Linux Finally Starts Removing Support for Intel's 37-Year-Old i486 Processor (phoronix.com)
- Reference: 0181357884
- News link: https://linux.slashdot.org/story/26/04/06/0358212/linux-finally-starts-removing-support-for-intels-37-year-old-i486-processor
- Source link: https://www.phoronix.com/news/Linux-7.1-Phasing-Out-i486
"A patch queued into one of the development branches ahead of the upcoming Linux 7.1 merge window is set to finally begin the process of phasing out and ultimately removing Intel 486 CPU support from the Linux kernel."
More [2]details from XDA-Developers :
> Authored by Ingo Molnar, the [3]change , titled "x86/cpu: Remove M486/M486SX/ELAN support," begins dismantling Linux's built-in support for the i486, which was first released back in 1989. As the changelog notes, even Linus is keen to cut ties with the architecture: "In the x86 architecture we have various complicated hardware emulation facilities on x86-32 to support ancient 32-bit CPUs that very very few people are using with modern kernels. This compatibility glue is sometimes even causing problems that people spend time to resolve, which time could be spent on other things. As Linus recently remarked: 'I really get the feeling that it's time to leave i486 support behind. There's zero real reason for anybody to waste one second of development effort on this kind of issue'..."
>
> If you're one of the rare few who still keep the decades-old CPU alive, your best bet will be to grab an LTS Linux distro that keeps the older version of Linux for a few more years.
[1] https://www.phoronix.com/news/Linux-7.1-Phasing-Out-i486
[2] https://www.xda-developers.com/linux-7-1-finally-dropping-support-intels-37-year-old-486-processor/
[3] https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?h=x86/platform&id=8b793a92d862c89055daa97ffa61a6929cf732f9
Re:Typical Stupidity (Score:5, Insightful)
name one other mainstream os that supports 486 processors.
Re: (Score:2)
There some business people that use OS/2.
Re: (Score:2)
"Support", not "in use". IBM sure as hell doesn't support OS/2 any more.
Re: (Score:1)
OS/2 isn't going to remove 486 support anymore than linux kernel 2 is going to remove it.
OS/2 has two future branches still in production.
eComStation which requires a 586 or newer, and ArcaOS which requires a 686 or the "pentium pro" version of the 586.
Neither one support 486.
Re: (Score:2)
From the most current OS/2 release:
"Hardware Requirements
Intel Pentium Pro or higher, or an AMD Athlon or higher. 64 Bit CPUs are supported (however ArcaOS will run in 32-bit mode). Computers with ARM CPUs are not supported. Apple Computers are not supported (regardless of CPU). The Vortex86 CPU is not sufficiently compatible to run ArcaOS and is not supported."
i.e. minimum hardware requirements are a 686 instruction set.
Re:Typical Stupidity (Score:5, Informative)
Not sure what you mean by "mainstream" but the BSD distros do. :)
Re: (Score:2)
NetBSD?
Re:Typical Stupidity (Score:4, Informative)
> No year will ever be the year of linux on the desktop with this stupid attitude of throwing working code away.
Who is still using a 486 and also needs modern kernel features? Let them run an older kernel. They are going to have to run older software anyway because a 486 with more than about 16MB is rare, and modern Linux distributions require multiple GB of RAM.
Re: (Score:1)
> Plenty of 486 around that had more than 16M of RAM, especially these days.
Impossible to figure out what you think you meant while you're mixing tenses. 486s with more than 16GB of RAM were extremely scarce and now they are even scarcer, they are not more common. Probably they are a larger percentage of the remaining 486s, but that is not the same thing as there being plenty around.
Re: (Score:1)
> if you can't figure that out I can only guess it is probably due to you mixing up M & GB.
Oh, are you one of the stupid clowns I've corrected on b and B? I bet you are.
> Plenty of 486 back in the days had more than 16M of ram.
This is bullshit. A percent of a percent of 486s had more than 16MB.
