Hi,

I've done some work on m68k FPE (floating point emulation).  It is
based on Gordon Ross' arch/m68k/fpe/fpu_emulate.c and Theo Deraadt's
Sparc FPE (derived from Chris Torek's original work, I think?).

Currently implemented (but not thoroughly tested... yet) instructions
are as follows:

Type=0: FMOVE (mem->FPr), FSQRT, FABS, FNEG, FDIV, FADD, FMUL, FSUB,
FCMP, FTST, FMOVE (FPr->mem), FMOVEM (FPr)
Type=1: FDBcc, FScc, FTRAPcc,
Type=2: FBcc (word, incl. FNOP)
Type=3: FBcc (long)

All data types except P (packed BCD) are supported.

I'd like you to try the code.  The gzipped tar source is located at:

	ftp://remus.rutgers.edu/pub/NetBSD/test_kernels/fpe.100595.tar.gz

and there's a README (written with mac68k port users in mind):

	ftp://remus.rutgers.edu/pub/NetBSD/test_kernels/README.fpe

Please direct any questions, complaints, suggestions, etc, to me at
kenn@remus.rutgers.edu.

Thanks for your help,

ken