Early-to-middle IBM-PC history (was: End-user uses for x86-64)

Ben Scott dragonhawk at gmail.com
Mon Feb 19 13:32:54 EST 2007


On 2/19/07, Michael ODonnell <michael.odonnell at comcast.net> wrote:
>> ... inability of the 286 to switch out of protected mode once it got in.
>
> Aww, shucks!  no need to let a little thing like that get in your way:
>
>    http://blogs.msdn.com/larryosterman/archive/2005/02/08/369243.aspx

/me runs away screaming

  Forcing a triple fault to *reset the CPU* to get out of protected
mode?  And then detecting that you're coming off a reset in boot code
and switching back to the OS (still in RAM) instead of resetting the
whole machine?  Ye Gods, did that actually work?  Well, I guess it
must have, but... *shudder*

> The huge body of software that eventually ran on the IBM PeeCee stands
> as a tribute to the creativity and perseverance of the programmers who
> overcame all the pointlessly idiotic obstacles placed in front of them ...

  I've always held programmers who get stuff done on such bitty boxes
with a mixture of awe and revulsion.  Awe, because they are both
incredibly competent and incredibly creative, eking every last drop of
performance out of small hardware (to do some truly impressive
things), and also finding solutions (like the above) when it seems
like no solution would be possible.

  Revulsion, because the need/desire to support lasting hacks like
that cripple progress in the field.  No, we can't fix the incredibly
brain-damaged PC architecture, because there's too much software that
will break if we do.  So instead we run everything on a platform
that's a mass of scar  tissue and band-aids.  Eeek.  :)

> ... made code run despite (what seemed to most of us to be) the
> best efforts of Intel/Microsoft.

  You forgot IBM, who perpetrated this alleged design in the first place.  :)

-- Ben


More information about the gnhlug-discuss mailing list