover linux for their needs.... Usually their stuff is always windows based because everyone has windows.. But they get tired of microslop too .... so still having native ax25 and others ham radio related stuff might be a good thing ... at least for a while still.

Replies (1)

@dazo@infosec.exchange 2026-04-25 16:42

@hackbyte @kernellogger I don't know much about the hamradio stuff, but do understand the removal can cause debate. I do think it might be worth considering reason why it really need to be in the mainline kernel. Convenience for end-users alone is not enough if nobody is actively maintaining and testing it closely with the upstream kernel developers. No developer or maintainer finds it interesting maintaining code which nobody seems to use. I see two alternative options: Have these drivers as out-of-tree modules, where interested users can collaborate on the further development. If there appears to be a larger than expected demand for these drivers, it may be accepted in the upstream mainline kernel again. The other option is to consider if these drivers really need to be in kernel space at all. Just look at all the USB devices available, there are many which just builds on a very generic kernel API and does the rest entirely in user-space. I wonder if the RDMA stack is somewhat similar too. Maybe that is just as good as a viable path? Instead of criticising kernel maintainers removing code they need to maintain which they don't feel enough people are using - look at this as an opportunity to revitalise the interest for this driver, from outside of the mainline kernel. Only time can tell if it was the right thing to do or not.

Open ##1890226