From owner-freebsd-hackers  Wed Aug 21 07:32:26 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id HAA19998
          for hackers-outgoing; Wed, 21 Aug 1996 07:32:26 -0700 (PDT)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id HAA19993
          for <freebsd-hackers@freebsd.org>; Wed, 21 Aug 1996 07:32:19 -0700 (PDT)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id AAA17293; Thu, 22 Aug 1996 00:12:38 +1000
Date: Thu, 22 Aug 1996 00:12:38 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199608211412.AAA17293@godzilla.zeta.org.au>
To: joerg_wunsch@uriah.heep.sax.de, kuku@gilberto.physik.rwth-aachen.de
Subject: Re: max math performance - how?
Cc: freebsd-hackers@freebsd.org
Sender: owner-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>Another idea comes into mind: (don't know 386 architecture so well though):
>Assumed you have an FPU inlined libm. Would it be possible to
>run the emulator through a trap (emulator trap instruction) or
>intercepting the illegal instruction trap, checking the stack for a valid
>FPU instruction and pass instruction execution to the emulator
>in that case.

That's what happens now.  The emulator sees all FPU instructions.
Unfortunately, it doesn't handle any of the instructions for functions.

Bruce