Re: (Score:2)
> and if it was coming from you I'd still ask the worst possible AI to verify your claims
OK bot. There isn't even a "you".
Re:Typical Stupidity (Score:5, Insightful)
> and also needs modern kernel features
This is a part everyone seems to miss when the get freaked out about Linux itself or some distribution dropping support for something 30 years old...
In 2026 if you are still using a computer older than mid-90s (and very more than likely even one from after the mid 90s) it is because it is part of some very specific process that almost certainly has you not making changes, which are almost certain to include software changes too.
Just because Linux 7.x can't be built for i486 any more does not stop you from grabbing any prior version and using that. Thinking about 486s specifically, I know there are actually a lot of odd things like hardened industrialized PCs and some routers and the like running licensed 486 cores and late manufacturing Intel parts; that are still in use. You can even still buy some new. It would not surprise me to learn people are running Linux on a good number of them, it would surprise me to learn people are running Linux newer than 5.10 or 5.15 on them. Even in the most exotic memory configurations a 486 is going to top out at 3.5GBs of memory, I guess you could do nearer to 4GB on a ISA only system (No PCI or VLB). You really going burn 16MB or more of that just on the kernel?
Let's be real if you are running a 468 you are probably using using Linux 2.0 - 2.6 already. Not being able to use 7 hardly affects you.
Re:Typical Stupidity (Score:5, Informative)
> Using IOT devices with kernel 2.6 in these days is just asking to be hacked.
Not really...
Almost all IoT devices work by phoning home. They call some remote server, and do some API stuff, send some message poll for new messages / instructions. They tend to have very little if anything listening.
If they do get onwd its because the infrastructure that supports them gets compromised, at which point its really the infrastructure that was hacked and not the device. The other thing that happens - all the gosh darn time - is what ever little web based interface they have for setting up wifi/IP settings/etc is some terrible CGI thing with some form of injection vulnerability. Again though if that gets pwnt, it is only after some ofther failure of your internal network security. That is a concern, I understand defense in depth, I get foothold and dwell time issues, However a newer kernel won't prevent that kind of compromise. Lack of shell escaping on calls to system() or bad choices around using eval() will get you popped on Linux 7.0 as easily as 2.0.
Re: (Score:2)
> Almost all IoT devices work by phoning home. They call some remote server, and do some API stuff, send some message poll for new messages / instructions. They tend to have very little if anything listening.
Are you talking about professional well made IoT devices designed for corporate management? Because holy shit are you wrong about general consumer IoT devices doing no listening. There's a reason for the running joke that the S in IoT stands for security.
In fact much of the community driven IoT interfaces for tinkerers rely on the fact that someone has hacked a device almost universally via an active open listening port to force it to work with something other than it's Cloud service.
Your best beat at secur
Re: (Score:2)
there is distros built to run on low ram tiny core dsl puppy. then when you get into just having cli nedding even less.
Pray tell, what modern desktop runs in 64MB of RAM (Score:5, Informative)
...Because that's the upper limit of high-end 486 motherboards.
The 80486 was essentially e-waste by the year 2000. Even ultra-conservative Debian completely dropped support for the i486 over decade ago (with Squeeze going out of LTS in Feburary of 2016 after a 5 year run)
Incidently, the first Linux install I ever performed was in early 1997, on an ISA-only 486DX33 motherboard +200MB pre-DMA IDE drive that I literally rescued from the trash.
Re: (Score:2)
The 80486 processor was _technically_ capable of addressing 4GB of RAM. But given that the largest 30-pin SIMM produced was limited to 16MB (due to 24-bit addressing) in practice the upper limit was 128MB, using 8* 16MB SIMMs. (I'm not aware of any mass-market motherboard with over 8 30-pin SIMM slots, but I'm willing to be corrected) Late 486 boards supported 72-pin SIMMs which could theroretically go up to 128MB, but the largest FPM (not the later/faster EDO!) SIMM I was able to find is 64MB, and those
Re: (Score:2)
For the anecdote, i saw adverts for 486 motherboards which accepted 128 Mo or RAM before the 486DX2 was even available. I never tested them, of course.
