Post by @kenshirriff@oldbytes.space

2026-02-14 17:32 UTC

In 1980, Intel released the 8087 floating-point chip, making math much faster. I&#39;m reverse-engineering this chip, 46 years later. Most of its instructions are implemented in microcode, but some are implemented in hardware. Let&#39;s look at the circuitry that decodes instructions and decides what to do...

Replies (5)

@kenshirriff@oldbytes.space 2026-02-14 17:38

Like many processors, the 8087 uses a "Programmable Logic Array" (PLA) to decode the bit patterns in instructions. A PLA consists of grids of transistors, with some of the transistors missing. The pattern of transistors controls the PLA's function. The grid is a dense and flexible way of implementing this logic.

Open ##1442659
@renardboy@mastodon.social 2026-02-14 17:37

@kenshirriff this is extremely cool, good luck 😁

Open ##1442670
@DianeBruce@bsd.network 2026-02-14 17:52

@kenshirriff Cool.I remember when they had to sell them in a matched set because of a NMI bug in the 8088. We discovered that bug the hard way at Dynalogic. ;)

Open ##1442676
@raymaccarthy@mastodon.ie 2026-02-14 19:36

@kenshirriff Did it have much speed up for 16 bit integer maths, or just Floating point (80 bit or 64 bit?)?

Open ##1442677
@TomF@mastodon.gamedev.place 2026-02-15 06:34

@kenshirriff That is a honking big register stack. I keep forgetting it's 80 bits per reg!

Open ##1442683