back to article Linux kernel dumps 386 chip support

Linus Torvalds has announced the Linux kernel no longer supports Intel's 80386 processors. Reg readers will doubtless recall that the 386 debuted way back in 1985 and made something of a splash when the chip found its way into PCs made by Compaq before an IBM PC bearing the processor reached the market. 386s screamed along, …

COMMENTS

This topic is closed for new posts.
  1. Anonymous Coward
    Anonymous Coward

    386?

    Hell, I remember programming the 8080. It was a ceramic DIP chip, too - none of this new fangled plastic or BGA stuff.

    1. Evil Auditor Silver badge

      Re: 386?

      8080, bloody Hell! Almost forgot, that's where I started with assembler. And I enjoyed it. Guess that makes me kind of a nerd.

      1. Fatman

        Re: 8080, bloody Hell!

        I am from the same time period, as I cut my teeth on a Z-80, and I toyed with assembler.

        The fact that I recall systems with free space on a 90k SSSD 5-1/4 floppy, has to "date me".

        1. Kevin Johnston

          Re: 8080, bloody Hell!

          Pah.....at least with a proper 8" floppy you could just use a hole-punch (very carefully) to double your capacity

          1. Goat Jam

            Re: 8080, bloody Hell!

            @Kevin Johnstone

            Umm, you could do that with 5 1/4 inch disks. The day I learned to do that with my apple ][ floppies was a joyous day indeed.

          2. N2

            Re: 8080, bloody Hell!

            You could also double the capacity in the same way with the smaller 720K floppy as well :D

        2. P. Lee
          Happy

          Re: 8080, bloody Hell!

          Apple ][+ 140k SSSD FD IIRC. Ah, the joys of being able to hook the reset button! :D

          Go-go Miner 2049er (but not "lemonade stand" - that's not a proper game.)

          My neighbour had a Heathkit Z80 on which we used to play Y-Wing.

          You are in a maze of happy memories, all alike. On the wall is an inscription.

        3. Blane Bramble
          Thumb Up

          Re: 8080, bloody Hell!

          @Fatman - the fact that you know what SSSD means without thinking it is a mistype of a new fangled non-spinning disk thingy dates you (and me).

          1. Richard North

            Re: 8080, bloody Hell!

            I suspect that might date quite a few of us...

            Whether that is a good thing or a bad one I shall leave to the reader.

          2. Charles 9

            Re: 8080, bloody Hell!

            Don't you normally split the term in the middle? Then again, by the time I cut my teeth on a C128, most disks were being labeled DS/DD or 2S/2D. Then by the time I went with 486's you had DS/HDs, in both sizes.

          3. NogginTheNog
            Thumb Up

            Re: 8080, bloody Hell!

            Me too - then I remembered: Single Sided, Single Density!

        4. Crisp

          Re: Z-80

          I'm going to let you finish, but first I'd like to say that the 6502 was the best 8-bit processor of all time

          1. Zack Mollusc

            Re: Z-80

            ...but 6809 had multiply! Multiply! That made it practically a supercomputer!

            I was young.

            1. Chemist

              Re: Z-80

              "6809 had multiply! Multiply! That made it practically a supercomputer!"

              Integer multiply of course, but far, far better it has wonderful addressing modes.

              1. John Smith 19 Gold badge
                Unhappy

                Re: Z-80

                The 6809 seemed to be one of the best of the 8bit breed. Good addressing modes, multiple accumulators, orthogonal instructions. I think it had only 2 real problems

                Long execution times.

                Late in the field.

                Motorola made many microcontrollers with all sorts of on chip jiggery pokery, and whoever they are now (Freescale?) probably still does but only on the 6800 instruction set.

                OT I wonder what a Psion II would have been like with the 6809 instruction set available?

                1. Chemist

                  Re: Z-80

                  "The 6809 seemed to be one of the best of the 8bit breed"

                  I think the problem was it WAS rather slow and the complex, elegant addressing modes and position independent code which made it such a great cpu to program in assembler were dying the death as compilers generally only used simpler modes. It was caught at the wrong moment in time really with the 68000 being developed on one hand and the shift over to RISC on the other.

                  I've often thought that an updated 32 bit with high clock speeds would have been interesting but past is past and I'd much rather have a multi-core modern cpu running at 3GHz even if much of the power seems to disappear somewhere these days.

                  (I do have a 6809 still running in a home-made Forth system but I guess I've not switched it on in years)

                  1. Crisp

                    Re: Z-80

                    Who knew 8-bit processors would be such a hot button issue!?

                2. Jon Green
                  Pint

                  Re: Z-80

                  The 6809 was lovely to program (I still have several Dragon 32s somewhere, and I'd written my first OS kernel for a 6809), but it was still hampered by a rather limited register file compared with the Zilog Z80, which had already been on sale for a couple of years by the time the 6809 launched. This compromise was due to maintaining source compatibility with 1974's 6800. Compared to that processor, though, the 6809 was a breath of fresh air!

                  I don't think there ever was a "perfect" 8-bit processor - large register file, efficient execution, concise code, logical and consistent instruction set, clean memory map, and no significant bugs - from that era, but the Z80 (+variants) and the 6809 probably were the best of breed.

              2. Peter Simpson 1
                Childcatcher

                Re: Z-80

                Nowhere near as wonderful as the 68K!

                //no antique computing icon???

                1. Thorsten
                  Coat

                  Re: Z-80

                  "Nowhere near as wonderful as the 68K!

                  //no antique computing icon???"

                  Microprocessors, bah! Antique computing means TTLs, not VLSIs.

                  Mine's the one with Wang 2200 BASIC-2 manual in the pocket...

            2. Seele

              Re: Z-80

              Re:M6809 - and double indirect address modes with increment/decrement all in one instruction, and despite its lower clock speed, most instructions were less than 3 cycles. It was a beautiful processor, especially compared to the Byzantine horror that was the 8086.

          2. fajensen

            Re: Z-80

            The 6502 was cool, but the Z80 had programmable interrupts for peripheral devices. The best part was that one could both read (and learn) the entire instruction sets and physically carry the documentation!

            I gave up after the 486, where the "CPU-book" was something like 600 pages and The Errata's three volumes of equal bloat (AFAIK). Last time I looked, on AMD Athlon, the memory controller was a volume of about 100 pages ... for nostalgia I sometimes hack embedded thingies like MSP430. I can still understand that "scale" of device and it is more fun than soduko and crosswords.

          3. Blitterbug
            Happy

            Re: 6502 was the best 8-bit processor of all time...

            Now, now. Let's be fair. I loved its zero-page addressing mode (bit of a shock coming from Z-80) and yes it was *fast* but I also loved the chunky opcodes in Z-80, like LDIR. Horribly wasteful of clock cycles though it was, my first sprite routine relied rather heavily on it...

          4. Anonymous Coward
            Anonymous Coward

            Re: Z-80

            6502 was a sort of low cost 6800 which itself was based PDP-11 instruction set. So you can sort of thank DEC as much as MOS for that.

            It's good to know that I was actually born when it was released too :)

          5. A J Stiles
            Thumb Up

            Re: Z-80

            Considering that the 6502 went on indirectly to inspire the instruction set of the ARM processor, which is probably the most widely-used processor architecture on the planet (it's found in mobile phones, routers, printers, wireless cards and even the Raspberry Pi), you might just be right with that.

            Actually, the 6502 wasn't a bad chip in its own right, if you knew how to deal with zero page mode.

          6. Dunhill
            Happy

            Re: Z-80

            the 6502 and the 6522 where nice chips that made $$$$ for me, they were always burned

        5. Measurer

          Re: 8080, bloody Hell!