Later, there were also souped-up versions of the 486 like that AMD with a quadrupled clock multiplier you could run at 160 Mhz.
I wonder (i don't know) if there are more modern clones running at a higher clocks for specialized tasks, maybe because a 486 design would be immune to some vulnerabilities and its design is out of patent right now.
A bit like those 6
Re: (Score:2)
the vortex86dx SOCs seem to be somewhat along those lines.
I don't know what cpu they are copying but it is used in rasteri's tiny WeeCee
Re: (Score:2)
> and even 486 could go beyond 64M of RAM.
Could and Did are two distinct words in the English language. Very few 486 machines ever existed with more than 64MB of RAM. They were for insanely niche applications. Now we change the debate from do we support what is today an incredibly rare architecture, to do we support what is today an incredibly rare architecture for the purpose of a niche that almost certainly doesn't exist anymore on that platform?
We can keep going down this rabbit hole of "but it did support", only to find there's a single machine
Re: (Score:2)
I have a high-end 486 motherboard with four 72-pin SIMM slots and PCI slots that accepts up to 192MB of RAM. For some reason it can't handle four 64MB SIMMs so it won't do 256MB, but it'll handle three fine. Or one 128MB and one 64MB one. But adding anything more makes it not boot and it seems to be a chipset limitation. At one point a couple years ago I had it maxxed out at that complete with an AMD 5x86 120mhz, PCI USB, video, and sound cards and Windows 98 and it did work but... other than for funsies th
Re: (Score:2)
"Those who forget the past are doomed to repeat it."
Linux literally is "repeating the past". Its very name is derived from the designs it copies.
"No year will ever be the year of linux on the desktop with this stupid attitude of throwing working code away."
LOL no, that's not it.
Re: (Score:3)
Who is throwing away working code? New versions of Linux will not support x486 processors. The keyword is "new". Existing versions will still work. All versions are archived for people to use. And Linux is still open source. Anyone who want to modify Linux going forward to support 486 is free to do so.
Re: (Score:1)
You are welcome to continue development on support for the 486 if you need continued support for it. The kernel is open source and you can easily clone yourself a local repo and continue 486 support maintenance while merging in new features as time goes on.
My first Linux CPU (Score:3)
We were I/O bound with a saturated network. Our boss bought a Pentium to see if it would help. He didn't understand what "I/O bound" meant.
Went to a Comdex and saw this new fangled thing called a network switch. 3-4 of those in our system solved our problems.
Not a 486 thing, but... (Score:4, Interesting)
Somebody thought it'd be a great idea to remove full 10Mb Ethernet support from two recently-purchased routers I tried at home (bought the second after the first didn't work).
Turns out this would've cost me my venerable and much-loved Roku Soundbridge M500 and M2000 network music players, which are working just fine, thanks.
I had to buy a cheap switch to put between them to straighten this out. Waste of money.
I understand that 486 support takes up needed and scarce dev resources, and it seems reasonable to me to remove it.
But I wonder what hidden breakage (like my case) happens as a result of making these "reasonable" decisions.
Re: (Score:2)
spending effort to maintain support for stuff nobody is using is not reasonable. Would you write a book, if you were certain nobody would ever read it or even want to?
Don't say "but what about a diary," even a private diary generally has one intended audience if it is the author themselves, for their own recollection.
writing software that no computer will ever run makes very little sense, even from an educational standpoint.
- as to the hidden breakage. Probably not much, because if you don't know about th
Re: (Score:2)
> The software is already written.
Yes it's already written. Use a kernel with the code still there. It's not like your 486 will have any application that requires the latest kernel, if your system even manages to boot at all.
