Linux 7.1 is dropping support for Intel's 486 processor. The chip launched in 1989. Ronald Reagan was still president. The Berlin Wall was still standing. And now, 37 years later, the Linux kernel is finally moving on.
A proposed patch from Ingo Molnar would remove CONFIG_M486SX, CONFIG_M486, and CONFIG_MELAN configurations. These are the flags that let you compile a kernel specifically optimized for 486 hardware. If the patch gets merged, future kernels simply won't boot on those processors. Not won't work well. Won't boot.
Linus Torvalds himself weighed in with characteristic bluntness: "I really get the feeling that it's time to leave i486 support behind." He questioned why developers continue wasting effort on hardware that's old enough to have grandchildren. The changelog captures his frustration clearly. There comes a point where backward compatibility stops being a feature and becomes an anchor.
The technical rationale is straightforward. Maintaining 486 support means keeping compatibility code in the kernel that creates overhead, occasionally causes bugs, and consumes developer time that could be spent on literally anything else. Every old architecture adds complexity. Every edge case requires testing. Eventually the cost exceeds the benefit.
But here's the thing Linux is famous for: supporting ancient hardware long past any reasonable expiration date. Run Linux on a Pentium. Boot it on a 486. Install it on that ancient laptop in your closet. The community wears this capability like a badge of honor. Dropping 486 support feels like an end-of-era moment because, in many ways, it is.
Reactions from the tech community mixed nostalgia with pragmatism. "Pour one out for the 486," wrote one developer. "That chip taught me assembly." Another was more practical: "If you're still running mission-critical infrastructure on a 486 in 2026, you have bigger problems than kernel compatibility." A third pointed out the obvious: "This is why LTS distributions exist. Your 486 isn't going to stop working. You just can't upgrade it."
Users with 486 systems still have options. Long-Term Support distributions will maintain older kernel versions for years. That ancient hardware won't suddenly brick. It just won't get new features, security patches for vulnerabilities that don't matter on a machine disconnected from the internet, or optimizations for hardware that doesn't exist yet.
There's a broader question here about longevity and sustainability. Linux's willingness to support old hardware has kept millions of computers out of landfills. That's genuinely valuable. But at some point, the marginal benefit of supporting a 37-year-old processor in cutting-edge kernel code becomes approximately zero. The developers who maintain this stuff are volunteers. They get to decide when enough is enough.
One commenter captured the moment well: "Linux supported the 486 longer than Intel did, longer than any commercial OS did, longer than anyone had a right to expect. That's a win, not a loss."
The 486 had a good run. It powered the early internet, taught a generation of developers how computers actually work, and stuck around long enough to see its own great-great-grandchildren. Now it gets to retire, knowing it lasted longer than any chip architecture has a right to expect. Thirty-seven years is long enough.
The technology is impressive. The question is what we're keeping alive and why. Sometimes the answer is because we can. And sometimes, after 37 years, the answer becomes because we can't anymore. Both are fine.