          Ahh, embedded stepper motor controller software development on a Z80 and PIO chip with NMI's, takes me right back....

        6. Charles 9

          Re: 8080, bloody Hell!

          For me, it was the 6502 on a Commodore 128. Simple as it was, the built-in machine language Monitor actually found some use for me in the latter days as with help from a book explaining all the opcodes, I actually learned how to make my own ML programs. Being able to interpret assembler code is still finding use for me when I stumble upon the occasional snippet. Generally, all I need is a lookup of the appropriate architecture and I can follow along.

          1. Anonymous Coward
            Anonymous Coward

            Re: 8080, bloody Hell!

            The C128 had a Z80 inside too for CP/M. Dual processor :)

            1. Charles 9

              Re: 8080, bloody Hell!

              Yeah, but standard mode (and the monitor) used the *8*502 (misread a bit, but it's a superset of the 6502). CP/M never really was that practical on a 128, even if you used fast mode in 80 columns. Plus by the time it came out the transition to DOS was already in progress. In any event, it wasn't really dual processor as only one or the other was on at a time.

          2. Erik N.

            Re: 8080, bloody Hell!

            Actually the C128 had an MOS 8502 and a Zielog Z80. The C64 started out using the MOS 6510 and later used the 8500. Everything except the Z80 were 6502 compatible, but each had its own additional capabilities.

        7. Palf

          Re: 8080, bloody Hell!

          8080? - pshaw. Around the same time GE MicroElectronics up in Bonny Scotland came out with the CP1600. Yeah, 16 bits in 1974. Stomped all over the 8086, but their marketing department must, in retrospect, have sucked mightily. I wonder who that was? Probably William McGonagle. He was writing poems instead of selling it.

          Ah saw a CPU up on a hill

          It's no there noo...

          It's still no there noo.

      2. Stretch

        Re: 386?

        snap

    2. Allan George Dyer
      Coat

      Re: 386?

      You didn't start on the 4040, like me? To be fair, mine was a plastic DIP package, but the IO chip was ceramic.

      I never tried running a *nix on it, and, IIRC, Linux never supported the 286 or earlier.

      I only switched off my last 386SX in 2010, though that was running DOS.

      1. Anonymous Coward
        Anonymous Coward

        Re: 386?

        @Allan George Dyer Pretty sure it ran Minix, though.

      2. charlie-charlie-tango-alpha

        Re: 386?

        "Linux never supported the 286 or earlier."

        Well, the linux kernel didn't appear until 1991 and Linus built it for a 386/486 target (he was playing with Minix 386 and was frustrated with its limitations).

        However, somewhere in my loft I still have a copy of v2 of Xenix (an MS licensed version of Unix) dating from the mid 80s which ran on a 286. The earlier version ran on 8086 I believe.

      3. lotus49

        Re: 386?

        Minix and Coherent did both support the 286. I remember them both fondly.

    3. Anonymous Coward
      Anonymous Coward

      Re: 386?

      You think you had it bad? Ha! When I was a lad I had to program an abacus in binary to factorise on an elliptic curve, while barefoot, sweeping the igloo 's chimney and fending off hungry polar bears.

      You don't know how lucky you've got it.

      1. Adam 1

        Re: 386?

        Luxury. We used to dream of hungry polar bears...

        1. Bob Merkin

          Re: 386?

          An abacus? We had to do our calculations with rings of giant stones, and it took the machine 31,557,600 seconds for one clock cycle!

          1. DaiKiwi

            Re: 386?

            "An abacus? We had to do our calculations with rings of giant stones, and it took the machine 31,557,600 seconds for one clock cycle!"

            I thought it was 31,556,925 and a half seconds.

            Or was that the mark II?

            1. ThomH

              Re: 386? (@DaiKiwi)

              There's no mark II; I've heard rumours some sort of internal disagreement about fjords.

      2. Field Marshal Von Krakenfart
        Coat

        Re: 386?

        You think you had it bad? Ha! When I was a lad I had to program an abacus in binary to factorise on an elliptic curve, while barefoot, sweeping the igloo 's chimney and fending off hungry polar bears.

        You don't know how lucky you've got it.

        Binary, you had Binary, Ohhhh we used to DREAM of havin' binary, there were 120 of us working in the outflow pipe of local sewage works, we couldn't even afoard ones and zeros, we 'ad to write all our programs using only a zero wher we worked twenty-four hours a day at the mill foxconn for fourpence every six years.

        But you try and tell the young people today that... and they won't believe ya'.

        Coat, with matching flat cap.

        1. Tezfair
          Facepalm

          Re: 386?

          Why am I reading this in my head with a yorkshire accent?????

          e i don't 'no

    4. Timmay
      Joke

      Re: 386?

      But didn't Apple invent both the 386 and the 8080? And all other chips in between, and ever since?

      I'm sure I read a patent application just last week about it.

      1. Simon B
        Devil

        Re: 386?

        No, apple invented the rectangle ... ;)

      2. Charles 9

        Re: 386?

        Nope. Intel gets all the credit and they have the applications to prove it, dating all the way back to their 4004 (the world's first self-contained CPU chip).

        1. Tom 13

          @Charles 9

          Shirley you must be mistaken! Everyone knows there were no computer until St. Jobs descended from the iCloud with the iPad.

        2. ThomH

          Re: 386? (@Charles 9)

          Surely CTC take a large chunk of the credit via the Datapoint 2200? They largely specified the 8008 instruction set and merely contracted it out to Intel. When Intel couldn't deliver on time they went with a TTL implementation, meaning that the very first commercial sale of a predecessor of the x86 architecture started in 1970 without Intel parts. As part of the contract termination negotiations, Intel got to keep the instruction set though at that point they'd never shipped a microprocessor — the 4004 wasn't available until late '71 and the 8008 itself wasn't completed until '72.

    5. Uncle Siggy

      Re: 386?

      Feh! That's nothing. Back in my day I had to do all my work on an abacus. Even the slightest nudge would shift my entire equation!

      1. Anonymous Coward
        Anonymous Coward

        Re: 386?

        Uncle Siggy:

        > Feh! That's nothing. Back in my day I had to do all my work on an

        > abacus. Even the slightest nudge would shift my entire equation!

        Shift happens!

        (courtesy of Mr Adams)

  2. Dazed and Confused

    Re: At that speed Windows 3.1 did not disgrace itself.

    Well not with typists of the one fingered variety perhaps. But the screen couldn't keep up with even my own random 10 fingered crawl across the keys let alone any moderately competent typist. I remember having to talk to quite a number of secretaries who were being driven up the wall by how much time it would take to catch up with them.

    1. Anonymous Coward
      Anonymous Coward

      Re: At that speed Windows 3.1 did not disgrace itself.

      You configured it wrongly then.

      1. Alan Bourke

        Re: At that speed Windows 3.1 did not disgrace itself.

        LOL

    2. yossarianuk

      Re: At that speed Windows 3.1 did not disgrace itself.

      It did compared to the speed on the Amiga (which only ran at 7Mhz)

      1. Anonymous Coward
        Anonymous Coward

        Re: At that speed Windows 3.1 did not disgrace itself.

        I touch typed on my Amiga 500, it was great to use! I still remember when I exchanged my A500 for an A1200, which I still have sitting in my garage, might boot it up sometime soon and slip one of the many 2.5" hdd's I have laying around it!

        1. Anonymous Coward
          Anonymous Coward

          Re: At that speed Windows 3.1 did not disgrace itself.

          I'd imagine the FFS of AmigaOS (unless you have a recent ROM/OS) has some limitations on disc size.

          You can get an IDE to CF adaptor that will be cheaper than finding an old 2.5" IDE small enough. Not having virtual memory means it won't wear out too fast.