The problem with written code is that if it remains "supported" it places a burden on all other code changes made to the product. Someone needs to do regression testing to make sure it's not broken. Someone needs to do security auditing and potential bug fixing. And above all, these are not reasonable requirements for h
Re: (Score:2)
To me, the 'fix' seems to be spinning up a new arch branch, call it x86-old or whatever. Use that to bring up a clean start 386/486/etc arch so that people playing in the main x86 area aren't tripping over the old code. We've got all the info needed to do so already sitting there in the src. Now, if there are enough users left to drive such a branch, dunno. I still play with my Soekris 4501 and 4801 as a hobby but I'm no longer using them in production, and my C knowledge isn't good enough to make this happ
Re: (Score:2)
Realistically, approximately none of the 486 code has had more than some automated testing in a virtual environment in a long time. We should assume it is bitrotted and only suitable for use on toys where subtle bugs in cache or memory management that leaked in from newer generations won't kill anyone. The time to fork the entire 486 arch and have a trustworthy result passed long ago.
It's the same situation as when Reiser3 was removed. Anyone who was using it in a new kernel could tell that performance w
Re: (Score:2)
My $DAYJOB's data center switch upgrade got switches that have 48x 10G SFP+ ports plus 8x 100G QSFP+ ports. When we installed them, we realized that some really old Dell Poweredge servers try to drop to 100M when using shared DRAC (with the dedicated DRAC port also being 10/100M only), and the switches don't support 100M. We also had to look at a bunch of rack PDUs to find options that were 1G rather than 10/100M.
100M uses less power than 1G, so I guess that's why Dell did that (sounded like a good idea to
Re: (Score:2)
My (admittedly anecdotal from the totally unscientific sample of random stuff I've had reason to work on) impression is that some 'shared' BMC ports had oddities related to network controller sideband interface speeds, since NC-SI is what the BMC is depending on if the NIC is on someone else's PCIe root. It's not like the BMC actually needs a faster link for much(normal management traffic probably doesn't fill 10mb and mounting virtual media may be literally once-in-a-lifetime) so the actual speed of the NC
Re: (Score:2)
Again, it's about power saving. Idle 1000BASE-T draws noticeably more power than idle 100BASE-TX (IIRC the drop from 100BASE-TX to 10BASE-T is not as significant). There are Energy Star ratings and EU rules about how much power an "off" and "standby" device can draw, so dropping to a lower NIC speed helps reach those levels.
There was a proposal for a "low power idle" mode extension of 1000BASE-T, not sure if that went anywhere or got implemented.
Re: (Score:3)
It's not that "they removed 10Mbps support" so much as the underlying ethernet hardware (MAC and/or PHY) used by that new router simply doesn't support it. Might not even support 100Mbps either; once you cross into multi-gigabit world, sub-1Gbps support is the exception rather than the rule. Why? Because that would require a more complex (ie expensive) design and be utilized by almost nobody.
Re: Not a 486 thing, but... (Score:2)
Sure. I understand well that things change.
However, it'd be nice if the box had a big yellow burst that said: "Now Without 10Mb/s Support That Has Worked Until Now On The Same Connector Ever Since We Came Up With It!"
Sucks to get it home and find that this one is borked, too.
Re: Not a 486 thing, but... (Score:2)
Using WoL (wake-on-LAN) also tends to use 10 Mbit Ethernet. There isnâ(TM)t an LED light combination for that speed on my 2.5 Gbps NetGear switch (so the port lights are all off) but otherwise it still works.
Re: (Score:2)
The answer is rather simple: don’t upgrade the kernel to a newer version. At this point, any kernel features added would not benefit a 486. Existing and working code is unaffected.
Re: (Score:2)
Actually, many consumer gigabit Ethernet switches lack 10Mbps support these days. They are 100/1000baseT only.
Business and enterprise switches though I've found (including Cisco ones, which you can find dirt cheap used) still are 10/100/1000Mbps. Even newer business and enterprise class switches retain support.
