Slashdot Mirror


Linux Letting Go: 32-bit Builds On the Way Out (theregister.co.uk)

An anonymous shares a report on The Register:Major Linux distributions are in agreement: it's time to stop developing new versions for 32-bit processors. Simply: it's a waste of time, both to create the 32-bit port, and to keep 32-bit hardware around to test it on. At the end of June, Ubuntu developer Dimitri Ledkov chipped into the debate with this mailing list post, saying bluntly that 32-bit ports are a waste of resources. "Building i386 images is not 'for free', it comes at the cost of utilising our build farm, QA and validation time. Whilst we have scalable build-farms, i386 still requires all packages, autopackage tests, and ISOs to be revalidated across our infrastructure." His proposal is that Ubuntu version 18.10 would be 64-bit-only, and if users desperately need to run 32-bit legacy applications, the'll have to do so in containers or virtual machines. [...] In a forum thread, the OpenSUSE Chairman account says 32-bit support "doubles our testing burden (actually, more so, do you know how hard it is to find 32-bit hardware these days?). It also doubles our build load on OBS".

378 comments

  1. Is it even possible to buy a new 32 bit chip? by Sasayaki · · Score: 3, Insightful

    I think that the trouble finding testing hardware is quite telling.

    Can end users even buy a new, off-the-shelf 32-bit system these days, except for specialized devices like embedded systems?

    Is there anything more than a relatively tiny fraction of the user base that is stuck on 32-bit hardware, that can't use virtual machines to run that software on something that's not a potato?

    And I mean, it's not like the old 32-bit versions of OS's are gone. Windows 95 is still around. It didn't go away. I'm willing to bet there are still Windows 95 machines running somewhere in mission critical systems in places around the world.

    Yes, there's no security updates, but just unplug it from the internet and you're safe from the vast majority of attacks, and if you're worried about local access to your Windows 95 machine... install a thicker door?

    At some point technology has to move on.

    --
    Check out my sci-fi book "Lacuna" at http://goo.gl/MVxX8
    1. Re:Is it even possible to buy a new 32 bit chip? by bugs2squash · · Score: 4, Interesting

      I don't happen to know, maybe someone has figures. But I would have thought that the 32-bit x86 embedded linux market was quite large.

      --
      Nullius in verba
    2. Re:Is it even possible to buy a new 32 bit chip? by Sasayaki · · Score: 1

      Sure and it is, but how many of them are running Ubuntu?

      --
      Check out my sci-fi book "Lacuna" at http://goo.gl/MVxX8
    3. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 1

      Yes, most 64 bit chips also run 32 bit software, so no, there's no problem finding testing hardware, that's absurd.

      The necessity is less the hardware, and more the amount of stuff that has to run in a 32 bit environment, from Firefox and Wine to old binaries - a problem that comes up more often than you'd think.

    4. Re:Is it even possible to buy a new 32 bit chip? by JustNiz · · Score: 3, Insightful

      Are you thinking just of people that still have ancient hardware, or do you also mean for new products?
      Is there really anything new that is still 32 bit? For sure Arm, Atom and of course desktop CPUs have all gone 64 bit ages ago. What's really left?

    5. Re:Is it even possible to buy a new 32 bit chip? by bill_mcgonigle · · Score: 1

      Can end users even buy a new, off-the-shelf 32-bit system these days, except for specialized devices like embedded systems?

      While 32-bit embedded is certainly a huge market, the use for 32-bit Ubuntu isn't new off-the-shelf systems, it's grandma's old XP machine that needs rescuing. Embedded gear is more likely to run a lite debian.

      But, perhaps Ubuntu isn't the right choice for Grandma at this point. Probably more people would go for Mint anyway (they're more likely to be impacted by Ubuntu's decision). My 'senior' mother is running CentOS 7 with XFCE and, not knowing anything about computers, doesn't see how it's any different than Windows.

      Ubuntu focusing on 64-bit probably won't affect many people at all. It might even give Mint new relevance if they maintain a 32-bit port.

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
    6. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      I would have figured that ARM (and RISC in general) had the embedded market in a tight grip.

    7. Re:Is it even possible to buy a new 32 bit chip? by LichtSpektren · · Score: 1

      Yes, most 64 bit chips also run 32 bit software, so no, there's no problem finding testing hardware, that's absurd.

      Nobody cares if 32-bit ISOs work on 64-bit CPUs because those people are already using a 64-bit ISO. The 32-bit ISO has to be tested on ancient hardware like i686, which there's a shortage of.

    8. Re:Is it even possible to buy a new 32 bit chip? by angel'o'sphere · · Score: 3, Informative

      The majourity of ARMs sold are still 32 bit, and there is no reason for an embedded system to go 64 bit, usually.

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    9. Re:Is it even possible to buy a new 32 bit chip? by MyFirstNameIsPaul · · Score: 1

      Would not having a maintained OS be reason enough for them?

      --

      I once took an excursion to Reddit, and later HN. Unlimited up/down voting sucks when dealing with a hive-mind.

    10. Re:Is it even possible to buy a new 32 bit chip? by gman003 · · Score: 1

      For computers? Quite some time. There was a one-off Atom netbook chip back in 2008, and before that was Core (the predecessor to the more popular, 64-bit capable Core 2) in 2006-2007 and some of the early Pentium 4s up to 2005. On the AMD side, you have to go back to K7, which stopped being made in 2005. So everything that you'd want to run a desktop distro on is at least eight years old.

      Intel did make x86-32-only chips for smartphones until much more recently, but you wouldn't want to run a desktop distro on those, anyways. And it's not like the Linux kernel is dropping support for it, so whatever weird hack project you might theoretically want to make with a bunch of old smartphones is still just as doable.

    11. Re:Is it even possible to buy a new 32 bit chip? by jandrese · · Score: 1

      Aren't most Atom chips IA32? While this is technically "embedded systems", in practice you generally run a full Linux distro on it.

      --

      I read the internet for the articles.
    12. Re:Is it even possible to buy a new 32 bit chip? by freeze128 · · Score: 2

      How would running a 32-bit program in a virtual machine on 64-bit hardware help? You would still need a 32-bit OS, right?

    13. Re:Is it even possible to buy a new 32 bit chip? by fredgiblet · · Score: 1

      "I'm willing to bet there are still Windows 95 machines running somewhere in mission critical systems in places around the world."

      As of about 5 years ago there was one in a paper mill I was working at. Since they were only using it because there was no new drivers for the thing it was plugged into I'm willing to bet it's still there unless it broke.

      The IT guy had a copy of Windows 1.0 in a drawer in the server room too.

    14. Re:Is it even possible to buy a new 32 bit chip? by fredgiblet · · Score: 1

      Chicken and egg. If there's still maintained OSes running 32-bit then there's no reason for them to swap over, so the swap won't happen unless it's forced. If it's forced then all the systems manufactured until then are made obsolete, and a lot of embedded systems don't have high turnover rates.

    15. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      New stuff has moved over to ARM. Embedded stuff doesn't really get upgraded very often (if ever), so those don't matter.

    16. Re:Is it even possible to buy a new 32 bit chip? by Fwipp · · Score: 1

      Cool, but this story is only talking about the i386 architecture (32-bit x86).

      32-bit ARM chips will still be supported.

    17. Re:Is it even possible to buy a new 32 bit chip? by bluefoxlucid · · Score: 1

      I think the argument is dumb.

      Duuuuuuuumb.

      Dumb.

      Boot an AMD64 into 32-bit mode.

      There were late-generation 32-bit Pentium 4 models without the CMOV prefixed instructions that became standard in i586. CMOV is a processor extension. That's how radically different 32-bit real hardware can be. A 64-bit Intel or AMD processor in 32-bit mode is no less of a complete test than any actual 32-bit processor, simply because testing Pentium-2 compiled IA-32 code on an actual Pentium-2 is as much of a test for a 32-bit Pentium 4 as testing that same code on a Core i7 in 32-bit mode.

    18. Re:Is it even possible to buy a new 32 bit chip? by phoenix_rizzen · · Score: 1

      32-bit only hardware? No.

      Able to run in 32-bit mode? Every single Intel and AMD CPU out there supports it. Why would anyone consider this to be "hard to find"?

    19. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      I think that 32-bit support may be offered much longer for specific generic VM platform variants, where testing can be virtualized and there are fewer chipset/controller driver variations to deal with. Similarly, multilib support for 32-bit processes could live on in a 64-bit distribution. But, abandoning support for bare-metal 32-bit installs in a mainstream distribution means that they no longer need to maintain a fleet of hardware and users to perform bare-metal testing to ensure that the distribution actually supports installation and operation on said hardware.

    20. Re:Is it even possible to buy a new 32 bit chip? by Misagon · · Score: 1

      It is not about the processor architecture, but the amount of RAM.
      4 GB is about the threshold of when x86-64's page tables are starting to not cost more than it provides.

      There are lots of new low-end machines sold even today with no more than 4GB RAM or even 2GB RAM - and a quarter of that is often going to be dedicated to the integrated graphics. These machines often come with 32-bit Windows preinstalled.
      If you want a 8" tablet, it is practically only the more expensive high-end models that come with more than 2 GB RAM.

      --
      "We mustn't be caught by surprise by our own advancing technology" -- Aldous Huxley
    21. Re:Is it even possible to buy a new 32 bit chip? by Carewolf · · Score: 1

      Aren't most Atom chips IA32? While this is technically "embedded systems", in practice you generally run a full Linux distro on it.

      Only the original Atom from the netbook days. All atoms since have had AMD64 support.

    22. Re:Is it even possible to buy a new 32 bit chip? by guruevi · · Score: 1

      unplugging never works. Security through obscurity is a bad plan especially when in the future your techs do need/want remote access.

      There are plenty of Intel 32 bits systems being sold for embedded devices in factories etc. Intel is still developing 32-bit compatible processors (Quark)

      --
      Custom electronics and digital signage for your business: www.evcircuits.com
    23. Re:Is it even possible to buy a new 32 bit chip? by Darinbob · · Score: 4, Insightful

      What the hell? 32 bit cpus are everywhere. The article is talking about PC builds, x86 clones in other words, only a Wintel person actually thinks that is the only arcthitecture out there. Meanwhile if you look at the Linux kernel it has 29 different architectures it supports.

    24. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      x86 and 32 bit ARM are 2 very different and incompatible architectures. Arm devices by and large are stuck with the OS they shipped with, this is why your phone and tablet are appliances and not computers, they're little different then the processor in your washing machine.

      But even there, with Cortex A-53 and Cortex A-57 ARM has moved to 64 bit, as these appliances are stuck with the OS they came with as well, the changeover will be far more rapid.

      Also, ram usage, even with only 1Gb of ram is minimal going from 32-64 still nets you higher all around performance.

    25. Re:Is it even possible to buy a new 32 bit chip? by Blaskowicz · · Score: 1

      With 1GB RAM your browser will easily fill 80% of it. Best if everything else fits under 100MB, plus a graphics you can hopefully get down to 32MB or something not too large.

    26. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      The reason they come with 32-bit Windows is because they use flash storage which is still kind of expensive, and using 64-bit Windows would basically make it use up twice as much storage.
      A problem that could be solved medium-term (together with most of the increased RAM usage) if Microsoft pushed for a "64-bit only by default" Windows, like the Linux distributions and possibly soon OSX are.

    27. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      An embedded system isn't going to be running Ubuntu or openSUSE.

    28. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      I'm willing to bet there are still Windows 95 machines running somewhere in mission critical systems in places around the world.

      Given that the OS has to be rebooted every 49.7 days, they can't be that critical. Also, a system like that is living on borrowed time already. Once the hardware dies, it's going to be pretty tough finding a replacement supported by the OS. However ultimately it is like other posters have said, no embedded systems vendor is going to care about what Ubuntu does.

    29. Re:Is it even possible to buy a new 32 bit chip? by Trogre · · Score: 1

      While 32-bit desktops are pretty much dead, there are plenty of laptops still running 32-bit processors. Not new ones, granted, but ones that came out in the last ten years.

      --
      "Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
    30. Re:Is it even possible to buy a new 32 bit chip? by caseih · · Score: 1

      But in embedded space 64-bit has no real advantage, and may even be a very slight disadvantage in terms of memory consumption (64-bit pointers, etc).

      The Raspberry Pi 3 is based on a 64-bit Arm processor, but all the distros available for it are all 32-bit right now. This might have to do with the lack of information from Broadcom. There's no data sheet available without signing an NDA about that chip, so it's hard for open source developers to support the more advanced features. In fact, even the exact electrical specifications of the gpio pins aren't really known.

    31. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      Thing is that there is a ton of still used 32 bit machines out there. For example my home desktop machine and one of my laptops. As with many, many people they continue to function perfectly well so I have no desire to upgrade.

      Many of my fellow techies are still stuck on the perpetual upgrade cycle and so fail to understand just how many people have dropped out and moved on to "use it till it breaks" approach.

    32. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      of course you can buy 32 bit x86 processors today. EVERY AMD64 PROCESSOR, whether it is made by AMD or Intel, ever sold, or is sold today, or will be sold tomorrow, supports 32 bit x86 i386 instructions. EVERY FUCKING ONE.

    33. Re:Is it even possible to buy a new 32 bit chip? by PRMan · · Score: 1

      I bought an HP Stream 7 about a year ago. At least on Windows 10, it's 32-bit.

      --
      Peter predicted that you would "deliberately forget" creation 2000 years ago...
    34. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      The mainline distros are not what you use for an embedded solution.

      Linux has matured where there are embedded distros for embedded hardware, and really, if you are talking about embedded hardware, you don't run a whole OS on it, you just run the one program directly on the hardware.

    35. Re:Is it even possible to buy a new 32 bit chip? by JustNiz · · Score: 3, Informative

      I think you're wrongly painting the entire embedded space with the same brush. I work in said space. Sure there are toaster controllers or whatever that would happily run on an 8 bit pic, but honestly there is a lot more heavy duty stuff going on than that in the majority of the embedded world.
      At least everything I've worked on in the last 20+ years would have benefited from 64 bit/moar powa!!!.

    36. Re:Is it even possible to buy a new 32 bit chip? by ChunderDownunder · · Score: 1

      Someone did mention recently hacking a 2013 era shoddy Windows tablet to run 32 bit Linux. The Atom chip itself was perfectly 64 bit capable but they'd botched the UEFI implementation to be only 32 bit.

      By way of anecdote, I'm not sure Canonical should expend the manpower to keep 32 bit Ubuntu alive solely just for such corner cases. Caveat emptor.

    37. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      The x64 box I am typing this on is 64 bit compat and 32 and 16. I think it can even go into 8 bit mode (my mind is a bit fuzzy on that detail).

      Finding hardware is dead easy. Pretty much every recent intel and AMD chip out there pretty much can do it.

    38. Re:Is it even possible to buy a new 32 bit chip? by shutdown+-p+now · · Score: 1

      Intel has some good offerings, too. But they're all x64 by now. There really isn't any case for i386.

    39. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      What is telling is this "story". It's based on a post by a Ubuntu developer.... which says fuck all about Linux
      Ubuntu (Swahili for mostly broken, Bantu for Debian is too hard - and doesn't phone home) is mostly just repackaged (with bugger all useful changes) Debian.

      tl;dr? Next up is a story about a guy who makes sandwiches predicting the end of sliced bread.

    40. Re:Is it even possible to buy a new 32 bit chip? by dryeo · · Score: 2

      Anything with 2Gbs or less memory might as well run on 32 bit, if only for the memory savings. Wife has a fairy new tablet running 32 bit Win 8.1, Atom processor and 1GB of memory with 16 GBs of storage. Be stupid to run a 64 bit OS on that. Son has a slightly older netbook with an Atom processor and 2 GBs of memory, runs 32 bit Linux (Ubuntu) on it. I have a TP 42, 12 years old and limited to 32 bit, still runs well but will never support 64 bit, (actually most Linuxes won't run on it without a custom kernel due to a bug in the CPU where cpuid doesn't correctly show support for PAE) runs very well with OS/2.

      --
      https://en.wikipedia.org/wiki/Inverted_totalitarianism
    41. Re:Is it even possible to buy a new 32 bit chip? by perpenso · · Score: 1

      The majourity of ARMs sold are still 32 bit, and there is no reason for an embedded system to go 64 bit, usually.

      Would not having a maintained OS be reason enough for them?

      Maintained by who, the desktop and server distro maintainers or the embedded distro maintainers? They aren't necessarily the same groups.

    42. Re:Is it even possible to buy a new 32 bit chip? by perpenso · · Score: 2

      Yes, most 64 bit chips also run 32 bit software, so no, there's no problem finding testing hardware, that's absurd.

      Nobody cares if 32-bit ISOs work on 64-bit CPUs because those people are already using a 64-bit ISO. The 32-bit ISO has to be tested on ancient hardware like i686, which there's a shortage of.

      No, 32-bit operating systems run just fine on 64-bit CPUs in the x86 world. Some people do so for improved performance. 64-bit sometimes has a performance hit.

    43. Re:Is it even possible to buy a new 32 bit chip? by perpenso · · Score: 1

      Able to run in 32-bit mode? Every single Intel and AMD CPU out there supports it.

      Can't the same be said for 16-bit mode?

      Why would anyone consider this to be "hard to find"?

      The problem is they are testing a complete PC environment. So its not just the CPU, its that AGP 3dfx voodoo card and PCI sound blaster card too.

    44. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      The majority of ARM are not i586 builds.

    45. Re:Is it even possible to buy a new 32 bit chip? by TheRaven64 · · Score: 1

      Add to that, the stuff where AArch64 really wouldn't make sense is typically M- or R-profile. ARMv8-M doesn't include AArch64 for this reason (amusingly, the early versions of the ARMv8 spec said that AArch32 and AArch64 were both optional and forgot to mention that you needed at least one, so it was possible to have a fully conforming implementation that implemented no instructions). If you're using A-profile then you're already at the beefy end of embedded, which goes all of the way up to 100+TB storage appliances with multi-socket multi-core CPUs.

      --
      I am TheRaven on Soylent News
    46. Re:Is it even possible to buy a new 32 bit chip? by TheRaven64 · · Score: 1

      64-bit sometimes has a performance hit.

      LP64 has a performance hit, but x86-64 doesn't require ILP64, you can happily run ILP32 code with the X32 ABI in long mode.

      --
      I am TheRaven on Soylent News
    47. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      Nobody cares if 32-bit ISOs work on 64-bit CPUs because those people are already using a 64-bit ISO.

      Windows users perhaps, because they need to reinstall every couple of years anyway, because of registry leak.

      My Broadwell Core i7 5775c machine runs 32 bit Linux, because that's all that was available when I installed Linux back in 1999. Guess what, Linux runs just fine without reinstalling, because Linux doesn't have a registry in the first place.

      My latest hardware replacement, I just moved the hard drive over. Oh, there was a couple of drivers I needed to add (that's what I get for running a custom kernel with only the drivers I need). The last couple of times I replaced my hard drive, I copied the OS over.

      Either of those took several orders of magnitude less time that setting up a new install to fit my way of doing things as perfectly as my current install which has been tweaked over 17 years.

    48. Re:Is it even possible to buy a new 32 bit chip? by drinkypoo · · Score: 1

      While 32-bit embedded is certainly a huge market, the use for 32-bit Ubuntu isn't new off-the-shelf systems, it's grandma's old XP machine that needs rescuing. Embedded gear is more likely to run a lite debian.

      Grandma's old XP machine can fuck off. You can buy a SFF C2D system with 4GB RAM for $50 on eBay. I know, as I've done it, I used it as my router for a while before the WRT1200AC came down to low low prices as a refurb. I have literally three C2D systems lying around my house rusting.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    49. Re:Is it even possible to buy a new 32 bit chip? by bluegutang · · Score: 1

      If you're a large business selling devices with embedded Linux, you should be able to compile your own build.

    50. Re:Is it even possible to buy a new 32 bit chip? by Eunuchswear · · Score: 1

      You want to tell my Grandma to fuck off?

      Well, you can't. 'Cos she's dead, you bastard.

      (Why do you want a Sydney Film Festival crash to desktop system anyway? And why does it contain enough iron to rust?)

      --
      Watch this Heartland Institute video
    51. Re:Is it even possible to buy a new 32 bit chip? by drinkypoo · · Score: 1

      You want to tell my Grandma to fuck off?
      Well, you can't. 'Cos she's dead, you bastard.