    3. I ain't Spartacus Gold badge

      Re: At that speed Windows 3.1 did not disgrace itself.

      I don't remember the typing problem. And I replaced an actual typewriter with my first PC, a 25Mhz 386 SX with 2 whole MB of RAM. It's amazing how little you can get on a 40MB hard disk, when it's already half full of DOS and Windows 3.1.

      My 8 notes of polyphony Soundblaster was amazing too. Wolfenstein 3D was my game of choice. The sad thing is I saw it somewhere a few months ago and played a bit, and I can still remember the way round all the levels (including where about half of the secret passages are).

      1. Conor Turton

        Re: At that speed Windows 3.1 did not disgrace itself.

        "It's amazing how little you can get on a 40MB hard disk, when it's already half full of DOS and Windows 3.1."

        HAHA. I remember those days. When Doom came out I had a choice of having Windows or Doom on so I got very good at installing Windows. I actually still have that 40MB Hard drive. I tried killing it once by driving over it in a 38 tonne lorry several times but it lived and worked perfectly. They dont build them like that any more.

        1. Field Marshal Von Krakenfart

          Re: At that speed Windows 3.1 did not disgrace itself.

          "It's amazing how little you can get on a 40MB hard disk, when it's already half full of DOS and Windows 3.1."

          40MB, We used to dream of havin' 40 MB...... etc. etc.

          The first PC I had had a 20MB HD with a RLL controller to trick it into giving you 32MB, when I got it a friend asked me "what do you need that much space for, are you going to use it as a file server?".

          That was in the days when the turbo C compiler and debugger lived quite comfortably on a 5-1/4" floppy and terabyte was something you heard of in SiFi films . I got rid of that PC when the install for MickeySoft 'C' (V6.0 ???) hit 40MB.

          1. sisk

            Re: At that speed Windows 3.1 did not disgrace itself.

            The first PC I had had a 20MB HD with a RLL controller to trick it into giving you 32MB

            Lucky bastard. My first computer had a pair of 5 1/4 in floppy drives and a floppy labeled "DOS Boot Disk". No internal storage whatsoever. I had at least 10 copies of that boot disk to, just in case something happened to it.

            1. I ain't Spartacus Gold badge

              Re: At that speed Windows 3.1 did not disgrace itself.

              You were lucky! My first computer were even worse. It only had 2 x 3" floppies, and mine was labelled CP/M boot disc. That were my Amstrad PCW 8256. The Vic 20 and Amstrad CPC464 before than were my brothers'.

              1. Thorsten

                Re: At that speed Windows 3.1 did not disgrace itself.

                My first own computer had just a single 5 1/4" floppy drive. Saving anything to disk from, say, Word, took about a dozen (maybe more) disk swaps. The 32 MB RLL was suddenly the highest priority on my wishlist...

    4. sisk

      Re: At that speed Windows 3.1 did not disgrace itself.

      I remember having that problem, and I didn't learn how to type properly until later. My typing speed at that point couldn't have been more that 20 or 25 wpm. That was only in Microsoft Works, however. In a proper word processor (my definition of proper at the time being Q&A.....did anyone else use that?) the computer kept up just fine with my mother, who was an excellent typist at the time, even if she did constantly complain about the differences between the computer keyboard and the one on her typewriter.

  3. Number6

    486?

    I assume the 486 still scrapes by? There are still products using 486-compatible processors on the market, I've used some recently. It would be somewhat annoying if Linux no longer supported those, but I haven't used a 386 CPU for many years (despite having a couple of old DOS 386 machines in the loft).

    1. Anonymous Coward
      Anonymous Coward

      Re: 486?

      Isn't 486 a 386+maths coprocessor?

      Anyone who needs a 386 can emulate it in software.

      1. Tom 7

        Re: 486?

        Can emulate it in software? Explain please how emulating it in software will make my hardware run?

        1. Anonymous Coward
          Anonymous Coward

          Re: 486?

          You run your existing hardware with an existing kernel.

          If you just have software which requires a 386, that you could do that in an emulator.

          Must I think of everything? ;)

          Actually, its getting hard enough to find a distro with a downloadable non-pae installation kernel on a pentium m.

      2. richardcox13
        Boffin

        Re: 486?

        > Isn't 486 a 386+maths coprocessor?

        Not necessarily: remember the 486SX?

        Plus it made some significant changes to the instruction set that made multi-threading significantly easier. For example the atomic interlocked increment and decrement operations return the new value rather than just a zero or not-zero indication. Hence (from the article:

        > which complexity has plagued us with extra work whenever we wanted to change SMP primitives

        (And also why the InterlockedIncrement and InterlockedDecrement Win32APIs where different on WinNT vs. Win9x.)

        1. Marco van de Voort

          Re: 486?

          The 486 also introduced some other instructions iirc, bound and bswap. (though for bulk swapping SSE solutions are faster)

          Note that this might not only kill 386s, but also 3rd party x86 chips that are not entirely 486 compatible in the sense that matters for the kernel. Cyrix and UMC among others had 486 alikes.

    2. tapanit
      Linux

      Re: 486?

      486 is still supported, judging by comments in the changelog. There are some sort-of-486-like processors that are no longer supported, however:

      "Note that the 386 is no longer supported, this includes AMD/Cyrix/Intel 386DX/DXL/SL/SLC/SX, Cyrix/TI 486DLC/DLC2, UMC 486SX-S and the NexGen Nx586."

      1. DrXym

        Re: 486?

        "Note that the 386 is no longer supported, this includes AMD/Cyrix/Intel 386DX/DXL/SL/SLC/SX, Cyrix/TI 486DLC/DLC2, UMC 486SX-S and the NexGen Nx586."

        Some chip manufacturers played fast and loose with the numbering so they might have said 486 and supported the instruction but internally they might have had a largely 386 architecture such as bus. Anyway I expect they're dumped for their rarity and disuse as much as anything else - who knows or cares if they work because nobody has the hardware to actually test them.

      2. PTR

        Re: 486?

        Cyrix!

        Wow, I'd forgot my first PC had one of their chips!

        1. Stuart Elliott
          Boffin

          Re: 486?

          I remember building a Cyrix M1 PC for someone, was their so-called 586 chip, was like excrement off a digging implement at the time.

    3. Anonymous Coward
      Anonymous Coward

      Re: 486?

      The more things change, the more they stay the same.... my current work laptop sometimes cannot keep up with my typing speed (Windows 7 + Word)

      Of course, back in those days we did not had to run the domain policy update, the antivirus scanner and who knows what else in the background.

      1. Anonymous Coward
        Anonymous Coward

        Re: 486?

        "The more things change, the more they stay the same.... my current work laptop sometimes cannot keep up with my typing speed (Windows 7 + Word)"

        Just this morning, on my 2 year old Core i5 w/6GB RAM, Win 7 and was typing an email in Outlook, it was struggling to keep up and I ended up with a word inserted into the middle of anwordother.

        Yet my 486 with Clarisworks (albeit I never ran email on it until it became a Pentium 233) never had this issue, was always snappy. At least until I put Win95 on it and had to DriveSpace the 170mb HDD to get any space.

      2. plrndl
        Linux

        Re: 486?

        MS has always acted as if the purpose of an operating system is to suck all the life out of the CPU, so you can't run any non-MS software, unless you invest in extreme hardware, like gaming machines. MS Word on my work 2GHz Core 2 Duo XP machine is slower than Wordstar on my old 12 MHz 286 DOS 3.3 from back in the last millenium.

    4. Anonymous Coward
      Anonymous Coward

      Re: 486?

