Re: *** next draft - press release ***

Albert D. Cahalan (acahalan@cs.uml.edu)
Fri, 15 Jan 1999 15:12:54 -0500 (EST)


Alexander Viro writes:
> On Fri, 15 Jan 1999, Albert D. Cahalan wrote:

>>>> new life to older 486-based computers. Support is widely available
>>>
>>> Here we should change that 486-based to 386-based ... don't limit our
>>> suporting to the lower end ... someone is still using 386 :)
>>
>> Hopefully that will be an option left for hackers only.
>> 386 support cuts into performance and SMP support.
>
> How so? When did you look at the code last time?

Bad WP bit on a 386 - need an extra test at least.
No invlpg instruction - must invalidate the whole cache.
No byte swap instruction - must shift and mask and add/or.
IRQ 13 is used for both SMP and FPU - trouble AFAIK.

The second problem at least could be mostly fixed with
self-modifying kernel code. Nobody cared enough though,
since the 386 is so uncommon.

Most 386 hardware doesn't even have enough memory for normal
Linux distributions. Linux in 2 MB takes some skill.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/