      I'm sure she was lovely when she was alive, unless she wasn't, and in any case I'm talking about her computer — which could have been replaced for so little money it'd just not have been worth messing with.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    52. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      They are selling UNICODE to YOU so that any Arab may snatch your chip and say he is a Chinese in Africa... written in Russian.

    53. Re:Is it even possible to buy a new 32 bit chip? by hsu · · Score: 1

      Besides, by replacing the pic toaster controller with bunch of modern 64 bit cpus putting out 100 Watts each you can get rid of those pesky resistor wires, and as an added bonus crunch bitcoins while toasting your breakfast slice.

    54. Re:Is it even possible to buy a new 32 bit chip? by hsu · · Score: 1

      There are plenty of embedded i386 stuff around there. However, a lot of this will not necessarily need ubuntu to provide binary distribution for them. Removing support from kernel would be a bad idea. However, I assume that was not intention, just that mainstream distros focusing on servers and desktops might drop non-64bit binary distributions.

      Embedded stuff has lot longer lifetimes than PCs, and it is not just ARM there. MIPS 32bit architecture is ubitous in WLAN routers, including ones being designed today, and there are new i386 architecture chips being sold today as well. 8 bit stuff is slowly being replaced with those, and as far as I can tell, there are very few 64 bit CPUs or SoCs at 2 USD range. I would not be surprised if there were more Linux-running devices being sold with 32 bit CPU today than there are 64 bit Linux anything.

    55. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      > Cool, but this story is only talking about the i386 architecture (32-bit x86).

      Please note that i386 != 32-bit...

      32-bit comprises 386, 486, 586 _and_ 686 (yes, there are 686 CPUs which are 32-bit only).

      386 is a special case because there is a lot to do to compensate the lack of some resources in that processor. Support by the Linux kernel has already been deprecated. I guess it's safe to say it's hard to come by a genuine 386 hardware.

      Not so with 32-bit: tons of apps work well with just 32-bits.

      Besides, as you mention:

      > 32-bit ARM chips will still be supported.

      A fact that makes wanting to abandon 32-bit even more silly IMHO...

    56. Re:Is it even possible to buy a new 32 bit chip? by Anonymous Coward · · Score: 0

      I don't know where you're working but the majority of embedded systems don't even need 32 bits, never mind 64 bits.
      You see, these chips are called controllers for a reason. Their main reason of existance it to control things, not to calculate large calculations or address piles of memory. In almost all controller applications 64 bits cannot be properly utilized anyway and will, in fact, make things slower and eat up more memory (which is relatively scarce in controller land).
      64 bits equals moar power only if you can utilize it, if it actually fits your problem.

  2. 32-bit hardware by Anonymous Coward · · Score: 0

    "doubles our testing burden (actually, more so, do you know how hard it is to find 32-bit hardware these days?)" I can find 32-bit hardware easily. Obviously, someone who didn't try and claims it not possible.

    1. Re:32-bit hardware by LichtSpektren · · Score: 1

      Specifically IA-32 hardware, not anything 32-bit.

    2. Re:32-bit hardware by Anonymous Coward · · Score: 0

      Specifically IA-32 hardware, not anything 32-bit.

      I don't disagree with a decision to drop support for old hardware, but what about some other level of support using emulated hardware to at least give some degree of support?

    3. Re:32-bit hardware by LichtSpektren · · Score: 2

      Specifically IA-32 hardware, not anything 32-bit.

      I don't disagree with a decision to drop support for old hardware, but what about some other level of support using emulated hardware to at least give some degree of support?

      I think you're missing what the point of "support" is. A company can release a binary and say they support it, without testing it and not fixing any bugs for it, but that makes them look like assholes.

    4. Re:32-bit hardware by TemporalBeing · · Score: 2

      "doubles our testing burden (actually, more so, do you know how hard it is to find 32-bit hardware these days?)" I can find 32-bit hardware easily. Obviously, someone who didn't try and claims it not possible.

      Agreed. There is a lot of hardware - even new hardware - that is 32-bit; while especially the case in non-x86 systems, there is even x86 systems that are still being shipped in 32-bit mode (e.g 32-bit OS) by default or are 32-bit only, especially in the embedded world - and yes, many of those embedded devices may still operate a GUI interface.

      Example: A previous employer was converting from DOS to Linux. We had a GUI interface and the developer just loaded up X-Windows (GNOME I think) and then made a full screen GUI app on top of that. The basic use case was an embedded system (VMIC 7805 Board, Pentium M, - which still sells new) and we were using Ubuntu 8.04 as a base at the time (8.04 was newly released). (FYI VMIC 7805 runs a number of difference OS systems from Windows, Linux, VxWorks, and even DOS.)

      So yes, this guy is extremely short sighted - probably looking at MicroCenter or Best Buy and saying "well, no 32-bit systems here, guess we can't buy any". There is plenty of manufacturers you can find to build a 32-bit only system that are desktop oriented.

      Now, where it may get harder is finding a 32-bit system that is server oriented since most server builders are looking towards packing in the memory beyond the capabilities of a 32-bit system. Even so, Canonical should have no issue there since they are also doing some system building and could just build their own for their dev farm or just build images in 32-bit only mode to be tested in servers in a datacenter running VMWare, KVM, Xen, etc - a 32-bit guest OS is no issue in a 64-bit host environment.

      --
      Truth is like the sun. You can shut it out for a time, but it ain't goin' away. - Elvis Presley (source: imdb.com)
    5. Re:32-bit hardware by techno-vampire · · Score: 2

      Now, where it may get harder is finding a 32-bit system that is server oriented since most server builders are looking towards packing in the memory beyond the capabilities of a 32-bit system.

      The whole argument is nothing more than a straw man. All you have to do is have a multi-boot system where one of the images is 32 bit. Sure, you won't be able to take advantage of all of the RAM, but it will run, just fine. And, you can have one using a PAE kernel so that you can test programs in that environment as well. (Yes, people do use PAE. It's for when you have more than 4 GB RAM but either don't want to nuke, pave and reinstall a 64 bit system or can't because you can't afford the downtime. All you need to do for that is install a PAE kernel and support packages, reboot into it and later, remove the non-PAE packages. And yes, I'm writing from personal experience.)

      --
      Good, inexpensive web hosting
    6. Re:32-bit hardware by TemporalBeing · · Score: 1

      Now, where it may get harder is finding a 32-bit system that is server oriented since most server builders are looking towards packing in the memory beyond the capabilities of a 32-bit system. The whole argument is nothing more than a straw man. All you have to do is have a multi-boot system where one of the images is 32 bit. Sure, you won't be able to take advantage of all of the RAM, but it will run, just fine. And, you can have one using a PAE kernel so that you can test programs in that environment as well. (Yes, people do use PAE. It's for when you have more than 4 GB RAM but either don't want to nuke, pave and reinstall a 64 bit system or can't because you can't afford the downtime. All you need to do for that is install a PAE kernel and support packages, reboot into it and later, remove the non-PAE packages. And yes, I'm writing from personal experience.)

      All true; however, you can't test chip sets that are only available on native 32-bit-only hardware. So there is a limit to what can be tested, which is almost entirely a kernel-land issue not a user-land issue. For user-land, what is not available could have kernel-level emulators used if you really wanted to do so; though timing might change slightly so it's still not the best but it would allow the user-land software to at least function.

      --
      Truth is like the sun. You can shut it out for a time, but it ain't goin' away. - Elvis Presley (source: imdb.com)
  3. 32 bit had to go by Anonymous Coward · · Score: 0

    You can blame Cankles all you want but it was time.

  4. What is OBS? by Anonymous Coward · · Score: 0

    (Tried googling)

    1. Re:What is OBS? by Anonymous Coward · · Score: 0

      OpenSuSE Build System found at https://build.opensuse.org/

  5. Ability to run on old hardware was a feature by DutchUncle · · Score: 1

    ... though I suppose one can continue to run an old version. And it's not reasonable to expect open source volunteers to do double work. But it is still a loss.

  6. really? by Anonymous Coward · · Score: 0

    So tried E-bay (feebay) and there seem to be a near unlimited number of 32-bit computers for sale.

    1. Re:really? by Anonymous Coward · · Score: 0

      Yeah, I'm on one now, that came off of Freebay, and it hums.

  7. Re:That's just great... by Sasayaki · · Score: 5, Insightful

    Well, I mean, if you're running Windows 10 right now... uhh.

    Lemme rephrase. If Ubuntu 18.10 is 64-bit only, is that a problem? What show-stopping problem for a 2006 MacBook is present in 18.04 but fixed in 18.10?

    What's wrong with running 18.04 until the hardware dies?

    --
    Check out my sci-fi book "Lacuna" at http://goo.gl/MVxX8
  8. it's easy to find 32 bit Hardware by Jazoray · · Score: 1

    At least for intel Archs, you can install a 32 Bit OS on a computer with a 64 bit capable cpu.

    1. Re:it's easy to find 32 bit Hardware by LichtSpektren · · Score: 1

      At least for intel Archs, you can install a 32 Bit OS on a computer with a 64 bit capable cpu.

      *facepalm*

      They're talking about x86 CPUs, like the i386, which cannot use the amd64 build because they don't support 64-bit. That's why it's hard to test.

    2. Re:it's easy to find 32 bit Hardware by jedidiah · · Score: 2

      They can still test the software on "better" hardware. They can also run it in a VM. That's what they expect everyone else to do.

      It may not be "optimal" but it's certainly possible.

      That's not even getting into the fact that they aren't really trying very hard to find 32bit x86 hardware.

      --
      A Pirate and a Puritan look the same on a balance sheet.
    3. Re:it's easy to find 32 bit Hardware by LichtSpektren · · Score: 1

      They can still test the software on "better" hardware. They can also run it in a VM. That's what they expect everyone else to do.

      It may not be "optimal" but it's certainly possible.

      That's not even getting into the fact that they aren't really trying very hard to find 32bit x86 hardware.

      I'm having difficulty apprehending what's so difficult to understand about this.

      VMs and 64-bit CPUs can use the amd64 image, so there's literally no point in testing the IA-32 image on an amd64 CPU. The point of supporting an IA-32 ISO is so people on i686s and Core Duos can still use them. And you have to *test* to make sure the IA-32 image actually works on those CPUs, else you're at best wasting those peoples' time, at worst wrecking their computers.

    4. Re:it's easy to find 32 bit Hardware by Jazoray · · Score: 0

      if you install a 32Bit OS on an AMD64-capable cpu, it will behave as if it was x86, no? So just use x86_64 (AMD64) cpus and use them to test both the 64 bit and the 32 bit builds.

    5. Re:it's easy to find 32 bit Hardware by tlhIngan · · Score: 1

      VMs and 64-bit CPUs can use the amd64 image, so there's literally no point in testing the IA-32 image on an amd64 CPU.

      Wrong. Just because it can run 64-bit code doesn't mean it should.

      There are plenty of systems SOLD TODAY that have 64-bit processors, but don't have >4GB of RAM. Lots of cheapie x86 tablets, x86 media sticks (Intel Compute Stick, anyone) etc? THey have 64-bit processors, but 1 or 2 GB of RAM.

      Yes, you can run 64-bit software on it, but should you? After all, the 64-bit software takes more RAM over 32=bit counterparts, and you're already on a RAM-starved system...

      (Those Intel Atoms can run 64-bit code, but have memory controllers that top out at 2GB...).

    6. Re:it's easy to find 32 bit Hardware by angel'o'sphere · · Score: 2

      If they expect "every one else" to run 32 bit software in VMs, they actually should test their 32 bit builds in VMs.

      "The devil is a squirrel" as we say in german ...

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    7. Re:it's easy to find 32 bit Hardware by 93+Escort+Wagon · · Score: 1

      That's not even getting into the fact that they aren't really trying very hard to find 32bit x86 hardware.

      It's always simple when it's other people's time and money you're spending.

      --
      #DeleteChrome
    8. Re:it's easy to find 32 bit Hardware by LichtSpektren · · Score: 1

      You seem to think that the only advantage amd64 has over IA-32 is that you can utilize more RAM, which is not the case.

    9. Re:it's easy to find 32 bit Hardware by armanox · · Score: 1

      Not what they mean by supporting. Building it isn't good enough. If you say your software supports i586 then it needs to be tested on a Pentium I, i686 on Pentium II and III, etc.

      --
      I'm starting to think GNU is the problem with "GNU/Linux" these days.
    10. Re:it's easy to find 32 bit Hardware by DarkOx · · Score: 2

      Yes, you can run 64-bit software on it, but should you?

      Yes you should, without a doubt, unless you know for sure the main bottleneck for your application will be memory.

      The 64-bit modes offer more registers, and instructions that can do more operations per cycle as well as other optimized instructions. In most cases the performance will be much better!

      --
      Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
    11. Re:it's easy to find 32 bit Hardware by DarkOx · · Score: 2

      Well yes and no, there are some surprises from time to time. Like you might expect passing arch=i486 to gcc would give you code that would run on a 486, not always unless you pass other options as well. The problem is testing on the latest x64 chip means you are going to have a super set of the the i686 ISA in most cases, and with the wrong compiler flags or kernel build options, you might discover it really does not work on older hardware.

      --
      Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
    12. Re:it's easy to find 32 bit Hardware by TemporalBeing · · Score: 1

      They can still test the software on "better" hardware. They can also run it in a VM. That's what they expect everyone else to do.

      It may not be "optimal" but it's certainly possible.

      That's not even getting into the fact that they aren't really trying very hard to find 32bit x86 hardware.

      I'm having difficulty apprehending what's so difficult to understand about this. VMs and 64-bit CPUs can use the amd64 image, so there's literally no point in testing the IA-32 image on an amd64 CPU. The point of supporting an IA-32 ISO is so people on i686s and Core Duos can still use them. And you have to *test* to make sure the IA-32 image actually works on those CPUs, else you're at best wasting those peoples' time, at worst wrecking their computers.

      You can do those tests on an amd64 system host using a VM just fine. The AMD64 architecture is backwards compatible to x86-32 (IA32). A 32-bit guest is fine for a 64-bit host (not the other way around though).

      What's funny is that it's not the Kernel Devs that are complaining. It's Ubuntu which does their own system building (https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&uact=8&ved=0ahUKEwi_r82Hld3NAhUMwYMKHYkGB2AQFggkMAE&url=https%3A%2F%2Finsights.ubuntu.com%2Fwp-content%2Fuploads%2FDS_The_Orange_Box.pdf&usg=AFQjCNGU-_hJXlWHrI2WpOZEOFwz3Er9ag&sig2=hygyf6xsRkCT702uT6ZCNQ). So it's not like Canonical couldn't put together the resources they want to. Even Debian doesn't have an issue, nor does Slackware - both of which have far fewer resources than Canonical does.

      --
      Truth is like the sun. You can shut it out for a time, but it ain't goin' away. - Elvis Presley (source: imdb.com)
    13. Re:it's easy to find 32 bit Hardware by Anonymous Coward · · Score: 0

      Without real hardware you can't even be sure your install images will boot, because of all kinds of weird BIOS and chipset crap. You also have to make sure you haven't slipped in instructions from a newer ISA by mistake. VMs are only as good as the emulation. And if the VMs are tested against the software that uses the VM as a testbed, you've just created a feedback loop of nonsense.

    14. Re:it's easy to find 32 bit Hardware by Anonymous Coward · · Score: 0

      Yes, you should. Have you ever heard of "virtual memory" and paging? You cannot, for example, memory map a 8 GB file if you only have 2, 3 or 4 GB of address space. Even worse, the process might already have fragmented that address space so that largest continuous available address space is only few hundred megabytes. Of course, if you want to much about with sliding the mapping windows about or use fopen be my guest and stay in the 1980's.

    15. Re:it's easy to find 32 bit Hardware by drinkypoo · · Score: 1

      "The devil is a squirrel" as we say in german ...

      But Germans can't actually say squirrel!

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    16. Re:it's easy to find 32 bit Hardware by dave420 · · Score: 1

      Hence them using the far-more-easily-pronounceable "EichhÃrnchen"... :-P

    17. Re:it's easy to find 32 bit Hardware by Anonymous Coward · · Score: 0

      Unless the application requires a newer video card or lots of ram, I have not found that to be true. You are correct about the hardware capabilities. So I have often wondered if it is the smaller binaries loading in to ram, or maybe we just haven't optimized our systems to fully take advantage of the performance capacities when using 64bit systems (poor coding habits?).

    18. Re: it's easy to find 32 bit Hardware by Anonymous Coward · · Score: 0

      It is easier if you speak German natively. The diphthong in squirrel would be tricky for someone whose native tongue doesn't really have them.

    19. Re:it's easy to find 32 bit Hardware by angel'o'sphere · · Score: 1

      Rofl.

      Was about to answer the same, and what is difficult to pronounce in "Squirrel" is byond me :D

      --
      Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
    20. Re:it's easy to find 32 bit Hardware by Lotus456 · · Score: 1

      Ask a Japanese native to say "squirrel." You'll probably hear something like "skuah-do" or "skuah-ru." :)

      --
      "It's a good computer... for I to BM on!" - apologies to Triumph, the insult comic dog
  9. Re: That's just great... by Anonymous Coward · · Score: 0

    I ain't voting for Cankles. Bernie isn't the nominee so Trump it is

  10. containers or virtual machines for apps? by Joe_Dragon · · Score: 1

    Why not just drop the boot 32bit part and only have the compact 32bit libs. Like how windows server 2008 and newer is on the windows side.

    Why cut off apps that can run today on a 64bit system with out needing any vm bs.

    1. Re: containers or virtual machines for apps? by LichtSpektren · · Score: 2

      Why not just drop the boot 32bit part and only have the compact 32bit libs. Like how windows server 2008 and newer is on the windows side.

      Why cut off apps that can run today on a 64bit system with out needing any vm bs.

      If you had read TFA, you might have found out that's exactly what most Linux distros are doing. 32-bit library support isn't going anywhere, just .ISO builds for the i686 and older.

    2. Re: containers or virtual machines for apps? by StayFrosty · · Score: 1

      That is the plan as described by TFA.

      --
      "Frequently wrong, never in doubt."
    3. Re: containers or virtual machines for apps? by Anonymous Coward · · Score: 1

      If you had read TFA, you might have found out that's exactly what most Linux distros are doing. 32-bit library support isn't going anywhere, just .ISO builds for the i686 and older.

      That's not what the Ubuntu mailing list post says. They're ditching 32-bit library support for Ubuntu 18.10+.

      16.10, 17.04, 17.10:
      * continue to provide i386 port to run legacy applications on amd64 ...
      18.04 LTS:
      * continue to provide i386 port to run legacy applications on amd64 ...
      18.10+:
      * Stop providing i386 port
      * Run legacy i386 only application in snaps / containers / virtual machines

    4. Re: containers or virtual machines for apps? by LichtSpektren · · Score: 1

      The thing you just quoted doesn't support what you just wrote. You seem to be confusing "i386" (a particular 32-bit CPU) with "32-bit library" (a library compiled to support 32-bit CPUs but also support everything thereafter).

    5. Re: containers or virtual machines for apps? by Anonymous Coward · · Score: 1

      The thing you just quoted doesn't support what you just wrote. You seem to be confusing "i386" (a particular 32-bit CPU) with "32-bit library" (a library compiled to support 32-bit CPUs but also support everything thereafter).

      I think you misunderstand how multilib support works on Debian based OSes. i386 in this case isn't referring to a specific CPU in this case. i386 refers to the 32-bit x86 architecture. amd64 refers to the 64-bit x86 architecture. multiarch 32-bit libraries are installed by installing the i386 version of the package.

      What I quoted says they're providing the i386 architecture ports of the libraries so that legacy applications can run on the amd64 architecture through 18.04 LTS. For 18.10 on, the 32-bit port of the libraries will no longer be provided.

      On my 64-bit x86 Debian system, both libc6:amd64 and libc6:i386 are installed to provide both the 64-bit and 32-bit version of libc. Without the i386 port, there is no 32-bit library.

  11. Re:That's just great... by LichtSpektren · · Score: 2

    openSUSE dropped IA-32 builds from their brand new releases, and the Ubuntu community is talking about it (nevermind that 16.04 will support their IA-32 build for another five years). It's still supported by CentOS and Debian and lots of other distros. In short, saying you're shit-outta-luck is totally not accurate.