      The 486 has a few MMU features the 386 lacks, notably page write-protect in kernel mode. That's probably the source of the complexity Linus was referring to. Could be the new instructions and builtin FPU too.

      1. Tom 13

        Re: 486?

        Ah, but 386 motherboards had a cool feature modern boards don't: an clock circuit that kept accurate time. I think they only disappeared with the advent of the Pentium. My first employer use to look for them because he could pick them up and sell the clock circuits to a client who repurposed them.

  4. Andrew Garrard

    How not to build a 32-bit CPU

    Ah, the 386. I remember when the first Compaq 386 machines came out (I believe at 20MHz, not 33MHz, though Wikipedia tells me that slower 386s were available), and my fanboyism disliking the fact that it took the performance crown for desktops back from the 8MHz ARM2 Archimedes machines (probably as measured by dhrystone, although BASIC may have been involved). I'm sure my copy of Structured Computer Organization contains some comment about Intel "finally making a decent CPU", but unfortunately I'm a few thousand miles away and can't check - anyone got the red edition?

    The 486 was a bit nicer as designs go, excluding the slight problem of getting everyone to optimize code in a way that was pessimal for Pentiums. I still wish IBM hadn't decided to use the chip from their printers and had gone with the 68000 series from the start (and if we were going to end up with thirty years of compatible machines foisted on us by Windows, we'd ended up with anything less crufty than x86), but at least it meant that near and far pointers weren't always obligatory...

    1. Ian Yates
      Windows

      Re: How not to build a 32-bit CPU

      "ended up with anything less crufty than x86"

      As always, it's the marketting and "ease" of use that wins, not the better implementation or sophistication.

      I used to be surprised at how well x86 coped with the exploding computing market, but (as you said) Windows lock-in pretty much guaranteed that no one really wanted to put money in to alternatives.

    2. Ian 5
      Go

      Re: How not to build a 32-bit CPU

      I would though like my PC to still have a turbo button, with a blocky 7 segment display to show my clock speed...

      20MHz? Nooooooo <click>TURBO POWER</click) 33MHz!

      Ah... memories. Overclocking was never so easy again...

      1. Ian Yates
        Thumb Up

        Re: How not to build a 32-bit CPU

        I remember doing that on my first home PC (massive steel-framed beige box that it was).

        When it was finally retired by my dad's purchase of a 486, I took the thing apart and was slightly disappointed to find that the values shown on the 7-segment display were purely down to the state of the toggle button (being now much wiser, it clearly makes sense).

        On the plus side, it meant that I was able to fit it to the 486 case and continue with the placebo "active turbo!" fun ;)

      2. Anonymous Coward
        Anonymous Coward

        Re: How not to build a 32-bit CPU

        Everything 80s and early 90s was TURBO :)

        Even Pascal!

        It was the 'Hybrid' buzzword of it's day.

        1. Captain Scarlet Silver badge
          Coffee/keyboard

          Re: How not to build a 32-bit CPU

          Ah so thats why Jeremy is facinated by Volvo Turbo's

  5. calagan

    Sad day

    I think that this great ideal which would aim at being able to run Linux on ANY device prevails over these complexities.

    I understand that Linux will not suddenly stop running on 386 boxes, but this is a step in the wrong direction, especially when you consider that with Linux , so much can be done with so little CPU.

    1. Anonymous Coward
      Anonymous Coward

      Re: Sad day

      >but this is a step in the wrong direction

      I'm not sure how dumping legacy stuff that causes issues supporting the majority of users is a "step in the wrong direction". I would go further than Linus.. I would remove all of the archs and boards that haven't been touched for years and only allow them back in when someone steps up and offers to support them and can prove that they still work or are fixable.

      >especially when you consider that with Linux , so much can be done with so little CPU.

      If you turn off everything that makes using a recent kernel worth using..

      1. Anonymous Coward
        Anonymous Coward

        Re: Not a sad day

        Anyone who has a 386 processor and wants to install Linux on it now can easily download one of the zillions of distros that support kernel up to 3.3 (or whatever) and it will run just fine. And then they can do the same for all of those other processors that haven't been sold in years.

        Just provide a document somewhere that says processor X is supported in kernel version [1, 3.3] or whatever with link to some distro locations that include that.

        Then the people maintaining the kernel can get on with supporting new stuff (only)

        1. Anonymous Coward
          Anonymous Coward

          Re: Not a sad day

          >Anyone who has a 386 processor and wants to install Linux on it now can easily download one

          >of the zillions of distros that support kernel

          I don't think that is true.. as someone else said glibc doesn't work on 386 anymore.. IIRC even Debian's i386 arch has only worked on 486+ for at least the last stable release, maybe the one before that.

    2. 123465789
      WTF?

      Re: Sad day

      Well, you can still run Linux on an 386 - just have to get a version before this merge. Yes, you can't use the latest and greatest version. But really, which of the newer features do you desperately need on an embedded 386-based device?

      1. Destroy All Monsters Silver badge
        Pint

        Re: Sad day

        The more so as embedded devices should really move to more modern CPUs than 386

        1. Bob H

          Re: Sad day

          Plus if someone is really concerned about it they can always fork support and merge the kernel modules that they need. If it is holding the platform back to support a chipset that hasn't been in production for years and isn't a commercial or societal need then it can go.

    3. mr_jrt
      Linux

      Re: Sad day

      Luckily, it's open source.

      If you have a pressing need to support the 386...then fork, and maintain your own kernel with support retained. You'll still be able to merge in the patches as the Linux guys do...it's just it will become more work over time...

    4. Ben Liddicott
      Pint

      Re: Sad day? The previous kernel versions still exist

      Erm, they aren't going back in time and deleting previous kernel versions. Just saying future kernel versions won't work.

      So just use the ones which do work. No problem.

      As for support, they will be supported in the same way as any other linux kernel - you support it in-house yourself, or buy support separately from a company or person with expertise.

    5. Stuart Castle Silver badge

      Re: Sad day

      There comes a point where you have to cut off support though, otherwise Linux would become bloated and inefficient. If relatively few people are using 386 class machines and supporting them is causing problems, why not drop support? After all, they'll probably not gain any advantage of using a newer kernel anyway as the machine won't be powerful enough to support a lot of new features.

      Earlier kernels probably aren't going to stop working because a newer kernel has been released, and, it being open source, there is nothing stopping someone restoring 386 support to newer builds.

  6. Anonymous Coward
    Anonymous Coward

    The joys of open software

    Being the head of one of the most successful and innovative open source movements in the world clearly has its freedoms. No paying customers means zero commercial obligation to keep supporting them, and one can just walk away without a care in the world. Bliss!

    So there's a little lesson. You really can't count on free software being sustainable in the long run. If your product life cycle isn't going to evolve at the same-ish rate as the free software you're currently depending on then it's not necessarily a good idea to use it in the first place. Many people use Linux as if it will be there forever, and will always be the best thing in town. With enough people using it that will probably become a self fulfilling prophecy (certainly no bad thing!), but it's not written into any contract anywhere that that really will be the case.

    In practise I suspect that it won't matter that much to embedded 386 users (I'm not one - PowerPC me). Embedded software is updated far less often than mainstream IT, and if it works now it'll keep working. I can understand Linus's point of view, and it will be quite a long time before it becomes a major headache for the users of these ancient chips.

    On a side note, a Linux driver dev I know moans that Linux is a bit Mozilla-ish. Every major release changes everything all over again and drivers have to be re-written, etc. etc. It's a ball ache to keep up. That may indeed be the price of innovation and progress, but on the whole many people would prefer a more considered approach. Device driver devs matter a lot, and if they get too fed up and stop bothering? Well, that sort of thing can do real damage to an OS.

