:: "David S. Miller" <davem@redhat.com>
>I think we should really investigate two issues before we really
>take your patch seriously:
...
>3) The effect of user/kernel time used for real life applications
>(sorry, I know you want improved SpecFP/INT numbers on Alpha
>but something which screws performance for what people mostly
>use a general purpose OS like Linus for is simply unacceptable)
>For example, what does "time make -s vmlinux" give with and
>without page coloring?
>
Your real life is my arbitrary benchmark and vice versa. If I knew
the identity of the objective performance test, I would have used and
cited it. It's true that I "want improved SpecFP/INT numbers on
Alpha", and I'm not ashamed. ;-) It is my opinion that contemporary
benchmarks are carefully designed and speak to the needs of some
users.
At the same time, I'm sensitive to the possibility that one size
doesn't fit all. In soliciting feedback, I'm looking to develop an
accurate characterization of my patch, and if I get lucky, to
improve it. I hope that, accurately described, the patch will be a
useful choice to some and not a screw of anybody.
Finally, here are my "time make -s vmlinux" results, which are much
like yours:
DS20 System (2 processors)
666 MHz 21264A (EV67)
8MB secondary cache each CPU
1 GB RAM
Unmodified 2.3.99pre3 kernel, median elapsed time of 5 runs:
269.95user 20.63system 4:42.11elapsed 103%CPU make -s boot
272.53user 20.22system 2:35.02elapsed 188%CPU -j 2
275.09user 20.96system 2:30.76elapsed 196%CPU -j 3
275.63user 21.05system 2:44.37elapsed 180%CPU -j 4
with coloring patch, median elapsed time of 5 runs:
270.63user 60.65system 5:18.52elapsed 104%CPU make -s boot
273.17user 57.86system 2:56.18elapsed 187%CPU -j 2
275.59user 59.07system 2:49.53elapsed 197%CPU -j 3
275.12user 59.20system 3:04.48elapsed 181%CPU -j 4
Regards,
\Joe [Joseph.Martin@Compaq.com Joseph A. Martin
Compaq Computer Corporation
200 Forest Street MRO1-2/K20
(508) 467-2100 Marlboro, MA 01752 U.S.A.
]
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Apr 15 2000 - 21:00:19 EST