  12. Re:That's just great... by angel'o'sphere · · Score: 1

    Be careful what you say here Dear creimer!

    Now you have three frontiers against you!
    a) the Apple/Mac hater
    b) the Mac/Apple lovers who hate you because you run windows on a Mac!
    c) the Windows haters

    And I believe the Windows 10 haters or Windows 10 upgrade haters are just in the queue behind them!

    Perhaps I should set up a shop with pitchforks close to your house and I finally become rich?

    --
    Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
  13. Re:That's just great... by phantomfive · · Score: 1

    It's hard to believe that testing 32-bit really doubles the testing effort, but whatever.

    --
    "First they came for the slanderers and i said nothing."
  14. Re:That's just great... by Anonymous Coward · · Score: 0

    Uh... as more time goes on, more exploits are found in all software, making all systems gradually more insecure. It's almost like there were a universal law governing such things *cough*.

  15. Re:That's just great... by __aaclcg7560 · · Score: 0

    If Ubuntu 18.10 is 64-bit only, is that a problem?

    For a 32-bit processor, yes.

  16. Re:That's just great... by phantomfive · · Score: 1

    Incidentally here's a list of architectures supported by Suse 12, so:

    aarch64 alphapca56 armv5tel geode ia64 ppc64 ppciseries sh4 sparcv9 alpha amd64 armv6hl i386 noarch ppc64iseries ppcpseries sh4a sparcv9v alphaev5 armv3l armv6l i486 pentium3 ppc64p7 s390 sparc x86_64 alphaev56 armv4b armv7hl i586 pentium4 ppc64pseries s390x sparc64 alphaev6 armv4l armv7l i686 ppc ppc8260 sh sparc64v alphaev67 armv5tejl athlon ia32e ppc32dy4 ppc8560 sh3 sparcv8

    I don't even know what all of those are.

    --
    "First they came for the slanderers and i said nothing."
  17. This is too early. by Anonymous Coward · · Score: 0

    But having no more 32bit machines with Ubuntu is probably a good thing.

  18. Re:That's just great... by LichtSpektren · · Score: 4, Insightful

    Uh... as more time goes on, more exploits are found in all software, making all systems gradually more insecure. It's almost like there were a universal law governing such things *cough*.

    Ubuntu's going to support IA-32 images for at least another five years (EOL for 16.04), probably seven (18.04 EOL). If your IA-32 system is still chugging by then, there'll still likely be Debian and CentOS to switch to.

  19. Re:That's just great... by Sasayaki · · Score: 1

    What show-stopping problem for a 2006 MacBook is present in 18.04 but fixed in 18.10?

    What's wrong with running 18.04 until the hardware dies?

    --
    Check out my sci-fi book "Lacuna" at http://goo.gl/MVxX8
  20. Re:That's just great... by StayFrosty · · Score: 4, Insightful

    Well, in 2018, the 2006 MacBook will be 12 years old. 18.04 is an LTS release and will have 5 years of support and security updates. By the time there are no more security patches, the machine will be 17 years old and software exploits will be the least of the user's concerns if it is still his/her main machine.

    --
    "Frequently wrong, never in doubt."
  21. Re: That's just great... by MightyYar · · Score: 2, Informative

    Gary Johnson? Jill Stein? Why pretend there are only 2 candidates?

    --
    W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
  22. This isn't about new hardware by l2718 · · Score: 4, Insightful

    The problem is not newly-bought consumer electronics or legacy software. The problem is legacy hardware. I'm still using the Thinkpad I bought in 2006 (4:3 aspect ratio display). Luckily it's a 64-bit processor, but others have older 32-bit machines.

    It's also not about the kernel -- Linux itself will support 32-bit architecture for a long while more, and most software will compile correctly on both 32-bit and 64-bit, though it will be less and less true as distributions stop their QA and you are left with only the upstream development team.

    Of course, these old machines are pretty few, so it probably does make sense for Ubuntu to drop 32-bit packages. Other more enthusiast-targeted distributions will probably keep 32-bit support. In particular Gentoo compiles everything locally.

    1. Re:This isn't about new hardware by Hognoxious · · Score: 1

      CERN still have plenty of crusty old hardware. They produce a 32-bit version of Scientific Linux for that reason. With a bit of faffing, you can even get it to run on non-PAE processors.

      --
      Confucius say, "Find worm in apple - bad. Find half a worm - worse."
    2. Re:This isn't about new hardware by serviscope_minor · · Score: 1

      The problem is legacy hardware. I'm still using the Thinkpad I bought in 2006 (4:3 aspect ratio display).

      Bear in mind that if 18.10 goes 64 bit only, then 18.04 LTS will be 32 bit and supported for 5 years. By the time that security updates stop your 2006 laptop will be 17 years old. I'm pretty sure that by 2010, almost no new consumer kit was being sold with 32 bit processors in (that's 4 years after the Core line went 64 bit and 2 years for Atom, where things seemed to have a much faster churn. AMD was all 64 bit earlier).

      If you bought a 32 bit Atom in 2010, chances are it would be a very low end machine, and would be 13 years old by the time updates stop. It would be quite suprising if such an old, cheap machine was robust enough to last that long.

      Nonetheless, there are other distros which will support it for even longer.

      --
      SJW n. One who posts facts.
    3. Re:This isn't about new hardware by FrozenGeek · · Score: 1

      I'd be interested to know what percentage of people running old hardware (you can decide for yourselves what you consider "old" to mean) actually keep the software up to date. Every place I've worked has a very hard time keeping all the software (even the OSs) patched as they ought to be. If patches stop flowing for some of the old stuff, will the owners, or the administrators, actually care?

      --
      linquendum tondere
    4. Re:This isn't about new hardware by Anonymous Coward · · Score: 0

      My desktop is coming up on 10 years old, and I still regularly use a laptop circa 2002 when I'm travelling (though I never connect it to the net - it's purely for programming).

      I care more about the desktop support as it is online and hence needs to be kept relatively up to date patch-wise. The laptop is runs xp and hasn't been patched (or connected to anything) for about 5 years.

    5. Re:This isn't about new hardware by Anonymous Coward · · Score: 0

      This is a bad idea. There are a lot of perfectly good 32-bit computers out there still working. I just sold a pair of IBM T43s to a good friend. I have upgraded to faster machines for gaming. I did however keep those T43s up to date with Linux Mint (KDE) 17.3 and installed updates as they came up in the update manager. These machines are still useful for routine web surfing, email, and everything but high end games. Hopefully there will still be some distributions that will keep doing 32 bit ISOs.

    6. Re:This isn't about new hardware by Anonymous Coward · · Score: 0

      non-pae processors are processors pre pentium pro plus some budget ones that came after so not a big loss and the people running such systems have been around long enough to be able to build from source anyway.

  23. 32-bit != i386 by Anonymous Coward · · Score: 5, Insightful

    Posts like this always confuse me. The terms i386 and 32-bit are not interchangeable. AFAIK, they were only talking about getting rid of i386 architecture (i.e. 20+ year old 32-bit hardware), but would maintain i686 (more recent 32-bit hardware) support.

    1. Re:32-bit != i386 by Anonymous Coward · · Score: 0

      And you are actually wrong...

      i386 support has been gone from glibc since glibc 2.3 came out.
      It has been gone from the kernel since 3.0 came out (may have been broken prior.)
      And I am pretty sure there were rumblings of getting rid of i486 and possible i586 in the recent past from gcc. (i386 AFAIK has been broken there as well for a while, although it still is an alias in -mcpu/march)

      When they talk about dropping i386 support, it is usually meant as an alias for i686 support or 32 bit x86. Yes it is confusing as hell, but this has been synonymous for 2 decades now, even when other more specific arch identifiers were available.

      One of the reasons the unix guys always mocked the linux guys. Because they were never pedantic enough.

    2. Re:32-bit != i386 by bsolar · · Score: 4, Informative
      They are actually discussing about dropping x86_32. This is from the original post which got "resurrected" at the beginning of the thread in their mailing list (the quoted text at the bottom):

      At some point we are going to want drop x86_32 kernel support and just have 32-bit compatibility libraries, but I don't know when that makes sense.

  24. math is hard by Anonymous Coward · · Score: 0

    It's hard to believe that testing 32-bit really doubles the testing effort, but whatever.

    One of the biggest parts of testing consists of keeping test systems up and running and ready to run the tests, every retired system is less labor.

    Virtual machines need updating and maintenance, too, they are not "free" by any stretch, again labor is saved when they are retired.

    Often tests are not "totally bulletproof", with failures from system problems or incorrect setup, much labor is involved here in investigation and/or mitigation, again directly related to the number of systems to test.

    1. Re:math is hard by luis_a_espinal · · Score: 1

      It's hard to believe that testing 32-bit really doubles the testing effort, but whatever.

      One of the biggest parts of testing consists of keeping test systems up and running and ready to run the tests, every retired system is less labor.

      Virtual machines need updating and maintenance, too, they are not "free" by any stretch, again labor is saved when they are retired.

      Often tests are not "totally bulletproof", with failures from system problems or incorrect setup, much labor is involved here in investigation and/or mitigation, again directly related to the number of systems to test.

      This and exactly this. Another thing people forget is that testing not only cover fresh installations, but also upgrades. So the testing permutations begin to fly off the handle with the number of supported platforms combined with the number of upgrade paths an organization needs to test.

      That shit is neither cheap, nor easy.

  25. Re:That's just great... by chipschap · · Score: 2

    My poor old Acer Netbook, 7 years old and going strong, isn't 64 bit, and runs Linux Mint very well. Oh no, what shall I do?

    Really, there will be 32-bit compatibility for a minimum of 5 more years (Mint 18 support cycle) and by then ... maybe the Acer will deserve retirement.

  26. Re:That's just great... by Anonymous Coward · · Score: 0

    That might have been true, except for the gift the Republicans received when Hillary bought the Democratic nomination.

  27. Re: That's just great... by slazzy · · Score: 1

    There are lots of great Linux distros.

    --
    Website Just Down For Me? Find out
  28. Re:That's just great... by Anonymous Coward · · Score: 0

    I guess I'm stuck using Windows 10 (32-bit) on my vintage 2006 MacBook (Intel Duo Core 32-bit processor).

    Given that you're using an ancient processor, I'm not really sure why you would be against using ancient software.

    I was surprised they bothered with 32-bit even back when windows 7 came out, but being a Linux user, I hadn't touched 32 bit for years even at that point. Why Windows 10 has 32-bit support I have no idea. Maybe for people like you that like really ancient, hot, slow processors.

  29. 32 bit cpus are everywhere by Anonymous Coward · · Score: 0

    Embedded, IoT and robotics are awash with 32 bit x86 stuff. And seriously, OBS load is the problem worth considering dropping x86?

    1. Re:32 bit cpus are everywhere by Merk42 · · Score: 2

      Embedded, IoT and robotics are awash with 32 bit x86 stuff.

      How many of those run Ubuntu or OpenSUSE?

  30. Re:That's just great... by __aaclcg7560 · · Score: 1

    What's wrong with running 18.04 until the hardware dies?

    I've never got Ubuntu to install successfully on the MacBook. Mint Linux install fine.

  31. GEEK POLICE RAID by Anonymous Coward · · Score: 1

    I guess I'm stuck using Windows 10 (32-bit) on my vintage 2006 MacBook (Intel Duo Core 32-bit processor).

    You have been found GUILTY of NON-GEEK THINKING!

    Turn in your geek card and proceed to geek jail! You are here by sentenced to 3 years first level technical support ("Have you tried turning it on and off?").

    Judges ruling:

    As we see, a geek would just compile his own kernel to his machine and wouldn't need to worry about if one already existed compiled for his machine. The defendant won't or doesn't know about or doesn't know how. Three years of low level technical support should re-educate him. Parole is available after one year if he compiles and runs the latest Linux Distro of his choice on his machine.

    Court adjourned.

    1. Re:GEEK POLICE RAID by __aaclcg7560 · · Score: 1

      As we see, a geek would just compile his own kernel to his machine and wouldn't need to worry about if one already existed compiled for his machine.

      I haven't compiled a kernel since the 1990's.

    2. Re:GEEK POLICE RAID by Hognoxious · · Score: 4, Funny

      I haven't compiled a kernel since the 1990's.

      Is that when you started or when it finished? I'm asking because it might have been Gentoo.

      --
      Confucius say, "Find worm in apple - bad. Find half a worm - worse."
    3. Re:GEEK POLICE RAID by Rob+Riggs · · Score: 1

      My kingdom for a mod point.

      --
      the growth in cynicism and rebellion has not been without cause
    4. Re:GEEK POLICE RAID by __aaclcg7560 · · Score: 2

      Is that when you started or when it finished? I'm asking because it might have been Gentoo.

      I started in 1997 with Debian in a book distro. Ran SUSE for many years. Used Ubuntu, Fedora or Mint at work, depending on whatever was popular with the engineers. When I worked at Google, I used Goobuntu (Ubuntu variation). These days I'm banging on Red Hat Linux to see if I want to go for the certification.

    5. Re:GEEK POLICE RAID by dbIII · · Score: 1

      Compiling kernel modules on the other hand happens with things like running the Nvidia installer.

    6. Re:GEEK POLICE RAID by Ed+Avis · · Score: 2

      I remember when a kernel build was an all-night job (on a machine with four megs of RAM).

      --
      -- Ed Avis ed@membled.com
  32. An Opportunity to Try Something New! by Anonymous Coward · · Score: 0

    The old laptop crowd should consider this as an opportunity to try something new - BSD!

    http://www.openbsd.org/i386.html
    https://www.freebsd.org/platforms/i386.html
    http://wiki.netbsd.org/ports/i386/

  33. Re:That's just great... by KiloByte · · Score: 3, Insightful

    Then use Debian. While discouraging i386 as default download is long overdue, judging from other old architectures, it'll be a long long time until i386 is retired from the first class arch set, and even then it'll be welcome in second class (AKA debian-ports), among stuff like m68k, alpha or sh4.

    Or, use Debian-hurd. It's available for i386 only!

    --
    The creatures outside looked from Alt-Right to Antifa; but already it was impossible to say which was which.
  34. Re:That's just great... by JackieBrown · · Score: 1

    Well, at least you can look forward to our next president not being Donald Trump.

    Are you kidding? Donald Trump has no money, no campaign staff in the battleground states, and the electoral map is stacked against him. The Republicans can kiss the White House, Senate, and maybe the House, goodbye.

    First off - you are agreeing with the poster above.
    Second, it's funny how the democrats complain about money in politics yet they are the biggest recipients from unions, businesses, and now international entities.

  35. Will create problems by Eravnrekaree · · Score: 4, Insightful

    So much for Linux being "great for old hardware". This is really just an dubious move by distros and really just ignores a huge area where Linux can see use: Old hardware where Windows wont run. You also have another aspect of this which is your basically trashing 32 bit app support if you do not include 32 bit libraries, or, providing a thunk between 32 bit apps and 64 bit libraries.

    Even if 32 bit libraries are not built, you should be able to run a 32 bit app by compiling the libraries yourself, so distros could at least allow people to build 32 bit libraries easily from source packages, (with the benefit of automatically building all dependancies).

    Another area this will create problems is with VMs on even recent hardware, Intel chips up to just a year or two ago didnt include VT-x or a Ring 2, which means that virtualization of 64 bit OSs will not work.

    1. Re:Will create problems by Groo+Wanderer · · Score: 3, Insightful

      You might want to think about what you just said, or read the blurb of an article you are commenting on. It specifically states "Major Linux distributions" which are not what tend to support ancient, embedded, long life, or related non-consumer/non-traditional server workloads. In short there are tons, hundreds likely of distros that will cater to 32-bit and even 8/16-bit hardware because that is all that is needed for the job they do.

      Go look at Linaro's work, it isn't technically a distro but is supports some pretty 'craptacular' hardware, at least by modern user perspectives. How long do you think your router can live with 'only' 32b SoCs? Do you think DDWRT will get a massive boost from 64b code? How about your dishwasher? There are distros that cater to all those markets and they are not moving to 64-bit only.

      In short nothing will change for 99.(big number) of users, those that need 8/16/32b code will still have distros to do it. Anyone wanting to run those distros as a modern desktop or server, well, enjoy it but I am not a masochist so I won't be joining you. For every one else, carry on, you won't notice anything but better wares and cheaper devices.

    2. Re:Will create problems by Anonymous Coward · · Score: 0

      The last released 32 bit only CPU was the Pentium D, released just over ten years ago. If you need to support hardware that old, save your money and buy new hardware. The electricity draw on that old hardware was high enough that even new (cheap end) hardware pays for itself in the first year or two of lower electricity bills.

    3. Re:Will create problems by TechnoJoe · · Score: 0

      "Great for old hardware" is a myth at best, and sometimes a farce.

      Myth example: I had an old Pentium III 650 Mhz machine with 256 MB of ram. It was consuming more power than my AMD Athlon 2.0 Ghz with dual core and 4 GB of ram. I have the same story with a bunch of old IDE 80 GB drives. I could string together 12 of them in a JOBD to get 1 TB, or I could just buy a 1 TB drive. At some point you have diminishing returns on that old hardware.

      Farce example: GCC was generating the CMOV instruction for 486, even though cmov was first invented on 586. Obviously, no one was using that old hardware, or their programs would have all crashed. (I think this is the thread. https://sourceware.org/ml/libc-alpha/2004-09/msg00157.html)

    4. Re:Will create problems by toddestan · · Score: 1

      Actually, the Pentium D supported EMT64, as well as some of the later LGA775 Pentium 4's. I'm guessing the last 32-bit x86 CPU from Intel was one of the Atoms. Otherwise, it's probably the Core Duo/Core Solo lines which were the mobile CPUs out about the same as the Pentium D.

  36. Re:That's just great... by angel'o'sphere · · Score: 4, Funny

    I don't even know what all of those are.
    That is easy to explain, phantomfive:
    All this are little bugs, usually black, with lots of golden legs sitting on the motherboard.

    Some of them even have arms as you can see, some are sparcling and the others are of greek origin, like the pentium but I wonder where the monon, and duon etc. are. I guess the alpha is also a greek and the i686 has an extraordinary high IQ, or had ... I believe he is out of service now.

    Regarding the 'motherboard' ... I'm not sure if you are old enough to get explained that yet. Considering that there are sometimes daughterboards sitting in strange positions on top of the motherboard ...

    Hope that helped!

    P.S. the Athlon is a long forgotten Spartan athlete. He always wanted to participate in Pankration at the Olympic games. But alas, Spartians were prohibited to participate in that discipline. (Something with killer instinct or something) So he finally decided to dress like a girl. Seems he did not know that women folk is not allowed at Olympic games either. Sad fate. Really sad.

    --
    Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
  37. Re:That's just great... by __aaclcg7560 · · Score: 1

    That might have been true, except for the gift the Republicans received when Hillary bought the Democratic nomination.

    The only thing that changed since Hillary won the Democratic nomination is Trump's sinking poll numbers, especially in the battleground states.

  38. That doesn't mean squat. by Ungrounded+Lightning · · Score: 2

    At least for intel Archs, you can install a 32 Bit OS on a computer with a 64 bit capable cpu.

    Which doesn't mean squat. We're talking Q.A. here.

    The goal is to determine whether the code will work on a real 32-bit architecture, not a 64-bit architecture running in 32-bit emulation mode. The two have differences. If you run the tests on something other than the real target you have no clue whether it will work on the real thing.

    --
    Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
    1. Re:That doesn't mean squat. by TemporalBeing · · Score: 1

      At least for intel Archs, you can install a 32 Bit OS on a computer with a 64 bit capable cpu.

      Which doesn't mean squat. We're talking Q.A. here.

      The goal is to determine whether the code will work on a real 32-bit architecture, not a 64-bit architecture running in 32-bit emulation mode. The two have differences. If you run the tests on something other than the real target you have no clue whether it will work on the real thing.

      IA64/AMD64 do not emulate IA32, they actually support IA32 natively. It's not an Emulation Mode, it's a defined part of the architecture. So yes, you know very well that they run properly. Furthermore, virtual environments for IA32 have been so thoroughly tested that they can be very well relied on for this kind of general testing.

      Now, if you want to do some testing of extremely specific hardware devices, that could be a different issue; but the general case of does an image work shouldn't be a problem.