    MS (and presumably Apple) long ago learnt the true value of drivers, API stability, etc. Like it or loathe it, the fact that Windows XP has been maintained all this time, still works, and so forth is fairly impressive and has meant that many paying customers have stuck with MS. Windows 7 looks like it will go the same way, and that's good news for many people out there.

    1. P. Lee

      Re: The joys of open software

      Must... not... feed... the... troll...

      Good luck with your W7 on a 386 or OSX support on powerpc.

      Bother, failed.

      1. Anonymous Coward
        Anonymous Coward

        Re: The joys of open software

        Lol better yet try running win98 on 386.

    2. Anonymous Coward
      Anonymous Coward

      Re: The joys of open software

      By contrast vendors of paid-for software withdraw support for their OSes afor their own commercial advantage, and with the objective of forcing their still-locked-in customers onto another platform controlled by that vendor... ka-ching! Exhibit A - HP, which has notched up - RTE, MPE, Domain/OS, WebOS at least and is probably trying to snuff out OpenVMS as we speak.

      We're talking about a part that ceased production 5 years ago (at which time it would have been made for anyone to be using it in new designs) - which embedded device vendor is still releasing major firmware updates for 5 year old products (yes I appreciate that device manufacturers' stocks of the part may take a year or two to be used up)?

    3. localzuk Silver badge
      Facepalm

      Re: The joys of open software

      Major flaw there. Try running Windows XP, which is still in some form of official support, on a domain with mandatory profiles and Internet Explorer 8 - the latest version to work on it and report back to me how well that works out for you. I know for a fact that it doesn't work as well as anyone would like, as IE creates files that have names that are too long, which messes up the locally cached profile so when a user logs in again it can't do anything with them and therefore doesn't let them log in. They also won't be fixing it, even though MS have admitted it is a problem, with the advice being... upgrade to a newer version of Windows.

      Linux has a heck of a longer support cycle than Windows, period. But if you need to support an ancient chip 27 years after it was released, I'd be advising that you make your own custom OS, and not expect anyone else to.

      That's stable and supported software too...

      1. Anonymous Coward
        Anonymous Coward

        @localzuk

        If you are using mandatory profiles then use a gpo to stop Windows making a local cached copy of the profile. If you are using mandatory profiles then this is the recommended way of using them. If your network is so slow that local cached profiles are necessary then you need to upgrade your infrastructure, if its because the server isn't always available then maybe local profiles would be a better option.

      2. david 12 Silver badge

        Re: The joys of open software

        Personally, I am finding that Win98 has better support than 2 year old EEEBuntu.

        I don't know that I agree with the idea that Linux has a longer support cycle than Windows: Red Hat has a 10 year support cycle for server products (XP is already older than that), but your workstation Linux from 10 years ago is well out of support now.

    4. Ronny Cook
      Facepalm

      Re: The joys of open software

      So, you're praising MS because they support a ten year old operating system, while criticising Linux for dropping support of a 27 year old CPU. Good luck in getting MS to support MS-DOS 4.0...

      For that matter, the embedded systems manufacturers are still free to fork their own copy of the Linux kernel that supports the 80386. They can backport patches from the mail kernel tree. They have full source access. Heck, if there's enough of a demand somebody is free to create a startup which does all these things for the embedded market - not that I can see it happening for a CPU that has not been manufactured for five years.

      With a closed source OS, none of these things are options. If your OS vendor drops support for your platform, you either find another OS or keep running the last version that works, tweaking your hardware to keep it functional.

      With regard to drivers, I can't speak from personal experience, but your friend did say "major releases" - 2.6 was released in 2003, about the same time as Windows XP, and many XP drivers work poorly if at all on Windows 7 (I hate to think how they are on Windows 8).

      1. Piro Silver badge

        Re: The joys of open software

        Not really 27 years old, because it was (and probably is) used commonly in embedded systems.

        http://www.reghardware.com/2006/05/18/intel_cans_386_486_960_cpus/

        March 2007 is when Intel stopped making 386s.

    5. Anonymous Coward
      Anonymous Coward

      Re: The joys of open software

      You're forgetting a major point that sort of completely invalidates your argument. Whichever kernel you choose up to the latest one that runs on the 386 can be recompiled, altered and upgraded at your will. Nothing stops you forking the current kernel and keeping your own 386-good branch. It will not be as cheap as free, but possible to do.

      Try that with your Windows/OS X/OS2 binary release of 15 years ago and come back with your results.

    6. Michael H.F. Wilkinson Silver badge

      Re: The joys of open software

      The joy of open software is that you can fork the code if you do not like where it is heading. Not trivial, but not impossible. You are allowed to. Try that with commercial code.

      1. Anonymous Coward
        Anonymous Coward

        Re: The joys of open software

        I have the entire (i hope) source code for CP-6 version E00, on microfiche. I don't remember what the 'with source code' option on the support contract cost, but it was significant. I'm sure if we had run our factory on Microsoft WIndows, I'd have a copy of that instead.

        Maybe one day I'll have the time, inclination and a microfiche reader and be abe to peruse the AARDVARK code...

        anonymous, just in case.

    7. Charlie Clark Silver badge
      FAIL

      Re: The joys of open software

      @deluded wanker

      XP is only still around because companies refused to buy Windows Vista. Neither of them will run on a 386. I can't remember when Microsoft dropped support for the 386 but it could have been as far back as Windows 95. More recently you can pity those poor fools who have bought Windows Mobile or Windows Phone devices only to see support for them dropped after about 12 months, Microsoft's approach to drivers was laughable for years - basically anything was possible and this was a major source of problems for many because buggy hardware would rely on fixes in untested drivers.

      There are lots of things to criticise about Linux and its development but this isn't one of them. Anyway if you want to keep your old 386s running up to date unix just switch to NetBSD!

      1. Soruk

        Re: The joys of open software

        > I can't remember when Microsoft dropped support for the 386 but it could have been as far back as Windows 95.

        Windows NT4 certainly ran on the i386. My university had a few compter rooms filled with Compaq 386SX-25s with 4MB RAM.

        OK, "run" isn't quite the right description. More accurately, the code executed.

        1. david wilson

          Re: The joys of open software

          Surely explicit old processor support in MS OSes is relatively moot anyway - even without things like memory size issues, few people would be likely to have the *patience* required to run such things on hardware much over a decade old.

          1. Gagol
            Happy

            Re: The joys of open software

            Day 1: push the power button

            Day 2, morning: enter credentials

            Day 2, afternoon: open mail client

            Day 3: read emails

    8. elstcb

      Re: The joys of open software

      "You really can't count on free software being sustainable in the long run."

      I was somewhat surprised and dismayed the other day when I went looking for "gadgets" for my windows 7 laptop. I don't normally go anywhere near them but I wanted something specific, I forget what now though.

      Anyway, upon clicking "Get more gadgets online" I was taken to a page that said:

      "Because we want to focus on the exciting possibilities of the newest version of Windows, the Windows website no longer hosts the gadget gallery. " (not even the ones that used to be there)

      If wikipedia is correct then Windows 7 is only about 3 years old. Microsoft is already dumping additional stuff from Windows 7 to "encourage" people to upgrade to Windows 8.

      1. Piro Silver badge

        Re: The joys of open software

        The removal of gadgets was over a security flaw - but they saw so few people bothered to use them, I guess they couldn't be bothered to fix it, and put their effort (which is to say almost no effort) into Windows 8 instead.

    9. Destroy All Monsters Silver badge
      Thumb Down

      Re: The joys of open software

      > So there's a little lesson.

      RICHTO, stop posting as an anonymous coward.

