Re: [PATCH] Simple Page Coloring [last edition] (2.3.99pre5 diffs)

From: Joseph A. Martin (jmartin@linux08.mro.dec.com)
Date: Wed Apr 19 2000 - 12:54:51 EST


>How did things go with my patch that I sent you? Every time
>I send you a copy of my version, I don't get any feedback yet
>I try to give you feedback on your work.

Here are some results. I'm relying on some other busy person for
SPEC, so I don't have anything current there.

My test system has two EV6 processors (64KB on-chip I&D caches) with
8MB L2 caches and 1GB of memory. I set up the cache sizes "by hand"
rather than porting the sizing code.

These are the runs with median elapsed time out of a sample of five on
a fresh system in single user mode. The two "davem" variations are
with and without USE_VIRTUAL_COLORING.

make -s -j 2 boot:
                          user system elapsed
2.3.99pre5 stock 273.04 20.28 2:35.21 188%CPU
simple coloring 272.14 41.55 2:47.05 187%CPU
davem coloring (virt) 271.52 21.90 2:35.38 188%CPU
davem coloring (novirt) 270.22 21.45 2:34.49 188%CPU

LINPACK 1000d.f:
                          user system elapsed
2.3.99pre5 stock 2.22 0.01 0:02.23 100%CPU
simple coloring 1.91 0.02 0:01.94 99%CPU
davem coloring (virt) 1.92 0.02 0:01.94 99%CPU
davem coloring (novirt) 1.91 0.02 0:01.93 100%CPU

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 : Sun Apr 23 2000 - 21:00:15 EST