      --
      Truth is like the sun. You can shut it out for a time, but it ain't goin' away. - Elvis Presley (source: imdb.com)
    2. Re:That doesn't mean squat. by tarpitcod · · Score: 1

      One thing I've always wanted to do was compile up with a really super-duper compiler an emulator for x32 on an IA64 first generation box and compare the performance to the built in emulation which was far from EPIC.

    3. Re:That doesn't mean squat. by F.Ultra · · Score: 1

      While AMD64 does not emulate the i386 as GP so erroneously claimed the problem is that the AMD64 has no i386 mode, it simply supports more registers and op codes so if your i386 build for any reason happen to include any op code from the AMD64 set then it will work on an AMD64 machine but not on a real i386 machine. How VM's do it, i.e if they also support the full AMD64 code set even if you set it to i386 mode, I don't know. If they don't want to keep two code bases they might allow AMD64 codes in i386 mode.

    4. Re:That doesn't mean squat. by TemporalBeing · · Score: 1

      One thing I've always wanted to do was compile up with a really super-duper compiler an emulator for x32 on an IA64 first generation box and compare the performance to the built in emulation which was far from EPIC.

      So Bochs Pentium Emulator (bochs.sf.net)? It's performance isn't the best either but it emulates the *entire* computer, but it's decently usable though.

      --
      Truth is like the sun. You can shut it out for a time, but it ain't goin' away. - Elvis Presley (source: imdb.com)
    5. Re:That doesn't mean squat. by TemporalBeing · · Score: 1

      While AMD64 does not emulate the i386 as GP so erroneously claimed the problem is that the AMD64 has no i386 mode, it simply supports more registers and op codes so if your i386 build for any reason happen to include any op code from the AMD64 set then it will work on an AMD64 machine but not on a real i386 machine. How VM's do it, i.e if they also support the full AMD64 code set even if you set it to i386 mode, I don't know. If they don't want to keep two code bases they might allow AMD64 codes in i386 mode.

      The AMD64 Architecture certainly does support 16/32-bit Real Mode and 16/32-bit Protected Mode with the same meanings and register usages as the IA-32 chipset. The additional registers are not available unless you're in Long Mode (64-bit mode).

      Checking AMD x86-64 Architecture Programmer's Manual Vol 1 section 3.1 (my copy is rev 3.07 from 2002) - yes the new 64-bit registers are available in name, but only in so far as they overlap their IA32 named values (e.g rAX -> EAX -> AX -> AL). But that's more a matter of ensuring that your compiler produces the correct values, and is easily enough to have the assembly searched for to validate.

      --
      Truth is like the sun. You can shut it out for a time, but it ain't goin' away. - Elvis Presley (source: imdb.com)
  39. Re:That's just great... by Anonymous Coward · · Score: 1

    It's hard to believe that testing 32-bit really doubles the testing effort, but whatever.

    So, are you volunteering for this effort? If one is actually doing proper testing (rather than "did it compile"?) every test has to be done again on each additional platform so that the release actually is verified as working. Sure, most of the time it just works, but there are always edge cases.

  40. Didn't Bill Gates say? by Anonymous Coward · · Score: 0

    20 bits ought to be enough for anybody!

    2^20 = 1MB address space, yielding 640KB of RAM.

    and yes, /me knows he probably didn't say it, and/or didn't mean it the way it was interpreted.

    1. Re:Didn't Bill Gates say? by Anonymous Coward · · Score: 0

      It's probably not funny when I have to explain it.

  41. Re: That's just great... by WarJolt · · Score: 1

    Because the founding fathers dream of not having a partisan democracy is dead.

  42. Does it really matter? by Anonymous Coward · · Score: 0

    You know, you can head over to newegg or wherever, and price together a very cheap low end machine (like whatever the cheapest AMD APU is) that would still run circles around a 2006 era Core Duo (last 32-bit only that I can think of). There really is no reason to keep running one of those old processors. Even the lowest of the low end new laptops would be better than a Macbook from 2006.
    If you're about to complain about the cost, keep in mind that even internet access and power requirements for an ancient machine would quickly cover the cost of a low end replacement.
    No, I think it's time to move on.

  43. Given that it's Linux by 93+Escort+Wagon · · Score: 1

    I don't see the problem here. While it may be tedious, it's not technically difficult to build your own 32-bit packages from source, at least most of the time - the (64-bit) package maintainers have done most of the hard work already, identifying dependencies and whatnot. You may occasionally have to do some troubleshooting, especially as time goes forward - but is this sort of thing where the Linux community really shines.

    Heck, you might even learn something.

    --
    #DeleteChrome
    1. Re:Given that it's Linux by guppysap13 · · Score: 4, Interesting

      This is certainly a place where *nix excels. I've started mucking around with an old Powerbook G4 because it's easier to carry around than my main workhorse. Debian, Gentoo, and FreeBSD all run on it happily even though it's hard to find new hardware to test on. Gentoo and FreeBSD treat ppc32 as a "second tier" platform - they'll still auto-generate the installers and configure package dependencies, but they won't check for errors during the build, and bugs in ppc32 won't delay a new release. It's up to users to submit bug reports/patches or fix issues as they come up. Transitioning i386 to this level of support is far from the end of the world.

  44. Re:That's just great... by __aaclcg7560 · · Score: 2

    Second, it's funny how the democrats complain about money in politics yet they are the biggest recipients from unions, businesses, and now international entities.

    Funny how much Republican money is sitting on the sidelines because Trump is the nominee.

  45. Just use gentoo, or devuan... by Anonymous Coward · · Score: 0

    Devuan's only support ATM is x86,x86_64, arm, and aarch64. Support them and make clear that 32 bit support is important to you.

    I'm sure some wannabe developers will start making 64 bit assumptions in their C/C++ code, but other than that there is no reason you can't keep using it. I still have lots of systems running x86 code (many predating x86_64, and a number of others without enough memory capacity to validate a switch to x86_64 userspace.)

    Furthermore, if you don't need binary compatibility, you should look into musl-libc.org, it is a complete reimplementation of the linux libc with better features, a shorter compile time, and support back to i486 (with the ability to compile simple c programs using compilers as old as gcc 2.95.3 and 2.7.x, although both need patches for the musl trinary to work with autoconf and company.) The only missing features in it are sunrpc and some of the old charset support (it is iso9660/utf-8 clean, latter assumed to provide most coverage for today's systems.) There are a number of musl based distros now, including a gentoo profile that can build a system using musl (individual ebuilds may fail obviously, but the support is there for the majority of the system, gcc-6.1.0 supports musl out of the box, and llvm/clang/libc++ can provide an alternative toolchain, although gentoo has too many incumbent gcc-reliant ebuilds to move to a clang-only system so far (although you should be able to do it manually with CC/CXX/CFLAGS/CXXFLAGS/LDFLAGS, as I have in the past.)

    The popular linux distros have already lost the faith of many of us with systemd and other questionable development decisions. One more disappointing choice like this doesn't matter in the big picture because it has been time to migrate for a while, they have just taken the choice of putting the migration off for another version out of your hands if you still run x86 hardware.

  46. Re:That's just great... by damnbunni · · Score: 1

    There are actually several new and new-ish 32 bit embedded x86 processors. Or systems where it makes more sense to use 32 bit Windows instead of 64 bit. Low-cost tablets, for instance, often use Atom chips that don't even support more than one or two gigs of RAM. They run Win10 okay, and browsing and casual games are fine on 'em, but a 64 bit OS would be a waste.

  47. Please don't kill 32-bit Wine by allquixotic · · Score: 5, Interesting

    We at least need enough 32-bit packages available in the 64-bit distro (whether by dpkg --add-architecture i386, or by installing "lib32" packages like we used to do) to install and run Wine.

    You see, to run Win32 programs, your Wine emulator binary needs to be a 32-bit Linux/ELF application. I suppose it could emulate cross-architecture, but wine prides itself on *not* emulating native code generation (for performance). Otherwise it would be as slow as a software virtualization solution like Bochs or (non-KVM) qemu.

    Wine, in turn, depends on a number of system libraries for core services. It then implements common Windows APIs "in terms of" available platform libraries. Direct3D in terms of OpenGL; DirectSound in terms of libasound2 or libpulse; etc. These libraries, linked into a 32-bit binary, must also be 32-bit.

    I agree that there's no point in testing 32-bit *hardware* any longer, but I hope they continue to ship 32-bit *builds* (even if they stop making 32-bit installation CDs). There's just too much software on the Win32 platform that needs to run on Linux (desktop OR server; see game servers) to abandon this segment of the market.

    1. Re:Please don't kill 32-bit Wine by LichtSpektren · · Score: 4, Informative

      Read TFA. Nobody's killing 32-bit libraries. Only .ISOs for 32-bit CPUs.

    2. Re:Please don't kill 32-bit Wine by waveclaw · · Score: 1

      16-bit Windows software can be run through Wine. Linux has never had a 16-bit implementation.

      Some business software is run through Wine. But it is heavily used for Windows games on Linux Mostly just 32-bit Blizzard titles and a few 32-bit or 64-bit MMOs like Eve Online.

      For these applications Windows-on-Windows (WoW) is something Wine should handle. Wow is a subsystem specific to Windows. Both the 16-bit and 32-bit versions. Thunks to Linux 32-bit compat libraries may not always be appropriate when WoW behavior is expected.

      64-bit Wine prefixes are considered experimental. But I would expect them to be very common. Ubuntu, MagiOS, openSUSE and Fedora provide it. Gamers playing on Linux likely will be using it the way from their distribution built it. That will be on 64-bit if their OS is 64-bit. On the flip side, competitive gamers looking for as much performance as possible are likely to try every combination to eke out that extra few fps. I have met people who dual box Windows and Linux for extra FPS on Linux when possible.

      At the worst, 32-bit compat libraries will have to remain around. Wine can use those instead of providing internal narrowing support. The compat libraries are needed anyway for closed-source applications. Things like Oracle products, random indie developer apps and any number of long gone companies that farted out a single Linux edition in the 90s.

      --

      "You cannot have a General Will unless you have shared experiences. You cannot be fair to people you don't know."
    3. Re:Please don't kill 32-bit Wine by fgouget · · Score: 4, Informative

      Read TFA. Nobody's killing 32-bit libraries. Only .ISOs for 32-bit CPUs.

      The fine article says

      His proposal is that Ubuntu version 18.10 would be 64-bit-only, and if users desperately need to run 32-bit legacy applications, the'll have to do so in containers or virtual machines.

      This suggests it's not just the ISOs that they plan to get rid of but also support for 32-bit applications, which includes Wine (for running 32-bit Windows applications). So yes, that's pretty worrying for Wine as a lot of Windows applications are either still 32-bit only, or depend on a 32-bit installer. Furthermore, one of the great advantages of Wine is that you do away with all the annoyance that are VMs. So using "containers or virtual machines" is really not much of a solution.

    4. Re:Please don't kill 32-bit Wine by stevied · · Score: 2

      To run 32 bit apps in a container, you'd still need 32 bit libs. So it implies some support would be kept..

    5. Re:Please don't kill 32-bit Wine by xvan · · Score: 1

      Is that possible today? to chroot/ lxc a x86_64 kernel to a 32 bit container?

    6. Re:Please don't kill 32-bit Wine by Anonymous Coward · · Score: 0

      To run 32 bit apps in a container, you'd still need 32 bit libs. So it implies some support would be kept..

      The 32-bit libraries would be in the container. The container could have the 32-bit libs from 18.04 or even compiled by the distributor of the container. It doesn't imply they will continue compiling and updating the libraries for 32-bit.

    7. Re:Please don't kill 32-bit Wine by Anonymous Coward · · Score: 0

      Yep. You can run a full 32-bit userland chroot with an x86_64 kernel. You can also run 32-bit docker containers on x86_64.

    8. Re:Please don't kill 32-bit Wine by allquixotic · · Score: 1

      Unless those libraries are shipped by Ubuntu, though, you'd have to either use a prior release or run a non-Ubuntu OS in your container in order to handle this. A lot of people would like to use the versions of libs shipped by Ubuntu on their 64-bit system, but compiled for 32-bit, to do their 32-bit work (e.g. Wine).

      So yeah, it would do a lot of harm to many use cases and workloads if they stopped providing 32-bit libraries at the very least. If they want to drop 32-bit kernel support, it would affect way fewer people, because there aren't many systems still in use today that only run x86 and not x86_64.

  48. Re:That's just great... by Anonymous Coward · · Score: 0

    I have a Compaq Preesario V6000 released in 2006, which is also a 32bit medium range machine, designed for "Windows xp".

    It runs Windows 10 reasonably, with only a hiccup or two now and then (needs more RAM). Most websites run perfectly fine on it, and can be used for banking, browsing (mostly, it's a bit choppy on HTML5 video - Flash video seems to have fewer problems). I would imagine if it was a high-end laptop, 17 years old shouldn't be an issue.

    So why would someone who just does that stop using it?

  49. Re:That's just great... by Hognoxious · · Score: 4, Funny

    I bet he's a systemd fan too, the bastard.

    --
    Confucius say, "Find worm in apple - bad. Find half a worm - worse."
  50. Re: That's just great... by WarJolt · · Score: 1

    Democrats like to obstreperously disagree. That way they can say a lot without making a point or actually having an idea that isn't nearly identical to the Republicans.

  51. Have they heard of Virtual Machines? by mi · · Score: 3, Insightful

    do you know how hard it is to find 32-bit hardware these days?

    What's there to "find"? You can kick-off a 32-bit VM under any hypervisor — both on the cloud or on your own desktop. You can automate the VM-creation and tear-down on your build-farm quite easily.

    I too strongly prefer to have a system, where size_t is equal to off_t (so you could mmap an entire file and not worry about it), but that is not "free". 64-bit pointers are, obviously, twice-wider than 32-bit ones, so "hairy" structures — with lots of pointers in them — nearly double in size. If none of your processes require more than 4Gb of virtual memory, there is no reason — other than the developers' laziness — to go 64-bit.

    Whether it is an OS embedded inside a router or a point-of-sale machine, or even a single-user web-and-email desktop, 32-bit is perfectly sufficient and the overhead of 64-bit not justified.

    And that laziness is what is keeping us back... Over the last 18 years, according to Moore's law, our computers have become at least 2^12 times more powerful. Now ask yourself, is the user-experience — however you choose to measure it — 4096 times better than it was in 1998? And, if it is not, where did the gains in hardware go?

    By refusing to setup/use tens or even hundreds of 32-bit test-systems, developers force thousands and millions of users to upgrade. That is not a fair trade-off.

    --
    In Soviet Washington the swamp drains you.
    1. Re:Have they heard of Virtual Machines? by Burdell · · Score: 4, Informative

      If none of your processes require more than 4Gb of virtual memory, there is no reason â" other than the developers' laziness â" to go 64-bit.

      First, addresses/pointers aren't normally the largest chunk of code or data memory usage, so the include in RAM usage is far less than double.

      Also, in the specific case of the Intel x86 architecture (which is what this is about, not general 32 bits vs. 64 bits), there is a significant reason to move from i386 to x86_64. The i386 architecture has a very small CPU register set, compared to most modern CPU architectures (and some instructions can only use certain registers). That means lots more things require memory loads/stores, which is bad for performance. When AMD created x86_64, they added a bunch of registers (and got rid of most of the usage restrictions), so 64 bit code performance is better.

    2. Re:Have they heard of Virtual Machines? by somenickname · · Score: 1

      I agree with you to a certain point but, how many people are running a 32-bit OS for legitimate reasons? Some, without a doubt. But probably not enough to justify the effort of keeping it around. I understand the argument of 64-bit pointer size increasing memory usage but, when running in 64-bit mode, the CPU has access to WAY more registers, things like SSE are implied, etc. It's a performance to memory usage trade off. Memory has been abundant and cheap for years and even lowly Atom chips have been 64-bit for quite some time.

      Now, if the kernel decided to drop support for 32-bit builds, that would be pretty crazy. But, having the major distros drop it isn't that big of a deal. If you have hardware/software that legitimately *needs* a 32-bit distro, there will be options available for quite some time. In fact, Ubuntu 16.04 is supported until 2021. That's a pretty large runway to get your poop in a group.

    3. Re:Have they heard of Virtual Machines? by Anonymous Coward · · Score: 2, Interesting


      When AMD created x86_64, they added a bunch of registers (and got rid of most of the usage restrictions), so 64 bit code performance is better.

      All true. But offset by the decreased performance of dealing with 64 bit pointers, it's mostly a wash. This is actually why Linus created the x32ABI https://en.wikipedia.org/wiki/X32_ABI to get the best of both worlds. Largely I think it's forgotten though.

    4. Re:Have they heard of Virtual Machines? by Anonymous Coward · · Score: 0


      Whether it is an OS embedded inside a router or a point-of-sale machine, or even a single-user web-and-email desktop, 32-bit is perfectly sufficient and the overhead of 64-bit not justified.

      Unless you happen to want to run your embedded machine past 2038, which is creeping up.

      In 2016, I'd really think twice about any embedded solution that runs 32 bit Linux. That POS machine might still be running in 22 years.

    5. Re:Have they heard of Virtual Machines? by mi · · Score: 1, Informative

      First, addresses/pointers aren't normally the largest chunk of code or data memory usage, so the include in RAM usage is far less than double.

      Define "normally"... On a busy web-server, for example, the number of simultaneous connections is limited by the maximum number of sockets and files, that can be opened and processes or threads, that can run. All of those structures have lots of pointers... And even where they don't, the alignment optimizations often pad them, leading to (much) bigger sizes. For example, right now on my two FreeBSD-boxes:

      sizeof(FILE) i386: 236, amd64: 312 sizeof(struct socket) i386: 424, amd64: 696 sizeof(struct proc) i386: 776, amd64: 1272

      So, if you use Apache, which uses a process (or even a thread) per active connection, you need about 50% as much memory for the same number of simultaneous connections.

      Something similar can be said for Firefox and any other user application.

      When AMD created x86_64, they added a bunch of registers

      Those registers are largely available to 32-bit processes — just specify the right -march to your compiler. Yes, 64-bit mode does give your something — no question about it. Point is, it is not free...

      --
      In Soviet Washington the swamp drains you.
    6. Re:Have they heard of Virtual Machines? by Anonymous Coward · · Score: 0

      It's a completely fair trade-off when, by and large, those users probably aren't their customers.

    7. Re:Have they heard of Virtual Machines? by mi · · Score: 1

      I agree with you to a certain point but, how many people are running a 32-bit OS for legitimate reasons?

      What's "legitimate"? Is it "legitimate" to want to be able to handle 50% more simultaneous TCP-connections to your server or from your desktop, for example? Look at my earlier post for differences in sizes of several vital data-structures.

      Now, is it "legitimate" for developers to "steal" the hardware gains for their own benefit?

      Unless you happen to want to run your embedded machine past 2038

      You can change time_t to int64_t at any time — with the next major release of your OS. You do not need to switch everything to 64-bit just for that.

      --
      In Soviet Washington the swamp drains you.
    8. Re:Have they heard of Virtual Machines? by somenickname · · Score: 1

      What's "legitimate"? Is it "legitimate" to want to be able to handle 50% more simultaneous TCP-connections to your server or from your desktop, for example? Look at my earlier post for differences in sizes of several vital data-structures.

      If pointer size is your primary concern, why not run a 16-bit OS? I say that in jest but, come on... If you are running a 32-bit machine (without PAE), moving from 4GB (well, more likely 3.5-ish gigabytes) to 8GB on 64-bits is going to cost you about $30. I can't even fathom a mission critical system unable to absorb a $30 upgrade.

      I was a holdout on 64-bit stuff as well. With PAE, you get 36-bit pointers in the kernel and that's pretty cool. But, it comes at a performance cost. And these days, the performance cost is a lot worse because it means you are missing out on the performance *benefits* you'd get by moving to 64-bits.

      Now, is it "legitimate" for developers to "steal" the hardware gains for their own benefit?

      I'm not even sure what this means. Who stole anything? And who benefited by this theft? As far as I can tell, distro maintainers have decided that the maintenance of an almost unused flavor of their distro is no longer beneficial. And, it makes sense: Almost all Intel/AMD hardware has been 64-bit for many years. But, again, as I said, even Ubuntu will support 32-bit hardware until 2021. More than a decade after 32-bit Intel/AMD hardware was completely fucking obsolete.