    10. Nuke

      @ AC [Microsoft Salesman?] = Re: The joys of open software

      Wrote :- "MS (and presumably Apple) long ago learnt the true value of drivers, API stability, etc. Like it or loathe it, the fact that Windows XP has been maintained all this time, still works, and so forth is fairly impressive"

      Please keep your analogies relevant. Torvalds is talking about the next version of Linux. That is equivalent to Windows 9, not XP. Will W9 run on 386's? I would not know and don't really care (who the hell would want to do that anyway?) and would not criticise MS if it didn't. Perhaps you can tell us, as you seem very close to MS?

      Also, XP is still used by a large percentage of people, including me when I use Windows. OTOH 386 processors are not much used now, and those that are are used for special applications for which the developers are quite capable of chosing an appropriate version of Linux.

      1. Blane Bramble
        Windows

        Re: @ AC [Microsoft Salesman?] = The joys of open software

        Windows 8 System Requirments:

        Processor: 1 gigahertz (GHz) or faster with support for PAE, NX, and SSE2

        So no 386's then.

        PAE: Pentium Pro or newer

        NX: Pentium 4/AMD64 or better

        SSE2: P4/AMD64 or better

        (from Wikipedia)

        So, Windows 8 doesn't support 386, 486, Pentium, Pentium 2 or Pentium 3 architectures it seems, or any of the AMD equivalents.

    11. Dazed and Confused

      Re: MS (and presumably Apple) long ago learnt the true value of drivers, API stability, etc.

      Argh, that might explain why no one ever had any problems moving from XP to Vista then.

    12. Ian Yates
      Facepalm

      Re: The joys of open software

      I know you're a troll but... "Every major release changes everything all over again"

      Surely, that's exactly the definition of a major release? If it doesnt include breaking changes, the convention is to call it a minor release or revision.

      Remember that (commerical) Windows versioning is more related to the GUI than any underlying architecture.

    13. Tom 13

      Re: The joys of open software

      I'd wager that 3 years from now Linux will still have better support for the 386 than MS does for DOS 6.12.

    14. A J Stiles

      Re: The joys of open software

      Being the head of one of the most successful and innovative open source movements in the world clearly has its freedoms. No paying customers means zero commercial obligation to keep supporting them, and one can just walk away without a care in the world. Bliss!
      Yes, because all the code that's already been written that did support the 80386 has suddenly vanished for all time. Oh, no, wait. It hasn't. It's all still there. Scrub that, then.
      a Linux driver dev I know moans that Linux is a bit Mozilla-ish. Every major release changes everything all over again and drivers have to be re-written, etc. etc. It's a ball ache to keep up.
      "Changing everything all over again" is pretty much the definition of a major release.
      MS (and presumably Apple) long ago learnt the true value of drivers, API stability, etc. Like it or loathe it, the fact that Windows XP has been maintained all this time, still works, and so forth is fairly impressive and has meant that many paying customers have stuck with MS.
      Windows XP has been about for what, 11 years or so, and Microsoft are going to be pulling support for it Real Soon Now. Linux has been around since 1991, and every version ever released is still available. For £0, and with commented Source Code and modification rights.

      Binary compatibility across releases was deliberately never a design goal of Linux. This was a deliberate decision: binary compatibility can potentially mean having to support things that, with the benefit of hindsight, turn out to have been stupid decisions. Having to recompile applications (or download new, pre-compiled versions from a trusted source) is ultimately less bother for users than security holes you could get a bus through sideways.

  7. Anonymous Coward
    Boffin

    The writing was on the wall

    glibc hasn't worked on i386 for some time now, a few years at least. So you were more or less stuck with µClibc on 386s.

    Most embedded things that would run on a 386, could equally run on an ARM or MIPS chip, with far lower power consumption and in a much more compact form factor than any 386.

    I will point out, that 386s running at 33MHz were common, and occasionally one did see a 40MHz 386 out in the wild. I've never heard of a 33mHz 386 chip — that'd be slow indeed!

    Last time I ran Linux on a 386 was in 2008. I grabbed this 386DX 33MHz with 20MB RAM to use as a stand-in for an embedded device as it was the closest thing I had to hand that would match the constraints of a typical embedded system. It ran a single floppy distribution of Linux based on µClibc and Busybox, had a (crude) web interface and was intended to drive a LED sign, taking feeds via RSS (using a self-written RSS parser based on libexpat). With a bogoMIPS reading of about 6 with the Turbo button turned on, it was the slowest Linux box at the uni.

    1. TeeCee Gold badge

      Re: The writing was on the wall

      "..occasionally one did see a 40MHz 386 out in the wild."

      It was AMD who produced a 386 clone clocking at 40MHz.

      1. TeeCee Gold badge
        Facepalm

        Re: The writing was on the wall

        Downvote? Ah!

        You're right of course, AMD's almighty clock-up was to 50MHz. In my defence all I have to offer is that it was a while ago.

    2. Marco van de Voort

      Re: The writing was on the wall

      9 times out of ten that you run such old systems, it is also attached to other hardware (read: cards) and software that are hard to replace.

      If the 386 was a database server it had been virtualized or migrated eons ago.

      1. Anonymous Coward
        Anonymous Coward

        Re: The writing was on the wall

        "9 times out of ten that you run such old systems, it is also attached to other hardware (read: cards) and software that are hard to replace."

        And such a machine would be running a modern Linux distribution? I think not.

    3. Sam Liddicott

      Re: The writing was on the wall

      My first 386 was a 386SX-20 someone gave me.

      Then, all of a sudden, a load of 386DX-40 boards WITH processor hit the market at £40. Everyone I knew bought one of those with 4MB of RAM which I think was another £120

      Then we all played doom.

    4. Eddie Edwards
      Pint

      Re: The writing was on the wall

      "I've never heard of a 33mHz 386 chip — that'd be slow indeed!"

      The great thing is it makes your HDD look like L1 cache.

  8. TheOldFellow

    386, what about the 4040?

    Damned new fangled chippery. who needs 64 bits, 32 bit or even 8 bits. 4 bits was enough for Apollo and the LEM!

    1. Anonymous Coward
      Anonymous Coward

      Re: 386, what about the 4040?

      "4 bits was enough for Apollo and the LEM!"

      No it wasn't; the Apollo Guidance Computer had 16-bit words.

      http://www.ed-thelen.org/comp-hist/vs-mit-apollo-guidance.html

      This book, by its designer, is well worth reading:

      http://www.amazon.com/Journey-Moon-Library-Flight-Eldon/dp/156347185X

      1. Dazed and Confused

        Re: 386, what about the 4040?

        You mean you don't have the Haynes manual for the Saturn 5?

      2. John Smith 19 Gold badge

        Re: 386, what about the 4040?

        "No it wasn't; the Apollo Guidance Computer had 16-bit words."

        Strictly it was 14 bits, with 2 bits for error detection and correction (1 bit correction, 2 bit detection).

        The internal architecture was bit serial clocked at 1Mhz (can't recall why as 16*32k is512k and its computing power was around 32Kips).

        So 14 bit or 1 bit, depending on your PoV.

        NASA quite liked the bit serial approach. Lower power, few chips -> less weight but slower (when you're trying to shoehorn a probe to Jupiter onto a relatively small rocket that makes a difference (and you get to define your own instruction set, which is pretty handy too). There's a book in the NASA history series on Spaceflight Computers. It makes interesting reading. A 386 was the upgrade to the Hubble control system.

  9. Ole Juul

    By the time the 486 and Pentium came along, 386s dropped out of sight.

    That is not entirely correct. People who were there, and the modern retro gaming crowd, will know that the AMD 386DX-40 was a hot seller precisely because it was faster than the 486SX. The reason being it had more cache. The Pentium of course beat them all. Here is a good writeup on the AMD386.

