Re: STREAM Bandwidth results

From: Michael Kukat (
Date: Wed Nov 13 2002 - 08:38:24 CST

  • Next message: Michael Kukat: "Re: STREAM Bandwidth results"

    Hi !

    On Wed, 13 Nov 2002, John D McCalpin wrote:
    > Which timer routine did you link with?

    second_cpu.c, as second_wall.c didn't link in the first try. But i found the
    problem fast now, i was a bit loaded the first test. second_wall.c declares
    a function second_ instead of second. I just removed the underscore and here
    we go with a bit better results i think:

    Again, ThinkPad A31, NetBSD 1.6, 1.6GHz P4

    This system uses 8 bytes per DOUBLE PRECISION word.
    Array size = 2000000, Offset = 0
    Total memory required = 45.8 MB.
    Each test is run 10 times, but only
    the *best* time for each is used.
    Your clock granularity/precision appears to be 10 microseconds.
    Each test below will take on the order of 23635 microseconds.
       (= 2363 clock ticks)
    Increase the size of the arrays if this shows that
    you are not getting at least 20 clock ticks per test.
    WARNING -- The above is only a rough guideline.
    For best results, please be sure you know the
    precision of your system timer.
    Function Rate (MB/s) RMS time Min time Max time
    Copy: 1019.3029 0.0520 0.0314 0.1334
    Scale: 1010.8317 0.0516 0.0317 0.1320
    Add: 1176.4404 0.0771 0.0408 0.2106
    Triad: 1173.5089 0.0604 0.0409 0.1407


    --     Home Powered by: (Net|Open|Free)BSD IRIX NonStop-UX
    Solaris AIX HP-UX Tru64 MUNIX Ultrix VMS SINIX Dolphin_Unix OpenStep MacOS A/UX

    This archive was generated by hypermail 2.1.4 : Wed Jan 08 2003 - 09:17:32 CST