    9. Re:Have they heard of Virtual Machines? by Anonymous Coward · · Score: 0

      By refusing to setup/use tens or even hundreds of 32-bit test-systems, developers force thousands and millions of users to upgrade. That is not a fair trade-off.

      Then maybe those thousands of users can find ways to support 32-bit build/test/ farms, and contribute their time and expertise for validation instead of boo-hoo. There's a reason why the build/test teams are pushing back against 32-bit builds, and it's not out of ignorance.

    10. Re:Have they heard of Virtual Machines? by mi · · Score: 1, Interesting

      If pointer size is your primary concern, why not run a 16-bit OS?

      Because with 16-bit pointers, you are limited to process-sizes of 65Kb only, which is too small. 32 bit gives you enough of process virtual memory.

      I'm not even sure what this means. Who stole anything?

      As I said — Moore's law gave computers, that are 4096 times more powerful today, than what we had 18 years ago. Would you say, user experience is 4096 times better today, than in 1998?

      No, it is not. It is a lot better, but on that scale, however you measure it. Various things continue to suck as they did before: databases and applications still want more memory. In 1996 I would take me three days to rebuild FreeBSD "world" — today it takes 3 hours. A 24-times improvement, instead of 4000-fold.

      Where did the hardware-gains go was my question and "developers ate it" is my answer. Every time somebody makes a choice of "complicated but efficient" vs. "simple but hardware-demanding" towards the latter, the users lose. Every time you choose Python/Java/Go/Scala over C, you sacrifice some hardware of all of your users for your own convenience.

      Almost all Intel/AMD hardware has been 64-bit for many years

      Yes, of course. But you don't have to switch the OS to 64-bit to make use of those hardware capabilities. And even where you do, that may be the user's choice — developers should not be forcing it. Certainly not for the bogus reasons like "we can't find i386 hardware"...

      --
      In Soviet Washington the swamp drains you.
    11. Re:Have they heard of Virtual Machines? by Anonymous Coward · · Score: 1

      There's also the x32 ABI option, where the pointer sizes are maintained at 32bit, but the x64 registers and calling conventions are used. Linux supports this. You get the performance advantages of x64 without the memory usage.

    12. Re:Have they heard of Virtual Machines? by KiloByte · · Score: 1

      You want x32 for that. The actual speed benefit is hardly ever better than several percent, or none at all for many common tasks, though -- it's worth the hassle only for significant memory savings. Memory is cheap, though -- so you want to bother only if you run many containers per machine. Only then it is a big gain.

      --
      The creatures outside looked from Alt-Right to Antifa; but already it was impossible to say which was which.
    13. Re:Have they heard of Virtual Machines? by Anonymous Coward · · Score: 2, Informative

      How the hell did this nonsense get +4 Informative?

      > So, if you use Apache, which uses a process (or even a thread) per active connection, you need about 50% as much memory for the same number of simultaneous connections.

      Um, no. You need an extra 1272-776=496 bytes per connection. The size of a "struct proc" is negligible compared to the total amount of memory each process will use. More generally, memory used for kernel structures is negligible compared to user-space memory consumption.

      The difference between 32-bit and 64-bit pointers is going to vary between substantial (but still nowhere near double) for code which uses pointers extensively (e.g. high-level languages such as Python, PHP, Java, JS, which use pointer-to-object everywhere), and negligible for programs whose memory consumption is dominated by large arrays of primitive types (char, int, float).

    14. Re:Have they heard of Virtual Machines? by Anonymous Coward · · Score: 0

      Another important source of waste is the 64-bit calling convention which requires reservation of room for at least 4 function arguments even the function doesn't use them and a 16 byte alignment. If you plan to use a 64-bit OS, you must plug in about twice as much memory as you would use with a 32-bit OS, otherwise it will be swapping to disk much more, killing performance. For example, comparing a 32-bit VM and a 64-bit VM, the 64-bit version of Firefox uses almost twice as much memory, although when I open a lot of images or a video, the relative difference decreases.

    15. Re:Have they heard of Virtual Machines? by Anonymous Coward · · Score: 0

      Maybe the old hardware has less backdoor issues, and this will help keep people from using large well developed distros on them.

      But then again large distros probably have plenty of back doors in them anyway.

      I hope systems like NetBSD, OpenBSD, and Minix don't stop supporting 32bit (Minix hasn't a choice yet).

    16. Re:Have they heard of Virtual Machines? by Anonymous Coward · · Score: 0

      I have yet to see that increase in performance. So maybe it hasn't been taken advantage of in all development practice yet. Even in really heavy loads I haven't noticed an improvement. However I haven't tried every processor available. In fact with light desktop tasks (low mem) I have actually seen snapper performance out of the 32bit machines, when the speed rating is the same as the 64bit machine. And that is with a sizable difference in FSB. I've often seen this even when it is between single core 32bit and multiple core 64bit machines. Its when large amounts of memory are required that I see the old 32bit machines fail, in comparison. Unless you're accessing an older server, 32bit machines just don't support the memory requirements of a bloated project. I would point out that if the desktop environment itself is graphically demanding, the 32bit machines tend to drag a bit due to lacking graphics power.

    17. Re:Have they heard of Virtual Machines? by david_thornley · · Score: 1

      What's there to "find"? You can kick-off a 32-bit VM under any hypervisor — both on the cloud or on your own desktop. You can automate the VM-creation and tear-down on your build-farm quite easily.

      Which is not testing on the appropriate hardware, which you'd want to do for at least some cases (like the kernel). The idea is to stop supporting 32-bit computers and use VMs instead, which is in line with what you're saying.

      --
      "When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
    18. Re:Have they heard of Virtual Machines? by david_thornley · · Score: 1

      There are limited resources, and the question is whether it's better to spend them to get maximum performance out of old hardware or to take advantage of newer features. As hardware becomes cheaper relative to software, it's overall more economical to use additional hardware to make the software tasks easier, and less economical to insist on supporting the older and rare hardware.

      If you have an old machine, try an old version of whatever Linux distro you like. It still should work.

      --
      "When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
    19. Re:Have they heard of Virtual Machines? by mi · · Score: 1

      There are limited resources

      Are there? We are talking about open source here. Programmers participate in such projects, because they like the process — not get paid (from a limited pool).

      whether it's better to spend them to get maximum performance out of old hardware or to take advantage of newer features

      False. An efficiently-written program will be faster on all hardware — old and new. A poorly-written one may be fast enough on the newer computer, but a well-written one will still beat it.

      As hardware becomes cheaper relative to software, it's overall more economical to use additional hardware

      That formula is only applicable for small audiences of specialized software. Things like desktop GUIs, web-browsers and e-mail clients, editors and word-processors are used by millions of people while the developer base is counted in low hundreds. There is no justification in forcing millions of users to throw out their old machines because a handful of developers can't be bothered to write well.

      try an old version of whatever Linux distro you like. It still should work

      Wow, condescending much? No, it will not work — old versions stop getting security updates. Oh, and I never liked any Linux distro — they tend to be for punks, not for professionals...

      --
      In Soviet Washington the swamp drains you.
    20. Re:Have they heard of Virtual Machines? by david_thornley · · Score: 1

      Are there [limited resources]? We are talking about open source here.

      Are you trying to tell me that there are infinitely many programmers participating in free/open source projects? As long as we have a finite number of such developers, we have limited resources.

      You also seem perfectly happy to make demands that such developers do things your way, without you actually paying them. You also seem to think that GUIs and browsers and the like should run at maximum speed, but in actuality they have to run fast enough. If a user-initiated action can happen within a tenth of a second, there's very little point in speeding it up. There are cases where performance is very important, but you haven't mentioned them, and most personal computers don't hit them very often. It's good if the kernel is efficient, since if it isn't most things can't be, but editors almost never need to be tremendously efficient.

      It would appear that you don't know much about Linux distros either, and if you knew anything you'd know that certain older versions tend to get security updates for a long time.

      --
      "When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
    21. Re:Have they heard of Virtual Machines? by mi · · Score: 1

      As long as we have a finite number of such developers, we have limited resources.

      Developer-numbers aren't the bottleneck...

      You also seem perfectly happy to make demands that such developers do things your way, without you actually paying them.

      I happen to know, how open source projects work better than 99.9% of people on the planet. Payment is not the issue. Open source developers work because it scratches a particular itch — and for the "bragging rights". Design-decisions of the senior developer(s) of the project are what rules it all.

      And the quiche-eaters, who would've been laughed at only 20 years ago, are increasingly making those decisions: yes, 100 million users may each need to buy $30 worth of RAM to use our next version ($3 bln), but we aren't going to spend an extra week coding the new features so that the upgrade does not increase memory-requirements. As I keep repeating in this thread, the bulk of hardware advancements have been spent on convenience of developers, rather than of users. And that's unfortunate...

      You also seem to think that GUIs and browsers and the like should run at maximum speed, but in actuality they have to run fast enough

      That's true, but the point is not entirely about speed. If the GUI desktop or a browser is taking up most of the workstation's RAM, then there is a problem. As I type this, my Firefox is using over 2Gb of RAM — WTF? Thunderbird is another 1.5Gb — and, although the two apps compile from near-identical code base, their authors make no attempt to share the shared libraries. Both and use their own separate libxul.so (73Mb each) for example — look on your own box for confirmation.

      The typical answer from the quiche-eaters — "Memory is cheap, go buy more" — is bogus.

      --
      In Soviet Washington the swamp drains you.
    22. Re:Have they heard of Virtual Machines? by david_thornley · · Score: 1

      Most of us on Slashdot know more about how free/open source projects work better than 99.9% of the population. You are correct in that it's largely a reputation economy, and developers scratch their own itches. (The corporate stuff is not usually for reputation but to scratch itches.) That means they aren't interested in doing exactly what you want, but what they want. Privately run projects focus on the interests of the developers, and things will be done for developer convenience. That's why there's tons of excellent F/OSS development tools of various sorts and much less software for routine business and accounting purposes. (I get off on development tools, myself, so I'm not complaining.) If there was more hack value in making something run more efficiently than in making something generally work better, things would go more according to what you want, but that doesn't seem to be the case.

      I think you're way overestimating the ease of packing new features into software without increasing the footprint. Spending an extra week on several new features is not going to reduce their additional CPU and memory requirements to trivial levels. It's going to take a lot of work of the sort that most people don't consider fun, and which won't get them much rep, and it's going to fail anyway because there's only a certain achievable level of efficiency. As software is improved in functionality, it is going to require more hardware. (If the functionality remains more or less the same, further development can reduce the footprint.) This means that, if you want improved software, you may have to improve your hardware. If you're unable or unwilling to, consider staying at an earlier version of the software.

      This results in more resource use for the user, but usually that isn't too bad. Very few people will suddenly find their RAM is inadequate (it tends to come in large quantities anyway nowadays) given a minor expansion.

      The fact is that hardware has advanced much farther than software. My current desktop is very roughly a million times better than what I started with (a TRS-80), but software has become nowhere near a million times better. Better hardware allows better software to be written, and compensates for the fact that more features is more memory and slower execution.

      --
      "When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
  52. Re:That's just great... by MyFirstNameIsPaul · · Score: 1

    I'm in the same boat with my Dell, but it is 9 years old. I use it for business trips because all I ever have time for is responding to email, maybe open some spreadsheets. Everything else I need I can get from my phone. I like the old Dell because it has a ridiculous battery life since it is just a Core Duo 1.2 GHz with a battery nearly half the laptop's weight.

    --

    I once took an excursion to Reddit, and later HN. Unlimited up/down voting sucks when dealing with a hive-mind.

  53. I think it's wrong, they're killing i386 not i686 by Anonymous Coward · · Score: 1

    i386 is a very old instruction set that nobody uses any more. i686 is the newer ia-32 architecture that started with the Pentium Pro around 1995. I believe that will still be supported. Maybe not by Canonical but who cares about them, they're corporate pinheads anyway.

  54. Re:That's just great... by LichtSpektren · · Score: 1

    I have a Compaq Preesario V6000 released in 2006, which is also a 32bit medium range machine, designed for "Windows xp".

    It runs Windows 10 reasonably, with only a hiccup or two now and then (needs more RAM). Most websites run perfectly fine on it, and can be used for banking, browsing (mostly, it's a bit choppy on HTML5 video - Flash video seems to have fewer problems). I would imagine if it was a high-end laptop, 17 years old shouldn't be an issue.

    So why would someone who just does that stop using it?

    Nobody's telling you to stop using it. The question is if distro maintainers want to do the work to provide full support to computers that are AT LEAST 17 years old. The answer is, the cutting edge distros don't, but distros focused on stability like CentOS probably will.

  55. Re:That's just great... by angel'o'sphere · · Score: 1

    Most definitely!

    --
    Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
  56. Terrible headline. DESKTOP DISTIES letting go by kwerle · · Score: 4, Informative

    There are a bunch of desktop distributions that will no longer do 32 bit builds. Makes sense.

    No effect on kernel or disties for 32 bit systems/embedded/etc.

  57. Unclear ... by Anonymous Coward · · Score: 0

    ... how much benefit there will truly be in dropping support for 32 bit arch. Essentially, people are saying "it's hard, and we don't want to keep doing it." Ok, fair enough, but then why not just say "we're going to stop supporting audio drivers and software, because it's hard and we don't want to keep doing it." Or fill in any other component or function. The fact is, there is clearly quite a number of people still using 32 bit, and apps still support it. As long as there is a need, and available apps, the decision to slice out a large section of the user base is an arbitrary one, at best.
    Maybe if you're a developer working on this project, and you're getting burned out on your job, instead of pushing changes like this that will negatively affect the user base, you should just make some personal life changes. In other words, maybe it's not the software that needs to change, but rather, the people working on it.
    Just a thought...

    1. Re: Unclear ... by Anonymous Coward · · Score: 0

      Oh for fuck's sake, another 7 years of 32 bit support not enough? Probably 10 for Centos? Plus various minor distros probably supporting 32-bit till about 2030?
      Whining retard.

  58. Re:That's just great... by Anonymous Coward · · Score: 0

    Said every QA engineer too. It more than doubles the efffort.

    For bug fixes and changes unrelated to processor type it doubles the testing, generally because some sales executive doesn't understand the concept of risk-based testing, demands "fully tested products" and throws his weight around the C-suite.

    For changes where processor type matters, it legitimately needs testing on both platforms. In this case engineering time is probably more than doubled because it's likely that two different fixes are needed.

    When you consider automated testing, the test needs to be designed and coded to run in both places. When running, the 32 bits test runs will take longer than the 64 bit test runs. So in business terms (time), that will more than double the effort.

  59. Re:That's just great... by StayFrosty · · Score: 2

    No reason to stop. If it does what you need, go ahead. I'm sure source-based distros like Gentoo will still be fine. Distros that focus on long term support like CentOS and Debian will probably still provide a 32-bit distro as well.

    I would also like to point out that your 10-year-old laptop is having trouble now. Add another 7 years to that and you will be compounding those problems dramatically. I don't see any (desktop or laptop) computers around from 1999 that are terribly useful today. Some parts from a 15+ year old PC are hard to find nowadays (unless you have a huge stash or like risking your money on feebay.) Think IDE hard drives and DDR1 or SDRAM.

    --
    "Frequently wrong, never in doubt."
  60. Re:That's just great... by oh_my_080980980 · · Score: 1

    *Sigh* that ASSUMES the exploits where there to begin with. Something does not become less secure just because of its age.

  61. Re: That's just great... by MightyYar · · Score: 1

    But those are two more political parties... It's still partisan.

    --
    W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
  62. Re:That's just great... by K.+S.+Kyosuke · · Score: 1

    Interestingly, Wikipedia says that all the models have 64-bit CPUs.

    --
    Ezekiel 23:20
  63. Re:That's just great... by sensei+moreh · · Score: 1

    My poor old Acer Netbook, 7 years old and going strong, isn't 64 bit, and runs Linux Mint very well.

    My ZG5 is currently running Win XP, Fedora 24 and Ubuntu 16.04 Mate. I'n hesitant to try Windows 10 on an Atom N270 with only 1GB RAM.

    --
    Geology - it's not rocket science; it's rock science
  64. Explanations: by Anonymous Coward · · Score: 5, Informative

    RPM non-binary package:
    noarch
    Used for packages that aren't arch-specific, documentation, configuration, etc.

    Sun/Oracle:
    sparcv9 sparcv9v sparc sparc64 sparc64v sparcv8
    sparc is sparc32 v7 abi (or should be... systems: ipx ss2, etc)
    sparcv8 is v8+ abi (32 bit with some new instructions, ss4/5/10/20)
    sparcv9/sparc64 is sun ultra+ systems (ultra 1 and above with 64 bit processors.)
    sparcv9v is (I assume) Niagra chips and above, containing virtualization/containerization tech.

    Hitachi SuperH:
    sh4 sh4a sh sh3
    Not sure what arch 'sh' is (sh2?3?)
    sh3/4 were used in some routers and I think the Sega Saturn/Dreamcast/Naomi 1/2 consoles/arcade boards. Have never actually seen one outside a console in real life, although in japan at least there were apparently some routers using them as embedded processors running linux.

    ARM:
    aarch64 armv5tel armv6hl armv3l armv6l armv4b armv7hl armv4l armv7l armv5tejl
    aarch64 is the 64 bit arm extensions. The rest are arm versions from 3 to 7 with different option flags. I think wikipedia has a comprehensive article on what they all mean. Lots of possible binary incompatibilities with arm binaries if you don't build to a specific instruction subset, which almost nobody ever did. v5->7 should I believe be forwards compatible. v6 and 7 definitely are.

    Alpha:
    alphapca56 alpha alphaev5 alphaev56 alphaev6 alphaev67
    DEC's legacy shat all over by Compaq+HP. Better chip than IA64 if they had only produced them on newer processes and provided PC-prices entry level systems to keep developers engaged. China is supposedly producing homegrown knockoffs of these chips for fpu calculations in one of their supercomputers.

    x86/x86_64:
    geode amd64 i386 i486 pentium3 x86_64 i586 pentium4 i686 athlon ia32e
    geode is amd's embedded x86, used in the original released OLPCs and various other embedded systems and devices. Not sure the exact x86 arch it is equivalent to (486->586, and maybe newer arch features)
    ia32e I think is the x32 or x86_64 using 32 bit pointer ABI which allows the register file of x86_64 but only using 32 bit pointer references to keep memory usage low for applications that don't require more than 4 gigs of ram.
    amd64 == x86_64
    And the rest of those are Intel/AMD designations up to to SSE2 (P4) Not sure why they have that many different versions.

    Itanic:
    ia64 - Nuff said

    IBM/Others PowerPC:
    ppc64 ppciseries ppc64iseries ppcpseries ppc64p7 ppc64pseries ppc ppc8260 ppc32dy4 ppc8560
    32 and 64 bit variants of PowerPC, dating from 90s era Macs to today.

    IBM S390:
    s390 s390x
    Some sort of mainframe/large workstation systems I think. Not sure if the supported models are all PPC derived or not. I believe they run a different microcode layer on top of the cpus intended for mainframe use. Also run a hypervisor(or equivalent) above linux (and predating hypervisor capabilities in x86 by many years.)

    Hope that helps! Check wikipedia for further info. They have rather comprehensive articles on all of these!

    1. Re:Explanations: by PhotoJim · · Score: 1

      Geode:

          processor : 0
          vendor_id : AuthenticAMD
          cpu family : 5
          model : 10
          model name : Geode(TM) Integrated Processor by AMD PCS
          stepping : 2
          microcode : 0x8b
          cpu MHz : 498.048
          cache size : 128 KB
          fdiv_bug : no
          f00f_bug : no
          coma_bug : no
          fpu : yes
          fpu_exception : yes
          cpuid level : 1
          wp : yes
          flags : fpu de pse tsc msr cx8 sep pge cmov clflush mmx mmxext 3dnowext 3dnow vmmcall
          bogomips : 996.09
          clflush size : 32
          cache_alignment : 32
          address sizes : 32 bits physical, 32 bits virtual
          power management:

          Linux fqdn.domain.tld 3.16.0-4-586 #1 Debian 3.16.7-ckt11-1+deb8u4 (2015-09-19) i586 GNU/Linux

      This is from an Alix 2D3 which is (but not for much longer) my router.

