--tsOsTdHNUZQcU9Ye
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: quoted-printable

On Fri, Jan 05, 2001 at 01:00:06AM +0100, Ignatios Souvatzis wrote:
> Hi,
>=20
> please test netbsd-current on 68LC040 and 68040V machines (well, and on
> FPU-less 68020/68030 machines, but I can do that myself) and report.
>=20
> 	"Should work now as well as on 020/030".

Preliminary benchmarks:

68060 / 50 MHz / disabled FPU:

whetstone 10:

Loops: 10, Iterations: 1, Duration: 7.360000 sec.
C Converted Double Precision Whetstones: 135.9 KIPS

FPE traps executed during this time:

5330161-5117052
213109
=2E/7.36
28955 instructions/second.

Compare to:

68060 / 50 MHz / enabled FPU:

whetstone 500:

Loops: 500, Iterations: 1, Duration: 5.360000 sec.
C Converted Double Precision Whetstones: 9.3 MIPS

scaling this with the above result gives me roughly 2 million FPU
instructions corresponding to that "9 MIPS" result. Hm. looks like
the compiler is optimizing more than the whetstone author allowed.

Results on 68030/25 tomorrow or after the weekend.

Regards,
	-is

--tsOsTdHNUZQcU9Ye
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: 2.6.3ia

iQCVAwUBOlUQt+4WPQce5FrdAQGiCwP/Zfmarjyxr+9kMDTiHm8Qccf8lq5OsqZA
nuKbswC1g6NnuO9P1HmCfAGwRsyWmIGXd/h0cgN7vHA3m3YwA1DqXVjfm4/FwKeQ
rhuBjoomwqPcukmXW+r9ifxjz9bgQVy3jMZDaE8g5hiS20zzg4xnGzwLAwPtoV4X
afgVcq6M/3I=
=11Bf
-----END PGP SIGNATURE-----

--tsOsTdHNUZQcU9Ye--