    1. localzuk Silver badge

      Re: By the time the 486 and Pentium came along, 386s dropped out of sight.

      I was still running a network consisting of 386 and 486 machines in 1997/1998. I believe it didn't end up being replaced until 2001!

      They worked well, running Windows NT4 I found.

    2. Fading
      Thumb Up

      Re: By the time the 486 and Pentium came along, 386s dropped out of sight.

      Thanks for that - good bit of nostalgia on the linked site.

  10. Anonymous Coward
    Anonymous Coward

    It was this decision, plus lack of Linux support for the 6502 that made me switch to Windows 8

    I'm much happier now.

    1. Will Godfrey Silver badge
      Happy

      Re: It was this decision, plus lack of Linux support for the 6502 that made me switch to Windows 8

      Buggritt. I had a similar comment lined up.

      Now I'll have to get all serious and point out that a lot of SOCs have more capability than a 386 based machine.

    2. P. Lee
      Angel

      Re: It was this decision, plus lack of Linux support for the 6502 that made me switch to Windows 8

      HEATHEN!

      WHY DID YOU NOT UPGRADE IN THE ONE TRUE WAY TO A LISA?

  11. Khaptain Silver badge

    C'mon own up

    Who the fuck still wants to own and run Linux on a 386. I can't even imagine any reasonable nostalgique for doing so.

    I learned assembler on a 386 but what would I accomplish by firing it up now, christ were talking 20ish years ago.

    Linus is perfectly correct in abandoning old archaich code that does nothing positive for anyone, anymore.

    How many of you keep your spare parts from your 20 year old Ford Escorts ? They are not really very uswefull today are they. ( Except for hoping to flog them to some antique Ford Escort nut)....

    FFS, lets move on.

    1. MJI Silver badge

      Re: C'mon own up

      Nothing wrong with a pre 1981 Escort especially with a BDA or Pinto in it.

      After that crap.

      Last were abbysmal according to Top Gear, and is the worst car I have driven.

    2. James Hughes 1

      Re: C'mon own up

      I've got a garage full of Mk1/2 escorts bits and pieces (well, engines, gearboxes, radiators, steering racks, axles etc)- very popular in kit cars and Locosts, and in the Historic Rallying scene.

      A decent 1600 Xflow is worth decent money nowadays!

      You should have said Austin Maxi.

    3. Nigel 11
      Alert

      Re: C'mon own up

      I think the right question is "Who the fuck still wants to own and run the latest Linux kernel on a 386". For which I cannot offer any good answer.

      The obvious answer to the question as posed, is anyone who has a large investment in a piece of hardware that's still useful and which would be very expensive to replace, which is controlled by an embedded 386 PC that runs linux. I don't have to look after any lab equipment that runs Linux. I do know of two pieces of lab equipment that still run Windows 3.1, and a couple more locked to Windows NT4. When I can no longer fix the computer, the bill for a modern replacement will be five, maybe six, figures.

  12. Lee Dowling Silver badge

    Intel NetPortExpress

    I have an Intel NetportExpress. It's the only thing I still own that has a 386 processor in it (and that's a 386SL - I know, because I cracked it open to have a look). And I doubt it runs Linux (I think it might be VxWorks but haven't bothered to look - it "just works" and has done for years).

    It networks a parallel port laser printer that you can't buy any more (Samsung ML-4500) which lets you use any toner in its refillable cartridges (until their attached drums wear out, when it's about £20 for a new cartridge + drum). I've had the same setup for so long, I've never had to add another printer on my home network.

    Though, it's a bit annoying because Windows 7 x64 doesn't seem to support the drivers for the printer (32-bit version works fine). There's probably a workaround that involves some sort of compatibility mode, but to be honest once I get to that point, I'll throw it onto a Linux machine as a CUPS printer and have done with it.... and then the last 386 in the house will be put into retirement.

    Not surprised, not shocked, not affected. I doubt whether anyone is. I got rid of my last 386 desktop machine something like 10+ years ago and that had been obsolete for a while, and run into the ground, and went to be a games machine for my cousin even after that. And embedded programmers have their own problems, and need to fix them themselves (if that means using an old kernel or patching the functionality back, or using a different chip, then they will).

    The oldest PC I have in the house now is something like a Pentium 133 laptop. Probably, what 3 generations above the 386? And that's never going to come off what it's currently running because of the problems of doing so (it is quite hilarious to have a ThinkPad that old, though, that was thrown into a skip nearly 8 years ago, have a PCMCIA card shoved into it and join a 802.11g network as if it was any other machine - and it still has one of nicest "feels" to using it of any machine I've ever owned).

    1. Cian Duffy

      Re: Intel NetPortExpress

      "Though, it's a bit annoying because Windows 7 x64 doesn't seem to support the drivers for the printer (32-bit version works fine). "

      Have you tried a generic PCL driver (not from a manufacturer but from a similar era PC - e.g. drive it with a HP Laserjet 4 driver)? This is what we do in work to get the various archaic Samsung and Brother printers that customers have and that will not die to work on x64 Windows.

  13. Nick Ryan Silver badge

    DX vs SX

    IIRC the 386 DX flew... the 386 SX wasn't really much faster than the 286 clock for clock, but could usually be found at slightly higher clock speeds. Pretty sure it wasn't just the lack of an integrated maths co-processor either.

    1. RichD

      Re: DX vs SX

      The 386 DX didn't have a math co-processor, difference between 386DX and SX was the size of the buses, 16-bit data bus and 24-bit memory bus, as opposed to the 32bit buses on the DX.

      It was the 486SX that lacked the math co-processor of it's DX sibling.

  14. Martin Huizing

    And I remember...

    combing my hair with a discarded 386 intel chip. The connectors worked like a fine comb!

    1. Chemist
      Joke

      Re: And I remember...

      "

      And I remember...

      combing my hair with a discarded 386 intel chip. The connectors worked like a fine comb!

      "

      Hope it got the bugs out !

      1. Lee Dowling Silver badge
        Angel

        Re: And I remember...

        I at least hope you earthed yourself first...

  15. Anonymous Coward
    Anonymous Coward

    Meh

    It was well over 10 years ago that we were forced to move from 386 to 486 on our embedded systems, as Intel was clawing back all the remaining 386 chips from manufacturers because NASA needed them for some project involving compatibility with existing system.

  16. Colin Miller
    Headmaster

    33 mHZ? Even Colossus ran at a few kHz. You probably meant 33 MHz.

  17. This post has been deleted by its author

  18. Anonymous Coward
    Anonymous Coward

    Methinks <b><i>this</i></b> would have been an appropriate juncture at which to jump from 2.6.xx numbering to 3.xx

    ho hum

    ...and to anyone jumping at the "grouse about your 386s being dropped" bait... what exactly is the hardware you have in your 386 systems which needs a kernel >3.7? Can't you just use a 2.6.xx or 2.4.xx kernel? Linux is FOSS so you can take your pick of any kernel there's ever been, download the source and do with it as you please.

  19. John Robson Silver badge

    No issue

    so long as the previous versions remain available from git for those who need them

  20. Anonymous Coward
    Anonymous Coward

    486 in 2013

    Presumably, someone could manufacture a 486 chip today running at 3ghz or higher. Out of interest, would there be any practical use for such a chip, even in a specialist field?

  21. BinkyTheMagicPaperclip Silver badge

    it was still supported?!

    Even NetBSD dropped 386 support over a year ago. I'm amazed Linux hasn't done it sooner.

    Also, IIRC the 286 was a bit faster than the 386, at least in DOS. 16MB protected mode memory did start restricting it though.

    1. /dev/null
      Facepalm

      Re: it was still supported?!