    2. Re:Explanations: by MachineShedFred · · Score: 1

      PowerPC stuff:
      ppc64 would be the generic 64-bit PowerPC chips, such as PowerPC 750 used in PowerMac G5
      ppciseries would be the IBM PowerPC-based iSeries servers, which are AS/400 and replacements.
      ppcpseries would be the IBM PowerPC-based P-series servers, which are the extension of the old F40 / F60 PowerPC based stuff running AIX.
      ppc64p7 is likely POWER7-based systems
      ppc would likely be 32-bit PowerPC, so PowerPC601, 603, 604, 620 (G3), 640(G4)
      not sure on 8260, 32dy4, and 8560.

      Yes, the S390 is for IBM mainframes not running OS/390.

      --
      Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
    3. Re:Explanations: by kjs3 · · Score: 1
      > sparcv9v is (I assume) Niagra chips and above, containing virtualization/containerization tech.

      I thought v9v was v9 + VIS instruction set extensions.

      > Hitachi SuperH

      SuperH gets used a lot in embedded, mobile and automotive applications. The SH3 & SH4 are really quite powerful 32-bit processors with MMUs, and I've always found the SH4 4x4 vector instruction set nifty (and amazingly fast if your problem can fit in that box). The biggest reason to continue SuperH support in Linux is the fact that the patents on the chip have expired and there is an active effort to produce an open source implementation (see: https://lwn.net/Articles/64763...). They currently have SH2 level functionality and are aiming at SH4.

      The SH5 was a 64-bit extension of SuperH. It apparently shipped but never made an impact on the market and quickly disappeared, but it does provide a roadmap to 64-bit once a working open source SH4 is solid.

      > s390 s390x Some sort of mainframe/large workstation systems I think

      Mainframes. Descendants of the IBM/370 and currently call "z/Series" or some such. Latest one is the "z13" line. Processor not derived from PPC or Power, though there is technology overlap. Can run Linux on bare metal or under the z/VM hypervisor.

    4. Re:Explanations: by kjs3 · · Score: 1
      > ppc64p7 is likely POWER7-based systems

      POWER 7 & 8 I think.

      > not sure on 8260, 32dy4, and 8560

      Those are PPC-based embedded processors.

  65. Re:That's just great... by Flavianoep · · Score: 1

    Well, in 2018, the 2006 MacBook will be 12 years old. 18.04 is an LTS release and will have 5 years of support and security updates. By the time there are no more security patches, the machine will be 17 years old and software exploits will be the least of the user's concerns if it is still his/her main machine.

    I have a 2009 32 bit processor Sony Vaio P for which the best option I found so far is Linux Mint, which is based on Kubuntu. The form fator of this model was abandoned long ago, so I have no appropriate alternative in size and weight (no tablet is not good enough). By the time the aforementioned MakBook be 17 years old, my Sony Vaio P will be 14, yet I have to admit that my least concern will be software exploits.

    --
    Linux is for people who don't mind RTFM.
  66. Re: That's just great... by Anonymous Coward · · Score: 0

    I think you have a partial valid point here about the "true" amount of effort.

    Anyone here seen the OpenBSD rack of all architectures. (Hint: it's on the main page), on real bare metal hardware like an old SPARC 20 pizzabox? I don't hear them bitching about this problem, and I can still run the recent OS on surprisingly very old hardware at ridiculously good performances. Granted my use case deployments are different because it's the right tool for my particular need, but that's not the point.

    So I'm not saying that it does involve SOME additional amount of effort, but I think the amount claimed may be exaggerated.

  67. Re:That's just great... by reboot246 · · Score: 1

    I think that we can kiss the country goodbye. A Hillary Presidency will be the end of the republic. It's on its last legs right now and cannot survive Bill's wife being President.

    Well, it's been a great 200+ years while it lasted.

  68. Re:That's just great... by chipschap · · Score: 1

    My poor old Acer Netbook, 7 years old and going strong, isn't 64 bit, and runs Linux Mint very well.

    My ZG5 is currently running Win XP, Fedora 24 and Ubuntu 16.04 Mate. I'n hesitant to try Windows 10 on an Atom N270 with only 1GB RAM.

    I bought the "upgrade" for my Acer Netbook, giving me a fantastic 2 GB of RAM --- but actually that makes streaming video, running LibreOffice, etc., no problem at all. I know there are Linux denigrators around here but really, when an OS gives you satisfactory performance on old and underpowered hardware, what's to complain about? A few hours of setup work? You're right, Windows 10 likely will not do so well on your ZG5 if it even runs at all.

  69. Re:That's just great... by __aaclcg7560 · · Score: 2

    A [Washington, ..., Lincoln, ..., Obama] Presidency will be the end of the republic.

    FTFY

  70. Re:That's just great... by cb88 · · Score: 1

    And you're going to assume they aren't based on the base evidence of the huge amount of exploits in just about any software that are discovered over time?

    You are the one making the reaaaaallly bad assumption.

  71. Re:That's just great... by StayFrosty · · Score: 1

    Assuming the hinges don't crack, the laptop doesn't get dropped, the power connector doesn't break, the RAM doesn't fail, there aren't any capacitors waiting to blow, and there aren't any hidden cold joints in a BGA socket somewhere, I guess you will have to find a new distro 7 years from now.

    Most laptops don't come close to lasting 10 years. Desktops are a bit better, but most of them have been 64-bit since 2006 or so. I don't think it's worth wasting a free software project's time and money to support such a fringe case. I'm sure there will be a demand for 32-bit distros in 2023. You will probably be able to use Gentoo, CentOS, Debian, one of the BSDs, or else some enterprising people will create a niche distribution to fill that gap in the market. You will be fine.

    --
    "Frequently wrong, never in doubt."
  72. If you'e still on 32bit... by Anonymous Coward · · Score: 0

    Just upgrade already, why are you purposely trying to extend the life of an old piece of garbage that most likely consume 10x more power than a new PC of much better performances today... Even if sub $300 PC/Laptop would probably perform better than a nearly 10 year old PC still stuck on 32bit.

  73. Re: That's just great... by Anonymous Coward · · Score: 0

    Pinch $120 out to your diamond-pressure level ass and upgrade to something built in the last 5 years.

  74. Re: That's just great... by phantomfive · · Score: 1

    Here is their list. Looks like they've finally given up on vax.

    --
    "First they came for the slanderers and i said nothing."
  75. Re: That's just great... by Anonymous Coward · · Score: 0

    No, it means that you've got 5 years before the final 32-bit version of Ubuntu goes out of support (and probably another 5 years after that if you switch to Centos).

  76. Re:That's just great... by Anonymous Coward · · Score: 0

    And you're going to assume they aren't based on the base evidence of the huge amount of exploits in just about any software that are discovered over time?

    You are the one making the reaaaaallly bad assumption.

    If Linus would have designed linux as a micro-kernel architecture operating system security could have been enhanced out of the gate and even if an exploit is missed the sandbox architecture could limit its effects.

  77. VirtualBox by rastos1 · · Score: 1

    if users desperately need to run 32-bit legacy applications, the'll have to do so in containers or virtual machines.

    The only reason why I install 32-bit stuff on my Linux machine is that it is needed when compiling VirtualBox from source code.

  78. Re:That's just great... by phantomfive · · Score: 1

    I didn't realize Athlon had such a tough life. I would have given him more support!

    --
    "First they came for the slanderers and i said nothing."
  79. Re:That's just great... by Yebyen · · Score: 1

    What show-stopping problem for a 2006 MacBook is present in 18.04 but fixed in 18.10?

    How can anyone honestly attempt to answer this question in 2016, when the release under development is still currently 16.10, and 18.04 won't be receiving its first patch release marking it as stable for more than another two years?

    --
    Restating the obvious since nineteen aught five.
  80. Re:That's just great... by Anonymous Coward · · Score: 0

    QUESTION: What's wrong with running 18.04 until the hardware dies?

    ANSWER: No security updates, and no bugfixes, and no support. That's what's wrong.

  81. Except.you abandon the hardware as well . . . by Anonymous Coward · · Score: 0

    Except for the fact that abandoning 32-bit software also abandons all the older 32-bit hardware that is still functional.

    Perhaps the solution is to keep backporting security and bugfixes to their last 32-bit release. That should be less of a burden that spinning an entire distro.

  82. Re:That's just great... by Hotawa+Hawk-eye · · Score: 2
    Doesn't it? Consider something like DES. If you had a file on your system encrypted with DES:

    In 1977, Diffie and Hellman proposed a machine costing an estimated US$20 million which could find a DES key in a single day. By 1993, Wiener had proposed a key-search machine costing US$1 million which would find a key within 7 hours.

    and:

    One of the more interesting aspects of COPACOBANA [a DES cracking machine] is its cost factor. One machine can be built for approximately $10,000.[26] The cost decrease by roughly a factor of 25 over the EFF machine is an example of the continuous improvement of digital hardware—see Moore's law. Adjusting for inflation over 8 years yields an even higher improvement of about 30x.

    DES hasn't changed, but the amount of computational power attackers can bring to bear has.

    Or to put it a different way: archers manning a castle's walls were a decent defense against melee soldiers ... but they'll do nothing but die when a bomber drops its weaponry inside the walls.

  83. A terrible disturbance by watermark · · Score: 2

    I felt a great disturbance in the Force, as if millions of Intel Atom netbooks suddenly cried out in terror and were suddenly silenced. I fear something terrible has happened.

    AKA, my netbook :(

    1. Re:A terrible disturbance by invictusvoyd · · Score: 3, Informative

      Many of the Atom based notebooks are 64 bit compatible with a minor performance loss.

    2. Re:A terrible disturbance by Voice+of+satan · · Score: 1

      I already switched one of my 900SD ASUS netbooks to debian because xubuntu is slow on my 32 bit celeron M @800mhz. Makes playing some SCUMMVM games more difficult. :/

      Netbook's hardware work fine. Bought them in 2009. No reason to switch.

    3. Re:A terrible disturbance by 4im · · Score: 1

      Indeed, I also fear for my couple of Netbooks, original Asus EEE 9" and 10". I guess the current (K|X)ubuntu 16.04 will be the last Linux distro they'll get.

    4. Re:A terrible disturbance by drinkypoo · · Score: 1

      AKA, my netbook :(

      I have one of those early Atoms, and that machine can't run Ubuntu worth a canned crap anyway

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    5. Re:A terrible disturbance by TeknoHog · · Score: 1

      Many of the Atom based notebooks are 64 bit compatible with a minor performance loss.

      What performance loss? The unbearable slowness of extra registers and instructions? (x86-64 guarantees SSE2, which might not be implemented in a generic i386/i686 binary distro.)

      It's a good point in general, though -- a lot of Atoms are 64-bit. I personally have a Mini-DTX board (aka Mini-ITX with another PCIe slot) with an Atom from 2010 in active service.

      --
      Escher was the first MC and Giger invented the HR department.
    6. Re:A terrible disturbance by Frederic54 · · Score: 1

      True I still use my HP MINI311, o/c to 2GHz, 3GB RAM, Mint17.3/Xfce runs pretty good on it, I am using it to develop on Atmel MEGA processor, it takes a few seconds to compile a 32KB program. It's for in-car apps so the netbook fornat is perfect.
      And even if it's more than 6 years old, it has a dedicated NVidia GPU (ion) and can decode in HW any 1080p video, and output it on the HDMI port plugged on the 55" LCD TV.
      However it's true that Chrome is a little bit slow and heavy sites like reddit with RES brings the CPU to its knees, but to edit/compile small C program on the go, it's wonderful.

      --
      "Science will win because it works." - Stephen Hawking
    7. Re:A terrible disturbance by Anonymous Coward · · Score: 0

      Many? Not mine.

      Citation needed.

    8. Re:A terrible disturbance by invictusvoyd · · Score: 1

      Hope you're not running KDE on it .. I personally recommend fluxbox http://fluxbox.org/

  84. Re:Hipsters are vermin by kwerle · · Score: 1

    Full subject of "Terrible headline. DESKTOP DISTRIBUTIONS letting go" attempted. Failed - too long.

    Fuck /.'s limited subject length.

  85. Re:That's just great... by Anonymous Coward · · Score: 1

    What's wrong with running 18.04 until the hardware dies?

    I've never got Ubuntu to install successfully on the MacBook. Mint Linux install fine.

    Which goes back to the original question:

    If Ubuntu 18.10 is 64-bit only, is that a problem?

    You don't need to reply to every comment, you know. If you don't understand the question or have anything useful to add, you can just sit there quietly.

  86. Re:That's just great... by Zontar+The+Mindless · · Score: 1

    My Acer laptop from 2006 has a single (32-bit, of course) core with 2 GB RAM, and still plays most video just fine. I fire it up every few weeks to see whether it still runs. So far, it always has.

    --
    Il n'y a pas de Planet B.
  87. Linux wants to force you to upgrade by Anonymous Coward · · Score: 0

    Thanks Hillary!

  88. Re: That's just great... by Anonymous Coward · · Score: 0

    I have the best Linux distros. Everybody always tells me I have the best Linux distros.

  89. not being picky but! by Anonymous Coward · · Score: 0

    use of 3 letters goes a long way here for those who know most ARM ports are 32-bit. Those 3 letters are "x", "8", "6".

    Like this: Linux Letting Go: 32-bit x86 Builds On the Way Out

    was that so difficult?

  90. I also don't want to do my job by Anonymous Coward · · Score: 0

    Aw, we don't want to do this because it takes time and it's hard.

    JOIN THE CLUB.

  91. Re:That's just great... by __aaclcg7560 · · Score: 1

    If Ubuntu 18.10 is 64-bit only, is that a problem?

    You don't need to reply to every comment, you know. If you don't understand the question or have anything useful to add, you can just sit there quietly.

    Let me rephrase this: A 64-bit operating system does not run on a 32-bit processor.

  92. 32 bit not supported even now by Anonymous Coward · · Score: 0

    Tried 64 bit Ubuntu on my cheap/disposable AMD A4 travel laptop, it worked but found it stressed the poor thing almost as much as Win 10 did. I tried a 32 bit Linux (which should be less stressing) only to find the 32 bit Linux didn't support several hardware functions (WiFi for example.) It would seem that 32 bit Linux is already depreciated...

  93. Re:That's just great... by __aaclcg7560 · · Score: 1

    How can anyone honestly attempt to answer this question in 2016, when the release under development is still currently 16.10, and 18.04 won't be receiving its first patch release marking it as stable for more than another two years?

    It's a rhetorical question. Especially since I have a 32-bit processor that can't run a 64-bit operating system. Not the same as running a 32-bit operating system on a 64-bit processor.

  94. Re:That's just great... by Tough+Love · · Score: 1

    I guess I'm stuck using Windows 10 (32-bit) on my vintage 2006 MacBook (Intel Duo Core 32-bit processor).

    There are many Linux distributions, some will certainly continue building 32 bit distributions for a long time to come.

    --
    When all you have is a hammer, every problem starts to look like a thumb.
  95. Re:Terrible headline. DESKTOP DISTIES letting go by Anonymous Coward · · Score: 0

    It's becoming more clear every day how the major Linux distributions have all been drinking the same cool-aid. If you think killing 32bit is no big deal you don't use it nor care to understand.

    CentOS 7 for example doesn't come with 32Bit, it had to be created by the community and was / is to a degree, still a mess. Like, Steam. You know, people do really use Linux for gaming. Their rather arrogant response is to
    push Fedora for desktop users. Fedora is a moving freaking target like Ubuntu - it's not coming anywhere near any system of mine.

    I could list off more but in the end I suspect like everything else Ubuntu/Red Hate are killing, it won't matter. Their shills are everywhere.

  96. it's the RAM, dummy! by Anonymous Coward · · Score: 0

    There are several modern computers (e.g. Intel Compute Stick) with only 1GB of RAM. And this is not enough for a x64 OS.

  97. Re:That's just great... by Anonymous Coward · · Score: 0

    Statistically, bugs happen per N lines of code. Given a large N, you will have a substantial number of bugs, of which some subset are security issues. Unless you have some mathematical proof showing the code is secure, we'll use statistics to make accurate but imprecise assumptions.

  98. Old Intel-Atom processors by dogvomit · · Score: 2

    Old Intel Atom processors won't run 64-bit code. My firewall/gateway machine is running an nice but old nano-ITX motherboard with such a processor. I had to download debian's 386 build to get it to work. So, I hope debian at least keeps the 386 build for a while.

    —G

  99. Re: That's just great... by ChrisMaple · · Score: 1

    About the only chance someone not a Democrat or Republican has of being elected President is if both the major party candidates die about a week before the election - and even then, it's iffy; there will be a large number of sentimental/stupid votes.

    --
    Contribute to civilization: ari.aynrand.org/donate
  100. I used to own 8 bit hardware by Anonymous Coward · · Score: 0

    I had 3 or 4 ... 8 bit computers. But haven't had one around for more than 25 years. For a while I had a few 16 bit machines (I think I might have one still around). I have an old 32 bit machine. Haven't turned it on in at least 7-8 years. It was odd going to 64 bits, and at the time there were people who were waffling on whether anyone should switch to 64 bit machines. But its been 64 bit for at lest 7 years now. It will likely be at least another 10 or maybe 15 before anyone says 128 bit (that is pure speculation, but its my opinion that each time you double the number of bits, the longevity of that number of bits also doubles). So 2016 + 15 will be 2031 before we go to 128 bits (more or less). 64 bit memory can address 16777216 terabytes of ram, and can have 16777216 terabytes of instructions. You don't need that many instructions, you won't need that much ram (16777216 terabytes should be enough for anyone), so I think we will be at 64 bits for a while.

  101. Re:That's just great... by ChrisMaple · · Score: 2

    Hillary has sold uranium to the Russians and military communications secrets to the Chinese. The risk of a Hillary presidency is the military defeat of the U.S. and the accompanying millions of dead Americans.

    --
    Contribute to civilization: ari.aynrand.org/donate
  102. Re: That's just great... by MightyYar · · Score: 2

    Donald Trump only has something like 20% odds, yet he'll probably get more than 1/3 of the popular vote. I say if you don't like Hillary, "throw your vote away" on a third party and not someone who has a 9-year-old's solution to immigration and runs a campaign targeting Archie Bunker.

    --
    W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
  103. Re: That's just great... by Anonymous Coward · · Score: 0

    Then there's NetBSD! LOL
    It can run on the Antikythera mechanism....

  104. 32-bit dates; the Y2K of 32-bit OS; 2K38 problem by knorthern+knight · · Score: 1, Interesting

    32-bit posix dates use Jan 1, 1970 as the zero point. The maximum range is...
    -2^31 seconds to
    (2^31) - 1 seconds
    from Jan 1, 1970, 0000 hr

    Try the following short bash script

    #!/bin/bash
    date --date="@2147483647"
    date --date="@2147483648"
    date --date="@-2147483649"
    date --date="@-2147483648"

    On a 32-bit linux system (real or VM), you get...

    Mon Jan 18 22:14:07 EST 2038
    date: invalid date '@2147483648'
    date: invalid date '@-2147483649'
    Fri Dec 13 15:45:52 EST 1901

    If you're a bank amortizing 25-year mortgages, you're already running into problems on a 32-bit linux. On a 64-bit system you'll get...

    Mon Jan 18 22:14:07 EST 2038
    Mon Jan 18 22:14:08 EST 2038
    Fri Dec 13 15:45:51 EST 1901
    Fri Dec 13 15:45:52 EST 1901

    The wraparound date for 64-bit time_t is 15:30:08 on Sunday, 4 December 292,277,026,596 by which time I don't expect to be around.

    --

    I'm not repeating myself
    I'm an X window user; I'm an ex-Windows user
  105. Re:Terrible headline. DESKTOP DISTIES letting go by jmccue · · Score: 3, Insightful

    It's becoming more clear every day how the major Linux distributions have all been drinking the same cool-aid.

    Seems so, but as of now 32 bit Linux has 22 years left (year 2038). I heard that may be fixed but AFAIK nothing yet. In any case for 32 bit I would use NetBSD or OpenBSD since the 2038 issue does not exist for them. I would like to know what "major dist" officially includes :) One distro I consider major just released a 32 bit version.

  106. Re:That's just great... by Anonymous Coward · · Score: 0

    I'm using a 10 year old laptop right now, you insensitive clod.

    Of course, I've had to replace a few components.