Of course, once you step into 10Gbps Ethernet, you have to be careful because many only are 10Gbps only, while some do support 1/10Gbps. 2.5Gbps support is iffy unless it's specified which is annoying
I was kind of shocked to find out how expensive (Score:2)
386 and 486 CPUs were back in the day. I used a commodore 64 for most of my school life and a word processor / typewriter when I needed something that could print text better than the ancient thermal printer I had (no joke I had an okidata thermal printer for my commodore. Quality was good but when you eventually couldn't get the regular paper you had to buy the rolls and I had the jury rig a feeder)
A lot of people complain about the Sega 32x and Atari Jaguar ports of Doom but it was mind-blowing to be
486 seemed magically advanced in the mid 1990s. (Score:3)
My first Linux installation was Redhat 3.03 on a 16MHz 386/SX system in mid-1995. For those of you without an AARP card, that's a 32 bit CPU with a 16 bit bus, which Intel released to cannibalize the market for the 286, which did not have a memory management unit. That means no swapping, you run out of ram, it was game over.
I think the 486/25 that replaced the 386/SX arrived in ... 1996 ... and it had an astonishing *eight megabytes* of memory. I had kept a one megabyte LIM/EMS 4.0 physical memory card from my 286 when I got the 386/SX, and that actually mattered with Windows 3.x. I put it in the 486, but given that vast eight megabyte expanse of dram it didn't last long.
Then in late 1997 my employer went bankrupt and as part of the dissolution I brought home the dual Pentium 133 system with 32 megabytes of ram. I remember all my IRC friends were so jealous of that monster ...
As long as my Pentium 90 still works (Score:2)
I'll be happy.
Interesting fact (Score:2)
Intel 486, and CPUs before it, did not include an unique ID. Pentium and all CPUs after it each have their own unique hardwired ID per unit.
But I guess using a 486 today is pretty unique anyway, so it doesn't really matter. :)
I was there... (Score:3)
I was there, three thousand years ago...
Ok, well it was only ~35 years ago, but I well remember cobbling together installable floppy images from Usenet to get Linux running on my 486DX with a bunch of GNU utilities. This took many hours of downloads and preparation over a dial-up connection, but this was the only way to install because even SLS hadn't come out with a coherent Linux distribution yet.
My 486 system had a whopping 4MB of RAM with a 200MB hard drive (my first). I massively overpaid for it and charged it all on my shiny new Circuit City credit card while I was still in college.
At my student job, I had an awesome, monochrome DECstation 3100 running Ultrix 3.1, so the thought of being able to run UNIX at home was just awesome.
Those were the days. :)
5x86 DX/133 (Score:1)
My very first linux box, which I still have and is still running today, is still on RedHat 3.0.3 that I got on a CD in a book from the Media Play in Poughkeepsie NY in 1996. Granted it is completely useless except as a samba server sharing the 1.6GB hard disk that is still in it (and still works). But, I keep it for posterity, and because I like having a monitor with xearth on it.
I could probably put a newer distribution on it but with only 24MB of RAM, the newer stuff would choke out on it.
Re: (Score:2)
> I like having a monitor with xearth on it.
You can't grab the [1]source [xearth.org] and build it for a newer architecture?
[1] https://xearth.org/
question (Score:1)
Can anybody seriously explain the use case for a still running mission critical system or something like that that may need a new kernel? is it feasable really? not trolling but trying to learn. second is why do they need an emulation in the kernel for old cpu's? i thought x86 was backwards compatible?
Hubble out of support (Score:5, Interesting)
Oh... How will Hubble do, since it runs a 486 DX 4 ? :)
Re: Hubble out of support (Score:4, Informative)
It uses VRTX, reportedly. Linux wasn't suitable as a real-time OS when the Hubble was designed, or really even when the Hubble got the 486 installed in 2009.
Re: (Score:3)
Linux isn't suitable as a real-time OS now either strictly speaking. In fact that one of the top hits from a search on Linux RTOS is a paper from NASA (from a comparatively recent 2019) discussing the performance of Linux with every RTOS relevant kernel feature set into the most ideal position. Their conclusion was... well you probably will hit your event deadline if you throw fast enough hardware at it, but it is still nothing like a true RTOS.