On Wed, Mar 27, 2013 at 10:43:53AM +0000, Stuart Henderson wrote: > On 2013/03/26 18:06, Creamy wrote: > > On Tue, Mar 26, 2013 at 09:00:39PM +0400, Franco Fichtner wrote: > > > On Mar 26, 2013, at 6:26 PM, Creamy <cre...@nocrater.com> wrote: > > > > Looking to the future, when are we going to drop 486 support, anyway? > > > > > > Now, that's a more interesting thing ask. > > > > How much of the hardware survives now, anyway? I mean at least the old > > Vaxen were, (and are), maintainable. 486 motherboard dies, what do you > > do? Chances are it's a multi-layer pcb, so if traces go bad within it, > > a repair is going to be almost impossible. > > Some of the 486-class embedded boards are quite solid hardware and > not likely to die anytime soon.
Agreed, but my experience was that those of us who were in the habbit of purchasing kit with decent build quality, in preference to the latest 'features', back in the day, were also the ones who tended to sell and replace it. The old boards that people are trying to keep in operation now, ironically, tend to be the rubbish ones. At least that's my experience, but others' may differ. > What advantage would there be to dropping 486 support anyway? In itself, perhaps not much, I very much doubt whether we'd see any use from being able to build the default distribution with 586+ compiler options, for example. However, on a practical level, if we took the decision to kill 486 support, we could, in effect, loose 99% of the ISA-related code, as excluding a few specialised pieces of hardware, (which OpenBSD doesn't support, and probably never will), ISA pretty much died by the 586 era, (as did VL-bus). As I pointed out in a previous post, we still have a Y2K workaround in the clock code, which is pointless on AMD64, anyway, and just a hang-over from taking the code straight from the i386 port. How many 586+ machines needed this workaround, anyway? Maybe some of the original P60 systems did, I honestly don't remember, but the number would be very small, if it is >0. I'm not claiming that dropping 486 support is the right thing to do right now, but I think it should be in our minds as an option. Look to the future, at what point did booting from CD-ROM become standard in BIOSes? I only used a few select brands of kit back then, generally the higher quality ones, so maybe I am off the mark here, but I never remember seeing a second-generation Pentium, (I.E. P75+), that lacked this feature. So, maybe we could eventually loose the need for boot floppy support, and we could overhaul the instructions in the official disc set, and make better use of those pages explaining the floppy install, which nobody uses, for something more useful. We could probably also loose the force-CHS code in the bootloader, which would save some very precious space, and allow us to use it for something more useful. For example, I'm obviously not using that on AMD64, so I added the feature to force booting of partition 3, regardless of which is flagged as active. Why? I was messing around with some assembler stuff on the raw hardware, (effectively writing my own OS, if you want to call it that, but all it did was print some text using the BIOS, it's a long story why I'm doing this, I'll tell the interested parties, (I.E. nobody), some other time), and I had flagged partition 0 as active, and had to boot from the 5.2 CD to set it back, as my 'os' has no fdisk program, (or any programs for the foreseeable future). However, it struck me that somebody dual-booting with Windows would probably have the same problem, because as far as I know you can't set an arbitrary partition active with fdisk in Windows, but I really don't know or care, because I don't use it. So, you see, killing 486 support might be no advantage in itself, but it opens up possibilities further down the line, that won't exist all the time we're dragging all this old stuff along with us. -- Creamy