    I've replaced the keyboard 3 times, the harddrive 5 times, the ram at least once, the CD drive at least twice, the upper case body 3 times, the lower case body 4 times, and this is the second screen. Oh, and I replaced the motherboard last year.

    The soul, however, is the same XP that it has run since I the day after I bought it.

  107. Re:That's just great... by cfalcon · · Score: 1

    You are on record with stuff like "Trump doesn't have the numbers to win the election, especially if he's getting less than 50% of the Republican vote", then later (once he started getting more than 50% of the Republican vote), "Trump will have less than 51% of the delegates to win the nomination outright" and "the only reason he's winning more than 50% of the vote... is that his his opponents are dropping out. He has yet to win an election by landslide."

    Since then, of course, he's won elections by a landslide, and has way more than 51% of the delegates.

    So previously, you had some set of reasons why Trump couldn't win. Then, when those were no longer true, you picked a different set of reasons why Trump couldn't win.

    I don't know who will win. But I know that you aren't applying any kind of constant standard to this- you chose your conclusion and now are selectively remembering things that back that conclusion.

  108. Re: That's just great... by Anonymous Coward · · Score: 3, Informative

    No they didn't. RHEL did, and since CentOS is more or less just a rebuild of the RHEL sources the base CentOS 7 distribution is also 64 bit only.

    But CentOS has also has an AltArch distribution for i386. It's right there on the download page if you want it.

  109. Fine, but don't blame test hardware availability by Fencepost · · Score: 1

    I'm not heartbroken by the end of 32-bit distros in a year or two, though I do still run a few 32-bit bootable Linux images on old systems being used as remote desktop terminals.

    That said, at least in the Intel-compatible world just about any x64 hardware out there will also run i386 32-bit just fine. You probably don't even have to take out the extra non-usable RAM though I confess I've never tried. Hardware to test a 32-bit build should be no harder to come by than hardware to test an x64 build.

    --
    fencepost
    just a little off
  110. Re:That's just great... by RoverDaddy · · Score: 4, Funny

    I assume the computer name is Theseus.

    --
    RETURN without GOSUB in line 1050
  111. Re: That's just great... by Anonymous Coward · · Score: 0

    Gary Johnson actually governed a state for a while. I think he's better than Trump.

  112. die in a fire by Anonymous Coward · · Score: 0

    There's nothing childish about suggesting we have a functioning border, you (((rootless cosmopolitan))).

    Modern technology allowed more foreigners to enter Britain in 2015 than every year between 1066 and 1950 inclusive.

    Are you willing to concede that the native Amerindians were a bunch of nasty xenophobes for opposing all those friendly European immigrants? They were opposing the almighty market and the neoliberal god Mammon... to send them upon a trail of tears was a great kindness.

    May death alight upon your wretched skull with the utmost haste.

    Trump 2016.

    1. Re:die in a fire by MightyYar · · Score: 1

      Look at what you wrote. Britain, an island with a natural "wall" 19 miles wide at its narrowest point, is having an immigration problem. Trump wants to solve it with a flimsy fence. Thank you, I could not better illustrate what a child-like solution to the problem it is.

      --
      W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
    2. Re:die in a fire by Anonymous Coward · · Score: 0

      So you're saying landmines would be a better solution? I like your thinking!

    3. Re:die in a fire by __aaclcg7560 · · Score: 1

      Britain, an island with a natural "wall" 19 miles wide at its narrowest point, is having an immigration problem.

      You're overlooking several things in regards to Britain. A rail-and-automotive tunnel that goes underneath the "wall" to connect Britain with the rest of Europe. As a member of the European Union, any citizen from any country in the EU can travel to Britain without a passport. Also, the British economy is doing a lot better than the rest of Europe. Hence, they have an immigration problem.

    4. Re:die in a fire by MightyYar · · Score: 1

      You're overlooking several things in regards to Britain.

      No, the Trump supporter is overlooking things in regards to Britain. I'm well aware that immigration is a complex issue with many variables.

      --
      W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
  113. Re: That's just great... by Anonymous Coward · · Score: 0

    of course, the inevitable turn to the political from a debate about linux. What's next? systemd?

      hahahahahahahhah

  114. Re: That's just great... by Bing+Tsher+E · · Score: 1

    My Asus Transformer Book Tablet/Convertible sports a modern fast and low power 64 bit Atom processor but it runs the 32 bit version of Windows 10. When I bought it, it had 32 bit Windows 8.1 installed. It runs 32 bits because it only has 2GB of RAM so there's no real reason to run 64.

  115. Or a Thinkpad. by sethstorm · · Score: 1

    Even the worst of them run longer than the support life of their intended software.

    --
    Twitter supports and protects racists - by smearing their critics with the "Hate Speech" label.
  116. Flexview T60p? by sethstorm · · Score: 1

    Now might be a good time to upgrade to at least a T7200 if not put in a 14" T61 board. Did the latter with a T60p, adding a SSD, 8gb of memory, and a T9500 without regrets.

    --
    Twitter supports and protects racists - by smearing their critics with the "Hate Speech" label.
  117. Re:That's just great... by Anonymous Coward · · Score: 0

    I heard that systemd is getting backported to 8 bit CPUs because with so few bits, one really needs proper daemon management.

  118. Re: That's just great... by Anonymous Coward · · Score: 0

    Roflmao

  119. Re: That's just great... by Anonymous Coward · · Score: 0

    Real Linux users compile their entire system from source. Who needs a distro?

  120. Re: That's just great... by Anonymous Coward · · Score: 0

    Translation: Waah! My 10 year old laptop won't be able to run the latest Ubuntu in 3 years! This is unfair!

  121. Re: That's just great... by jrumney · · Score: 1

    18.04 will be a LTS version, supported with security patches until 2023.

  122. Re:That's just great... by cfalcon · · Score: 1

    What distro do you currently have loaded on your 32-bit 10 year old MacBook? Lets find out exactly when you are going to lose support. I would be shocked if whatever distro you are using will have no updates for you before your machine is totally dead and you are using a newer box.

  123. Re:That's just great... by cfalcon · · Score: 1

    I'm sad you didn't get more Funny mods for Laptop of Theseus lol

  124. Re:That's just great... by cfalcon · · Score: 1

    > *Sigh* that ASSUMES the exploits where there to begin with.

    This is a good assumption. It has been true of 100% of operating systems so far, right?

    > Something does not become less secure just because of its age.

    If it is in the 0% of software with no flaws, this is true. But, it is 0%.

  125. Re: That's just great... by cfalcon · · Score: 1

    There are several candidates, but a vote for a non-Democrat or non-Republican must be cast from a position of protest, principle, or pragmatism. Protest says that you are not ok with any either of the mainstream candidates, principle says that you are seeking a mostly-perfect solution and will not compromise, and pragmatism says something like "be more like this guy, and you'll have my vote".

    A third party candidate cannot currently win. If you are voting based on the idea of "if my vote was the deciding one, who should it be?", you must choose one of the two major parties. I believe that most people vote this way, based on the "I have only two choices" meme I see pretty much everywhere.

    Our system is built to be terrible with three parties. First, we have a plurality system at almost every state, so that a candidate with 15% of the vote everywhere will round those out and deliver effectively zero electoral votes. Second, everyone is afraid of a popular conservative candidate spoiling for Democrats, and a popular liberal candidate spoiling for Republicans- this is the exact opposite of what voting SHOULD do. Instant runoff, multiple choice, and Condorcet all seek to address this second point, but the two groups with power have no motivation to change this, as it will only hurt them as organizations. Thirdly and finally, should a third party candidate be popular enough to carry a couple key states (or say Jill Stein turns some blue states green, and simultaneously Gary Johnson turns some red states yellow), then no candidate will have a majority of electoral votes. In this case, the House of Representatives just chooses a candidate. A third party candidate with 49% of the electoral votes could be passed over by a Congress that installs the third place dude.

    Any of these rules could be changed to fix the electoral system and allow meaningful third party choice. But until then, a vote for a third party candidate is protest, principle, or pragmatism, and it will never be a vote for a winning candidate.

  126. nike tn pas cher Homme by zhenmeihua · · Score: 0

    for every individual. Nike, originally known as Blue Ribbon Sports, was founded by university of Oregon track athlete Philip Knight and his coach Bill Bowerman on January 25, 1964. Nike produces a wide range of sports equipments and his famous products were track running shoes. Where you can take active part in any sports activities with a pair of Nike shoes. Well, Nike is a part of wide ranges of sports activities like baseball, ice hockey, tennis, Association football, lacrosse, basketball and cricket. Well Nike gives inspiration to its fans and viewers. Nike lovers have the freedom to choice amongst the brands appeal them. To prove it value of being a high quality brand there are various benefits where every individual love the creativity of Nike. Nike is created with greater technology such as: synthetic and mesh upper, phyton, midsole with encapsulated air-sole unit for cushioning, rubber outsole with herringbone pattern for traction and durail, Comfort technology such as follows: light weight, super responsive performance cushioning, incredibly thin, Nike zoom brings the foot closer to the ground for optimum feel and aggressive maneuverability nike tn pas cher .

  127. Re: That's just great... by Anonymous Coward · · Score: 0

    OpenBSD makes new releases every 6 months? Didn't know that...

  128. Re: That's just great... by MightyYar · · Score: 1

    I mostly agree with everything you wrote, but I would add that a vote for a third party candidate makes it easier for that party to get ballot access in the next election. The more elections a third party is in, the more likely the changes you describe are to happen. We will never have a change to the system if we keep voting for the two entrenched parties.

    Personally - and I'm not sure how you'd classify this, perhaps principle, I'm more likely to stay home than vote for Trump or Clinton. They are both wrong for this country in different ways and I'm not sure who would be worse, though Trump is certainly more of a wild card. But essentially they both play for the same team, so I don't really think it matters too much who wins. At the end of the day, the people paying for the $10,000 plate dinners still run the show - and they really don't care much about the R or the D. Given that there are two other choices, if I actually go to the polls, I'll pull a lever for one of them. I like libertarians (of the less wacky variety) in general - I'd probably even describe myself as one except where the ideology runs into pragmatism... like everywhere :) But even Stein is preferable to the mainstream candidates.

    --
    W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
  129. Some 2016 stuff still 32 bit by dbIII · · Score: 1

    Yes but there are plenty of things based on 32 bit ARM CPUs being sold in 2016.

    1. Re: Some 2016 stuff still 32 bit by p91paul · · Score: 2

      The headline is click-bait, but they are referring to intel 32 bit CPUs. I see no plans to drop arm 32 bit support, since arm64 is still in its early days.

    2. Re: Some 2016 stuff still 32 bit by dbIII · · Score: 1

      Good point.
      I have far too many 32 bit Pentium IV things still lurking about but most have FreeBSD on them anyway.

    3. Re: Some 2016 stuff still 32 bit by Ilgaz · · Score: 0

      Yea, in 2 years Dimitri will notice everyone using 64bit ARM in his local Starbucks and propose to drop support of 500.000.000 32bit ARM machines.

      These are the only guys who propose something against Google Android BTW. One of my Android shareware just had an update that brings back Android 2.0 support since Google make it possible. Give up Android OS for something these guys ship. No thanks.

    4. Re:Some 2016 stuff still 32 bit by MachineShedFred · · Score: 1

      and they are specifically talking about dumping the i386 distributions, which none of those ARM CPUs run.

      --
      Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
  130. Re:That's just great... by dbIII · · Score: 1

    You may want to plug some shiny thing that doesn't exist yet into it but have no driver - so then it would be time to give up on Ubuntu or SUSE to go with Debian, Slackware, centos, etc, etc or FreeBSD to use it.

  131. Re:That's just great... by dbIII · · Score: 1

    With respect she's not really very different to her husband and the sky didn't completely fall in when he was running the place. Trump on the other hand doesn't seem to understand what a Republic actually is and seems to want to be King, so he's unlikely to get much of the party behind him especially the very conservative types. If he does get in expect a lot of shouting and not much getting done.
    Remember that Obama couldn't close GITMO and that the health plan had to be watered down a lot to pass. The President does not wield absolute power and there is plenty to stop a Hillary Presidency being the end of the republic - once again probably a lot of shouting and nothing extreme getting done.

  132. Re:That's just great... by dbIII · · Score: 2

    So? Reagan sold a lot of stuff to Iran and even Hezbolla.
    "Treason doth never prosper, what's the reason? For if it prosper, none dare call it Treason."

  133. Re:That's just great... by dbIII · · Score: 1

    Hey slashdotters - go upstairs and ask your mother why she's not going to vote for Trump to get an idea of why people are saying he can't win.

    You don't want to ask? The short story is the conservative types that make up a lot of the Republican party and do all the fetching and carrying at election times think he is a sleazy slimeball and not a "real" Republican. The longer story is that he's said a lot of stuff that people are taking personally. Completely angering the conventional Christian voters, the Hispanic voters and woman of all ideologies (especially the conservative ones) pokes a bit of a hole in his chances and adds up to a lot of people that would rather not turn up than vote for either Trump or Clinton.
    Since Trump cannot rely on the rusted on Republican vote I doubt he's going to win unless not many people turn up to vote for Clinton.

  134. Re:I think it's wrong, they're killing i386 not i6 by Wyzard · · Score: 1

    "i386" is still the name that Debian and its derivatives (like Ubuntu) use for the 32-bit x86 platform, regardless of the specific chip. Debian actually dropped support for pre-686 CPUs a few months ago, and had required at least 586 for several years prior, but the overall architecture is still called "i386", because that's what it's always been called, and there's no real benefit (and lots of inconvenience) in changing it. Same reason why 64-bit x86 is called "amd64" even though Intel implements it too.

    This Ubuntu proposal is about dropping 32-bit x86 entirely, not just certain old chips.

  135. Re:That's just great... by Anonymous Coward · · Score: 0

    "This is the actual axe that George Washington used to cut down the cherry tree."

    "I had to replace the handle... and the head. But it occupies the same space, intrinsically!"

  136. Re:That's just great... by dbIII · · Score: 1

    So he finally decided to dress like a girl. Seems he did not know that women folk is not allowed at Olympic games either. Sad fate. Really sad.

    Especially with the amount of effort required to carry that off at a naked Olympic games!

  137. Re:That's just great... by Anonymous Coward · · Score: 1

    (Tap tap tap...) Now it says all the models have dual overhead cams, neon ground effects, hydraulics and LED spinners on the wheels!

  138. Re: That's just great... by cfalcon · · Score: 1

    > I'm more likely to stay home than vote for Trump or Clinton

    That just means you don't give a fuck. If you skip the section that says "President of the United States" but vote the REST of the ballot, then that sends a message. Staying home just means you were high that day or whatever.

    > But essentially they both play for the same team

    I think Trump is enough of an outsider that he may not be able to have any say in legislation. I don't feel they are that similar. Most years I'd agree though.

  139. Re:That's just great... by cfalcon · · Score: 1

    > get an idea of why people are saying he can't win
    > I doubt he's going to win unless not many people turn up to vote for Clinton

    See, these are two different things. The top one is wrong. Trump can win. Anyone saying Trump can't win should have a lot of money on Hillary in the prediction markets- free money!

    The bottom one is a much more reasonable statement. Certainly, Trump is not favored to win. But his odds aren't terrible. Personally, I find the "Trump has alienated a lot of Republicans" point to not be very compelling, because the Republican party has been casting Hillary as an archdemoness for over two decades. I think this will guarantee an above-average Republican turnout in ANY election she runs in. Meanwhile, Trump has record unpopularity (Clinton WOULD be the least unpopular candidate ever measured, except that Trump is in this same election), and that could motivate some normally unmotivated Democrats.

    My real point is simple: this is an unprecedented election, so claiming that one candidate is totally fucked based on X Y Z doesn't seem very predictive.

    If I had to put money on it, I'd put money on Trump. He's demonstrated a political resilience unseen, and seems to have a whole new rulebook he's playing by. But I don't have to put money on it, so I don't. My one personal belief is that the election *will not be close*. That is to say, I think either Trump OR Clinton will have a solid margin of victory. I think both are telling stories that make the other one out to be Very Scary, and while THAT is nothing new, people are putting a lot of faith in those narratives this year, and I think one will be a reasonably clear victor.

  140. Re:That's just great... by dbIII · · Score: 1

    The top one is wrong

    The top one is correct because "people are saying he can't win". Please READ the items you cut and paste.

    and seems to have a whole new rulebook he's playing by

    Not really. Look up "carpetbagger" from years back or look at other places around the world today.

  141. Re:That's just great... by cfalcon · · Score: 1

    Lol, not disputing that you are correctly referencing a sentiment. I said why the sentiment was incorrect.

    > Not really.

    I think it's too early to say. If Trump had been smashed in the primary then we wouldn't be having this conversation. I didn't see a lot of Trump haters in, say, November, saying "Trump will dominate in the primaries, force everyone out of the race by early May, easily reach the 1237 bound delegates needed, get more primary votes than any Republican in history, but then lose to any Democrat because $REASON_ARRAY".

    That's not to say he can't lose, or that $REASON_ARRAY isn't solid. But it IS to say that everyone talking about him getting blown generally has a totally shit record this election thus far. All the political prediction guys I normally follow got fucking SHRILL man. They all were predicting him getting blown out, then him having a ceiling, then a larger ceiling, then a contested convention, then some fucking arcane bullshit about secret Republican councils, and then they started talking about how he's gonna lose the general. Well, he could lose the general. But so far, taking anything out of their mouths and immediately betting on the opposite state would have been mad profit.

  142. Re: That's just great... by MightyYar · · Score: 1

    Well, he has played golf with his opponents husband on more than one occasion... how different can they really be?

    And by different, I don't mean in effectiveness... it's hard to even say how effective Trump would be since he has zero track record. I mean in their worldview. There will be little change to any major policy.

    --
    W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
  143. 64 bit builds are a waste of resources by iamacat · · Score: 1

    You are doubling size of typical data structures made up of ints and pointers, while RAM and especially CPU cache is a very finite resource. Why would you want to do that when modern hardware supports 32 bit software perfectly well?

    1. Re:64 bit builds are a waste of resources by drinkypoo · · Score: 1

      You are doubling size of typical data structures made up of ints and pointers, while RAM and especially CPU cache is a very finite resource. Why would you want to do that when modern hardware supports 32 bit software perfectly well?

      CPU cache is an issue, RAM is not. The additional RAM used up by ints (which don't actually have to be long ints) and pointers is basically irrelevant today, because RAM is practically free today. $100 bought me 16GB of very fast stuff. I can afford double-size pointers. Meanwhile, a simple recompile to x86_64 can easily produce over a 10% performance improvement because register renaming doesn't actually solve register starvation, it only mitigates it; and because shoveling data is twice as fast, and there's considerable shoveling in the most intensive typical task for home PCs: gaming. It's also a big help while transcoding, which comes up more and more often.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    2. Re:64 bit builds are a waste of resources by iamacat · · Score: 1

      Not every Linux device is a consumer desktop or laptop, RAM in mobile/embedded space is far from free. And again, cache is much smaller than RAM.

    3. Re:64 bit builds are a waste of resources by drinkypoo · · Score: 1

      Not every Linux device is a consumer desktop or laptop, RAM in mobile/embedded space is far from free.

      At the point at which you're considering 64 bit, you probably also have at least 1GB, and the additional pointer length is not going to cause you any real-world harm. I don't like to mess with anything with less than 2GB any more, unless it's a really dinky platform like Arduino or some other microcontroller.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    4. Re:64 bit builds are a waste of resources by david_thornley · · Score: 1

      If you're trying to economize on hardware in your embedded or mobile device, what are you doing running a desktop distro? Certainly you could shave off some money by not running all of Ubuntu. TFS implies that this is for desktop distros like Ubuntu.

      --
      "When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
  144. Re:That's just great... by dbIII · · Score: 1

    I normally follow got fucking SHRILL man

    He's not playing by their rules, or anybodies, or with little reference to reality, so he really pisses them off.

    The rest of the world just looks on with cynical pity and says - "so an Atlantic City Gangster wants to cut out the middle man and be President, how is that really going to change much about the USA?" - they may be wrong but not by as much as we would hope. The line between casino boss and gangster may as well not be there for example which makes the populist angle with Trump especially weird. Why do people think someone who is likely to get them hurt if they can't pay a debt to him has their best interests in mind?