      Actually, NetBSD dropped the 386 in 2007, though the first formal release without 386 support was 5.0 in 2009.

      Ironically, the x86 port of NetBSD is still called "NetBSD/i386"...

  22. Richard 26

    Attribution?

    "This tree removes ancient-386-CPUs support and thus zaps quite a bit of complexity," are Ingo Molnar's words.

  23. LaeMing
    Holmes

    Whelp!

    No-one wanted to do the work for fun. No-one wanted to pay to have it done. That's the way OSS flys: no-one wants a feature enough to maintain it, it naturally drops out. (That is a good thing BTW!)

  24. Lockwood
    Trollface

    Evil Linux dropping support

    How can you people support an OS raising its minimum processor requirements a whole generation?

    It's an outrage to think that they can just axe support for the 386 like that!

    Where will it end? The next version requiring a 486 with 487, not just a plain 486?

    The line must be drawn here! This far, and no further!

    1. Michael Thibault

      Re: Evil Linux dropping support

      Captain! Relax.

      1. Destroy All Monsters Silver badge
        Devil

        Re: Evil Linux dropping support

        Clearly we need Netanyahu for a little pep speech about red lines etc.

  25. John Smith 19 Gold badge
    Thumb Up

    It's pragmatic.

    Who needs a desktop OS hosted on a processor last mfg 4 years ago?

    No one. This focuses finite resources on active platforms. The "centre of mass" of the community has shifted.

    The last people running 386 are people with ones buried inside various bits of exotic (and not so exotic) equipment. They already have a tool chain in place to support their bits and pieces. Will future upgrades benefit them? There UI is a 2 line LCD with a set of buttons and switches, or a set of 422 serial lines. At the bottom is the latest Linux assembler that much better?

    And it's not like previous versions will magically vanish from all websites worldwide is it?

    I wonder if this will actually shrink the kernel or will they pack other stuff instead?

    BTW Intel does supply compilers. I wonder if they still support the 386.

    1. Sir Codington
      Windows

      Re: It's pragmatic.

      I think it will be replaced by arm things. Everything is going arm and the guys making Linux realize that a crappy windows RT arm edition won't cut the mustard, so they are thankfully building all sorts of arm support into the kernel.

      Microsoft, take the hint. Support arm fully you plebs!

  26. Paul 37

    Windows 3.1

    The stupid thing is I'm not really doing anything today that I wasn't doing then

    WordPerfect, Lotus 123, Foxpro, Monarch and Corel may have bitten the dust (in varying degrees) but I'm esentially using something I could fly to the moon to write roughly the same documents, spreadsheets, small database systems and draw the odd diagram. The ony real difference is my email is now intregal, rather than stuck on an IBM green-screen in the corner.

    Progress anyone ?

  27. Justicesays
    Joke

    What he should have said

    Is that if the 386 lot don't like it, they can fork off. (the previous linux kernel)

  28. Morrie Wyatt
    Boffin

    6502 / Z80? Hah!

    Doesn't anyone else remember the fun to be had hand-keying the PDP 8 bootstrap sequence on the front panel switches?

    The 6502 was simple, but the Z80 had so many more registers to play with.

    Ah the memories!

    1. John Smith 19 Gold badge
      Happy

      Re: 6502 / Z80? Hah!

      " but the Z80 had so many more registers to play with."

      Enough to host an actual compiler in fact. Remember Turbo Pascal for CP/M?

      1. preppy
        Unhappy

        Re: 6502 / Z80? Hah!

        No...don't remember Turbo Pascal for CP/M.....but I do remember BDS C and dBASE II - both good tools in their day.

        The bad memories are all about trying to use Prestel using my Osborne 01's 300 baud modem.

    2. Anonymous Coward
      Anonymous Coward

      Re: 6502 / Z80? Hah!

      Not PDP-8, but one of its descendants the HP 1000.

      One one occasion when there was a disk failure we used the front panel switches to load a program which flashed the display register in meaningless patterns - we had the HP customer engineer going for a while when he arrived with the head crash fixit kit (they used to hate the 3 hr drive and fear the possibility that the problem they were facing did not correspond to the parts they'd brought).

  29. BinkyTheMagicPaperclip Silver badge

    Gone from Xen now too

    Perhaps a co-incidence but i386 support has just been dropped from Xen -current source today!

    Not sure I'd be masochistic enough to run Xen on that low a configuration, though..

  30. Anonymous Coward
    Anonymous Coward

    well if you will use linux or windows...

    what do you expect??? OSX runs like a dream on my 386

  31. Anonymous Coward
    Anonymous Coward

    ELKS

    There's always ELKS!

    http://elks.sourceforge.net/

    Or port RetroBSD: http://retrobsd.org/wiki/doku.php

  32. John L Ward

    We're all so busy thinking back...

    ..., yet as someone who went from home electronics (ah the joy of cutting Veroboard) through the early efforts of Sir Clive (dodgy 16k RAM packs anyone?) to the joy of the BBC B and on, I'm just wondering what kind of conversations my children (and theirs) will be having in years to come...

    "What? Apple are dropping support for the A4??? What about my iPad? Steve Jobs will be spinning in his grave!"

    "Who?"

    etc. etc.

    1. Nelbert Noggins

      Re: We're all so busy thinking back...

      Heh, I doubt it'll be anything like the thoughts where i "helped" my dad build a ZX81 because the kit form was much cheaper than the ready built version...

      or the joy of finally getting a program running on said ZX81....

      or the portable compaq luggable that was larger and heavier than his suitcase with a 4" green screen

      I'm fairly certain if he tries to get Internet near me it'll still result in "BT expect upto 512Mb broadband available" on your line though ;)

  33. Daniel von Asmuth
    Facepalm

    April fool

    This is just a rerun of the first of April, 2005.

    http://amailbox.org/april_fools

  34. Henry Wertz 1 Gold badge

    Nostalgia aside....

    Nostalgia aside... I did run Linux originally on a 386. But, that said, the hard memory limit on the 386 was 16MB (you could still have 4GB virtual memory...). So, shoving a 386 full of RAM to watch a newer distro crawl to a start is realistically not an option. Even most small distros (puppylinux, damn small linux) are not small enough to run in 16MB. Also most distros have been built for 486+ for years, the few new instructions 486 added over 386 ended up improving performance so significantly that distros dropped 386 years ago (and some are built i686 -- Pentium Pro -- on up.) So I'm not going to cry out over this.

    I think cmpxchg (atomic compare and exchange) is probably why the kernel guys are jonesing to drop 386, not having this instruction would make spinlocks and mutexes much harder (these are used on SMP systems to make sure a given resource is only accessed by one CPU at a time).

  35. Donald Becker

    Dropping 386 support impacts zero users.

    I initially used Linux ('MCC Interim') with a 386, but a bit over two decades ago (!) switched to a 486.

    I think I still have that 486 around for sentimental reasons.

    That 386 has no chance of running a modern Linux kernel. It had too little memory. You would have to strip the kernel down to uselessness to get it to load at all, then there would still be too little space for buffers.

  36. Wombling_Free
    Trollface

    But we was 'appy...

    Ceramic casing? LUXURY!

    All we 'ad were a burnt stick an' a coupla Napiers Bones.

  37. Anonymous Coward
    Devil

    Oooooo Errrrrrrrrrrrr 386?

    Ha... if you think that is fast - you MUST see this... It's a 486..................

    Oooooooooooooooooooooo I think I paid about $4000 for that when it first came out.

    $180 for a 4 or 8 meg stick of RAM - or was that 2 sticks of 4 meg of RAM?

    I wish I had of had of had what I have today back then...... twas such a bucket of shit.

This topic is closed for new posts.

Other stories you might like