Timers and timing, was: MySQL Performance 6.0rc1

From: Chuck Swiger (cswiger_at_mac.com)
Date: 10/27/05

  • Next message: KubaTyszko: "weird hangs with 26oct current and releng_6_0" 
    Date: Thu, 27 Oct 2005 16:14:57 -0400
To: "Yuriy N. Shkandybin" <jura@networks.ru>

    Yuriy N. Shkandybin wrote:
    >>> Check gettimeofday syscall, it follows every I/O syscall, I think
    >>> our gettimeofday is tooooooo expensive, if we can directly get time from
    >>> memory, the performance will be improved further.
    >
    > It's true:
    > run next on same PC -- freebsd and linux and compare
    [ ...snippet of timing code deleted, see attachment instead... :-) ]

    FreeBSD 4.11-STABLE i386
            null function: 0.01069
                 getpid(): 0.51729
                   time(): 3.51727
           gettimeofday(): 3.48715

    FreeBSD 5.4-STABLE i386
            null function: 0.01278
                 getpid(): 0.51329
                   time(): 2.54771
           gettimeofday(): 2.54982

    Linux 2.6.5 i686
            null function: 0.01858
                 getpid(): 0.01979
                   time(): 0.44811
           gettimeofday(): 0.55776

    Darwin 8.2.0 Power Macintosh
            null function: 0.01889
                 getpid(): 0.03590
                   time(): 0.20913
           gettimeofday(): 0.17278

    SunOS 5.8 sun4u
            null function: 0.05051
                 getpid(): 1.29846
                   time(): 1.26596
           gettimeofday(): 0.29507

    [ These are representative results (in seconds); running the test three times
    per host shows the null function time value is stable to two digits, or three
    on some hosts; the other values seem to vary by less than 10%. ]

    The Intel boxes are all Intel P3, between 700MHz and 1Ghz, the Sun is a
    dual-proc E450 @ 450MHz, and the other is a Mac Mini @ 1.3Ghz, I think.

    Real numbers are are well and good, but I don't want to start yet another
    thread about microbenchmarks or statistics.

    People who are doing timers are generally looking for one of two things, a
    cron-like system which schedules periodic or one-shot events over time
    intervals of minutes, hours, days, etc (for which time() and alarm() work
    fine), or they want to deal with high-resolution time in order to see how long
    a call like a SQL query takes or update the display every 10ms, 16.67ms, and so
    forth to do realtime graphics (via gettimeofday() and usleep()/nanosleep()).

    It's clear that the Linux getpid() syscall is merely keeping the pid around
    locally, rather than doing a full context switch, and Darwin seems to be doing
    similarly, only with some locking or tracing overhead.

    It doesn't make sense to keep invoking a hardware clock from the kernel for a
    timer which is updated at a one-second resolution. Can't we just keep a static
    time_t called __now in libc for time() to return or stuff into *tloc, which
    gets updated once in a while (have the scheduler check whether fractional
    seconds has rolled over every few ticks)? 

    -- 
-Chuck

    #include <errno.h>
    #include <stdlib.h>
    #include <time.h>
    #include <sys/time.h>

    typedef void (*null_t) (void);

    void null_function(void) {}

    void gettimeofday_(void) {
       struct timeval unused;
       gettimeofday(&unused, NULL);
    }

    void time_(void) {
        time_t now = time(&now);
    }

    void getpid_(void) {
        pid_t p = getpid();
    }

    timeit(null_t f, char *name)
    {
        struct timeval start;
        struct timeval stop;
        struct timeval tmp;
        int unused;
        
        register unsigned i;
        double diff_time;
        static double null_time = 0.0;

        if (null_time == 0.0) {
            /* null function call */
            gettimeofday(&start, NULL);
            for(i=0; i<1000000; i++) {
                null_function();
            }
            gettimeofday(&stop, NULL);
            null_time = (double) (stop.tv_sec - start.tv_sec) * 1.0 + \
                (stop.tv_usec - start.tv_usec) / 1000000.0;
            printf("%20s: %0.5f\n", "null function", null_time);
        }

        gettimeofday(&start, NULL);
        for(i=0; i<1000000; i++) {
            f();
        }
        gettimeofday(&stop, NULL);
        diff_time = (double) (stop.tv_sec - start.tv_sec) * 1.0 + \
            (stop.tv_usec - start.tv_usec) / 1000000.0;
        printf("%20s: %0.5f\n", name, diff_time - null_time);
    }

    main() {
        timeit(getpid_, "getpid()");
        timeit(time_, "time()");
        timeit(gettimeofday_, "gettimeofday()");
    }

    _______________________________________________
    freebsd-current@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-current
    To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"

  • Next message: KubaTyszko: "weird hangs with 26oct current and releng_6_0"

    Relevant Pages