  145. Re:That's just great... by cfalcon · · Score: 1

    I mean, I'm with you on that point. The article cracked did ( http://www.cracked.com/blog/tr... ) was pretty informative, but the part that stuck out to me was: "People are not angry at Washington; they are totally over Washington. They don't feel Washington can do anything to make their lives better."

    This level of distrust, derision and populist hatred was earned over the years. It did not come instantly, it did not come from Rush Limbaugh, it was not a top-down phenomena. To get candidate Trump, you have to EARN candidate Trump, with years of bullshit, weakness, and pandering. If you spend years promising some group that you are opposed or in favor of something, and you never actually press for the change you promised, you'll eventually lose that group. It doesn't even matter if the thing you promised is idiotic and impossible, if you keep promising it, you'll eventually be branded a panderer, whatever the word chosen to brand you is. To get president Trump, it is mostly the same formula, just expanded over more voters, and we'll see if only the Republicans feel this way soon enough.

    I am worried that if Trump wins, the take-home message to most of the politicians will be "be bold and brash, appear independent". That's just one of the things Trump is doing to appeal to people, and I think he is doing it because Trump is inherently bold, brash, and independent- it isn't some poser bullshit, but nor is it relevant. The fact that people who feel disenfranchised are swarming to Trump is VERY IMPORTANT, and all the attempts to break his campaign down into tiny bite sized tactics that future operatives can employ on the field is doomed to failure. Trump will succeed or fail based on how much faith America has lost in its highly educated oligarchic leader-class (that it pretends doesn't exist, because class doesn't exist and everyone is equal). If Americans feel that only Donald Trump has their best interests in mind, that is not because Donald Trump is a High Wizard Of Illusion (though he appears to be!), it's because they stopped believing that their leaders are leading in the correct direction. That's not just a communication failure, it's also a failure of direction.

  146. Re: That's just great... by Hognoxious · · Score: 1

    CentOS 6 is supported till 2020.

    By then hopefully someone else will have made systemd secure and reliable or bitten the bullet and chucked it.

    --
    Confucius say, "Find worm in apple - bad. Find half a worm - worse."
  147. WTF? by Anonymous Coward · · Score: 0

    > the OpenSUSE Chairman account says 32-bit support "doubles our testing burden (actually, more so, do you know how hard it is to find 32-bit hardware these days?).

    WTF is "hard to find 32-bit hardware", dude? Nobody asks you to dig out an old 486 computer. Last time I checked, all x86-64 CPUs had full i386 support built-in.

  148. saw that coming by luther349 · · Score: 1

    but there rite thers so few 32 bit systems left in fact i dont own any. it may be a porblem for embedded systems but most of those run old versions of a os anyways.

  149. Re:That's just great... by TheRaven64 · · Score: 1

    If Trump were not running, Hillary's popularity would be the lowest of any nominee since they started recording polls. As he is, she's the second least popular - but not by a very large margin. Both parties have managed to pick candidates that most of the electorate regard as unelectable. The election is basically down to whether anyone can be bothered to turn up to vote for who they regard as the slightly lesser of two evils.

    --
    I am TheRaven on Soylent News
  150. Re: That's just great... by TheRaven64 · · Score: 1

    VAX never got ELF support. I believe OpenBSD GC'd it when they removed all of the a.out-only architectures. The OpenBSD GNUstep port maintainer was the only one to ever test Objective-C on VAX and we fixed a couple of bugs as a result.

    --
    I am TheRaven on Soylent News
  151. Re:That's just great... by TheRaven64 · · Score: 1
    You still want a 64-bit OS on x86, even if you don't care about the larger memory. Targeting x86-64 gives you (at least - I've probably forgotten some things):
    • A guarantee that SSE is there, so you can use SSE for all floating point (much faster than x87, even if you don't use the vector functionality) and a calling convention that uses SSE registers so you aren't doing expensive x87-SSE moves all the time to actually do calculation (which you get if you target i686+SSE).
    • More registers, which reduces instruction count noticeably.
    • Cheap PC-relative addressing, which makes position-independent code (all libraries and typically the main binary if you're doing ASLR too) around 10% cheaper.

    If you really don't want to use more than 4GB of virtual address space, the X32 ABI lets you get all of these benefits in an ILP32 environment, so brings all of the performance benefits and none of the costs of x86-64.

    --
    I am TheRaven on Soylent News
  152. Re:That's just great... by dbIII · · Score: 1

    This level of distrust, derision and populist hatred was earned over the years. It did not come instantly

    Reagan pushed that line HARD. “Government is not the solution to our problem; government is the problem.”
    A lot of P.R. money has been spent over the years pushing that line.

    I am worried that if Trump wins, the take-home message to most of the politicians will be "be bold and brash, appear independent"

    See that Reagan quote above as to why the horse has already bolted and Trump is just someone who took home the message. He's not a sign of things to come, he is that thing.

  153. And yet Valve still doesn't have a 32-bit Steam by drinkypoo · · Score: 1

    It's sad and pathetic that you literally can need multiarch on your Linux system just to run steam. The games usually are 64-bit (actually, they usually come with both 32- and 64-bit binaries) but Steam still isn't. It's amazing how incompetent Valve can actually be, and still lead the market.

    --
    "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
  154. Re:That's just great... by drinkypoo · · Score: 1

    I have the original Acer Aspire One and it cannot play fullscreen youtube, even at 480p. It just doesn't have enough CPU to pull it off.

    --
    "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
  155. Re: That's just great... by Anonymous Coward · · Score: 0

    It will, however, be able to run 32-bit Windows 10. You know, the OS that gives life back to older hardware.

    Oh how the world has turned upside down the last couple of years. Linux requiring new hardware, Microsoft products being something you only run if you can't afford anything better, Apple being as for from bankrupt as technically possible...

  156. MS sales guys will have another argument by Ilgaz · · Score: 1

    "Look, you have thousands of 32 bit only machines with 2 GB of memory. We support them with Windows 10 32 bit while Linux will drop their support"

    This is how Linux could never be a credible player on corporate desktop. Some guy who has zero experience in the real World, real business scene propose something with corporate wannabe talk, everyone claps and says "Yea drop those old stinky CPU support".

  157. Even Apple won't support them with the new OSX by Ilgaz · · Score: 1

    One can't demand support for Macbook unless he/she donates several machines and money for this purpose to the distribution and if distribution agrees to support them.

    I am seriously pissed off by these Starbucks Latte drinking, trendy types poisoning the Linux scene too but let's not forget Apple are the guys who shipped 32bit only machine with some weird EFI back in 2006. They always pick 2 generation old CPU and add some non standard weird firmware. You can't expect support from Linux or BSD guys, they need access to real hardware.

    1. Re:Even Apple won't support them with the new OSX by Anonymous Coward · · Score: 0

      > They always pick 2 generation old CPU

      WTF are you smoking? Apple's first Intel MacBook Pro in 2006 featured a 1.83 GHz Core Duo, which was brand spanking new.

  158. Re:That's just great... by Anonymous Coward · · Score: 0

    Windows is another whole matter. For one, the primary reason why those low-cost tablets run Windows x86 is because they have 32 bit UEFI, and Windows x64 does not support booting from that.
    Then there is the issue of disk space, which is already tight. Turning such a device on for the first time and letting it install updates will typically see only 16 GB of the 32 GB in popular models remaining free. With Windows x64, this would be much worse because it needs to have both 32 and 64 bit libraries present.
    Finally, the little RAM is more of a problem on Windows. On Linux x86_64, almost all applications that a user typically runs are 64 bit. On Windows however, many programs are still 32 bit. This means both 32 and 64 bit libraries need to remain in memory all the time, filling it up quickly.

  159. Re: That's just great... by Anonymous Coward · · Score: 0

    Mild nitpick, but important when discussing the removal of a platform target from a distributor:

    IA32 is 32bit Itanium, not x86 compatible.

  160. When will they start on the 128-bit linux? by Anonymous Coward · · Score: 1

    When will they start on the 128-bit linux? We haven't increased word size for a while

  161. Realtime linux? by Anonymous Coward · · Score: 0

    What about realtime Linux that can run on 32, 16 and maybe 8 bit cpus acting as micro controllers?

  162. Not Everybody Buys A New Computer Every Year by BattyMan · · Score: 1

    And some of their bosses don't just cash out for new machinery every decade.
    My own boss' policy is to run the old junk until it breaks down.

    Neither of us have 64 bit computers.
    Neither of us even have computers new enough to do useful virtualization (ie run WinXP ina box).
    Neither of us have money to buy new computers.

    --
    Exceeding the recommended torque is not recommended.
    1. Re:Not Everybody Buys A New Computer Every Year by david_thornley · · Score: 1

      If what you have right now works, do you really need to run the latest version of Ubuntu? Particularly when there's going to be a LTS version for quite some time?

      --
      "When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
  163. Too bad... by Anonymous Coward · · Score: 0

    ...for me. All I have left are 32-bit processors. My 64-bit ones have all failed and I really don't want to spend the money to replace them.

    Which brings up a new point, why do 64-bit systems so failure prone when compared to 32-bit?

  164. Like there's a lack of distros ... by Qbertino · · Score: 1

    ... that will continue 32bit support. I bet there are a few out there who are glad to jump to the occasion and get a larger userbase.

    --
    We suffer more in our imagination than in reality. - Seneca
  165. Re: That's just great... by Anonymous Coward · · Score: 0

    Or to put it a different way: archers manning a castle's walls were a decent defense against melee soldiers ... but they'll do nothing but die when a bomber drops its weaponry inside the walls.

    And yet CIV would have me believe that spear infantry dug into a mud hut city can destroy a main battle tank

  166. Re: That's just great... by Billly+Gates · · Score: 1

    Well in 2 years it will be nearly 10 years old!

    Did you cry that XP didn't run on your 9 year old 486 in 2001 or Windows 95 won't run on your 9 year old 8086 IBM XT with 1 meg of ram? It happens. If you're on slashdot you are probably a geek who runs alternative OSes anyway. Run freebsd or get with the times?

    Technology moves forward for the rest of us

  167. Re: That's just great... by __aaclcg7560 · · Score: 1

    Did you cry that XP didn't run on your 9 year old 486 in 2001 or Windows 95 won't run on your 9 year old 8086 IBM XT with 1 meg of ram?

    I only had a used 486 computer for six months before I got a Pentium. DOS ran just fine on my IBM AT until I retired the system in 1997. This week I'm replacing the nine-year-old system that I built for Windows Vista in 2007.

    Technology moves forward for the rest of us

    Sometimes old technology keeps on working just fine.

  168. Hiow hard is it to find 32 bit hardware by Eunuchswear · · Score: 1

    Every fucking PC on sale today runs 32 bit code just fine.

    What an idiot.

    --
    Watch this Heartland Institute video
  169. Re:Terrible headline. DESKTOP DISTIES letting go by Anonymous Coward · · Score: 0

    Glad there are more people mentioning those systems. FreeBSD is nice. But those two operating systems beg to be mentioned after an article like this one.

  170. Re:That's just great... by MachineShedFred · · Score: 1

    That hardware is a decade old, and the LTS distributions will still give another 5 years of maintenance before pulling the plug.

    Do you really expect to be using a 15 year old notebook and get current software? Really?

    --
    Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
  171. Re: That's just great... by __aaclcg7560 · · Score: 1

    Pinch $120 out to your diamond-pressure level ass and upgrade to something built in the last 5 years.

    I paid $1200 for my MacBook, including the $200 premium to get a black case. It's still a useful machine. I'm not ready to pop out another diamond out of my ass for the newest and greatest MacBook.

  172. Re: That's just great... by slazzy · · Score: 1

    True, 12 years old for a main machine is pretty old. For some secondary tasks no big deal though

    --
    Website Just Down For Me? Find out
  173. Re:That's just great... by fgouget · · Score: 1

    Lemme rephrase. If Ubuntu 18.10 is 64-bit only, is that a problem? What show-stopping problem for a 2006 MacBook is present in 18.04 but fixed in 18.10?

    You're missing the point if you only worry about 2006 MacBooks.

    Making 18.10 64-bit-only means it will no longer be able to run most Windows applications on account that 32-bit Windows applications and installers can only be run by a 32-bit Wine process. Not just on old 32-bit-only MacBooks but on any Intel based computer. Also Wine is still receiving lots of updates to run new Windows applications so being stuck with the version shipped in 18.04 means getting behind the times pretty quick (unless they really keep updating the 32-bit Wine package long after 18.04 is in maintenance mode).

  174. Re:That's just great... by Anonymous Coward · · Score: 0

    RHEL7 (and hence CentOS 7) has already dropped IA-32. RHEL6 will only be fully supported until 2020.

  175. ...Aaaannnd You Missed the Point by Anonymous Coward · · Score: 1

    You can use 64-bit CPUs as stand-ins for pure 32-bit CPUs because of instruction set compatibility.

    That's why the OpenSUSE Chairman sounds dumb, talking about how hard it is to get 32-bit hardware. In fact it's drop-dead easy to get 32-bit hardware.

    You seem to have missed this point completely. Thanks for playing!

  176. Re: That's just great... by Anonymous Coward · · Score: 0

    Corporate interests once again. At one time Linux was famous for keeping old systems alive when Microsoft made them obsolete. Now corporations are making Linux machines obsolete. What about all the 486 systems being developed today for industrial applications? Are corporations going to screw them over?

  177. Re: That's just great... by Anonymous Coward · · Score: 0

    Great! I won't have to worry about a 32-bit distribution for my 64-bit Raspberry Pi any longer! What a relief!

  178. Re: That's just great... by Anonymous Coward · · Score: 0

    You fucking idiots couldn't even be bothered to RTF summary could you? This is IA32, not ARM.

  179. Re:That's just great... by Anonymous Coward · · Score: 0

    And let me rephrase what the previous poster asked: What exists in the absolute newest distribution 5 years from now that you must have on your then 17-year-old laptop, that isn't already in the final i386 distribution released 4.5 years from now? What critical functionality will you be missing that you can't spend $50 at that time to buy what is today a brand new laptop to get around it?

    Just because they will cease i386 development in the future, doesn't mean that all previous releases somehow disappear or self-uninstall on that magic date.

  180. Re: That's just great... by Anonymous Coward · · Score: 0

    I can just see Patrick Volkerding chuckling in the background over this discussion. Another 50 years of Slackware, right Patrick?

  181. Re:That's just great... by Anonymous Coward · · Score: 0

    the better question:

    If you are concerned with fixes and support, why are you running 12 year old hardware and have any expectation of receiving fixes and support?

  182. distros or linux by trigggl · · Score: 1

    When is the kernel letting go?

    --
    Ops, I shuld have usd the prevuwe but in.
  183. Re:That's just great... by MachineShedFred · · Score: 1

    You really think that the GOP is going to lose 70+ seats in the House?

    Even the most blind sycophantic Democratic operators aren't predicting that one.

    The Senate is up for grabs, and the White House likely stays with the Democrats, but the House will still firmly be in Republican hands.

    --
    Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
  184. 32-bit compat out too? by fluffynuts · · Score: 1

    TFA seems to suggest that 32-bit compat is out too (by the suggestion to run 32-bit processes in containers / virtual machines).

    If this is also the plan, they can take a long walk off a short pier. There are plenty of 32-bit images still out there (Steam games? Other stuff) which work just fine.

    I've already ditched Ubuntu, going back to Debian simply because my system rotted through updates to the point where sound was a hit and miss affair. This just gives a solid reason not to care about Ubuntu any more. Pity, as it's "home-grown" for me )':

  185. multiple distributions by Anonymous Coward · · Score: 0

    You know what else is a waste of resources...having a billion Linux distributions.

  186. Back when Linux was "cool"... by bobmajdakjr · · Score: 1

    So, back when Linux was still "cool" to me, in Jr High, so around 1999 I remember arguing with friends and family about how awesome it was. One of the arguments I used to parrot all the time was "they build it for every machine ever! hardware capitalism gave up on can still be made viable because it installs on everything!" And you could do it without having to go full gentoo-kun. My first Linux was on an old PPC Mac, good ol Yellow Dog. How far Linux has come now that they are considering dropping a shitload of binaries for still widespread machines. It used to be for the people by the people. I guess it is not anymore. Maybe I've just been hanging out with Hancock too long.

    1. Re:Back when Linux was "cool"... by ebvwfbw · · Score: 1

      You're right. Keep it alive. Buy 32 bit hardware and volunteer to build the various distributions and maintain them.

      From me - "Thanks!"

  187. Re:That's just great... by MachineShedFred · · Score: 1

    Whinging about a 2006 Core Duo based notebook that was famous for having the fans fail and fry the GPU probably isn't the best case.

    Whinging about hundreds of 2014 thin client using a 32-bit Intel Atom is a bit more of a concern.

    --
    Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
  188. Re:That's just great... by Shirley+Marquez · · Score: 1

    18.04 will be an LTS release that continues to get security updates until 2023. By then all the 32 bit hardware is likely to be dead. The downside is that you won't get some new features that will be in releases after 18.04.

  189. Re:That's just great... by Shirley+Marquez · · Score: 1

    The 32 bit version of Windows 10 will run with 1GB RAM but it won't be great - basically you'll be able to run one program at a time, and web browsers won't work well. That's the configuration of a lot of the small tablets on the market. The Anniversary Update will increase the official RAM requirement to 2GB, and the 64 bit version already requires 2GB.

  190. Re: That's just great... by cfalcon · · Score: 1

    > DOS ran just fine on my IBM AT until I retired the system in 1997

    The last version of DOS came out in 1994. The IBM AT came out in 1984. That's a ten year difference. Its 2016, and the 2006 laptop still has current versions of Linux with a 32 bit OS on them, and will for years. What's the problem? If you retire the 2006 machine in 2019, you'll still have a current version of Linux even on the timelines discussed in the article.

  191. Re:That's just great... by david_thornley · · Score: 1

    Typically, people with really old computers don't pick up new shinies to plug in. In most cases, if they were interested in new shinies they'd be on a 64-bit system now, and Ubuntu will be supporting 32-bit systems for quite a few years to come.

    --
    "When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
  192. Re:That's just great... by david_thornley · · Score: 1

    One take-home message if Trump wins is that outsiders are ineffective. Trump would get less done than Jimmy Carter.

    Sanders was his counterpart in the Democratic race, and Sanders did a lot better than most people would have expected. I think the Brexit vote is more of the same.

    In the very likely contingency that Clinton is elected, she's going to have to deliver something good to the Trump/Sanders constituencies, or they'll be back and in greater numbers.

    --
    "When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
  193. Re:That's just great... by david_thornley · · Score: 1

    It's far too early to claim that the Republicans will keep the House. There's a distinct possibility of the Democrats winning very big indeed.

    --
    "When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
  194. Re:That's just great... by dbIII · · Score: 1

    Outsider? Trump was BORN on the inside! Daddies contacts in the Party helped get him to where he is today.
    See also Reagan, he played the "outsider" card as well despite being also part of the Republican party machine (for less years than Trump has been though).

  195. Re: That's just great... by tepples · · Score: 1

    Which 10 inch netbook designed for 64 bit use should I buy to replace my 6 year old Dell Inspiron mini 1012?

  196. I sense, much butt hurt by Anonymous Coward · · Score: 0

    If you want to keep your 32 bit POS up to date, just fork what you want to keep.

  197. Re:That's just great... by david_thornley · · Score: 1

    Politically, he's an outsider. If he was an insider a year ago, he's pretty much changed that. He's going to get worse cooperation than Obama.

    As far as business goes, he's an insider, and a poster child for the Occupy movement's idea of the 1%. Personally, I think the establishment should have taken the Occupy movement more seriously, as it was a forerunner of Trump and Sanders. It was partly nihilistic, as it was mostly a rejection of things without a decent idea of what they wanted done about it, but that's Trump's constituency.

    --
    "When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
  198. Re:That's just great... by Anonymous Coward · · Score: 0

    Do Ubuntu or openSUSE even do any testing? Doesnt seem like it whenever I tried them. Gave up on crappy distros like that long ago

  199. I don't mind by gexacor · · Score: 1

    Looks like he is right I can't remember when I'd use x386 Linux last time if we didn't talk about some legacy servers All new installations are installed at x64 base and old ones can utilize old x386 distributions Maybe they can offer x386 images for a pay?! ))) Nah, there will be a shitstorm for that Otherwise most utilities and even control panels can use both x64 and x32 distributions, beginning from cPanel and ending with ServerSuit, so I don't mind.