in reply to Re^2: Question about curious performance of if...elsif block
in thread Question about curious performance of if...elsif block

Even with 36 cases, I still see nothing like the 5 second difference you apparently are:

#! perl -slw
use strict;
use Time::HiRes qw[ time ];

sub test1 {
    my $a = $_[0];

#   return;
    if ($a==1) {}
    elsif ($a==2) {}
    elsif ($a==3) {}
    elsif ($a==4) {}
    elsif ($a==5) {}
    elsif ($a==6) {}
    elsif ($a==7) {}
    elsif ($a==8) {}
    elsif ($a==9) {}
    elsif ($a==10) {}
    elsif ($a==11) {}
    elsif ($a==12) {}
    elsif ($a==13) {}
    elsif ($a==14) {}
    elsif ($a==15) {}
    elsif ($a==16) {}
    elsif ($a==17) {}
    elsif ($a==18) {}
    elsif ($a==19) {}
    elsif ($a==20) {}
    elsif ($a==21) {}
    elsif ($a==22) {}
    elsif ($a==23) {}
    elsif ($a==24) {}
    elsif ($a==25) {}
    elsif ($a==26) {}
    elsif ($a==27) {}
    elsif ($a==28) {}
    elsif ($a==29) {}
    elsif ($a==30) {}
    elsif ($a==31) {}
    elsif ($a==32) {}
    elsif ($a==33) {}
    elsif ($a==34) {}
    elsif ($a==35) {}
    elsif ($a==36) {}
}

sub test2 {
    my $a = $_[0];

    return ;
    if ($a==1) {}
    elsif ($a==2) {}
    elsif ($a==3) {}
    elsif ($a==4) {}
    elsif ($a==5) {}
    elsif ($a==6) {}
    elsif ($a==7) {}
    elsif ($a==8) {}
    elsif ($a==9) {}
    elsif ($a==10) {}
    elsif ($a==11) {}
    elsif ($a==12) {}
    elsif ($a==13) {}
    elsif ($a==14) {}
    elsif ($a==15) {}
    elsif ($a==16) {}
    elsif ($a==17) {}
    elsif ($a==18) {}
    elsif ($a==19) {}
    elsif ($a==20) {}
    elsif ($a==21) {}
    elsif ($a==22) {}
    elsif ($a==23) {}
    elsif ($a==24) {}
    elsif ($a==25) {}
    elsif ($a==26) {}
    elsif ($a==27) {}
    elsif ($a==28) {}
    elsif ($a==29) {}
    elsif ($a==30) {}
    elsif ($a==31) {}
    elsif ($a==32) {}
    elsif ($a==33) {}
    elsif ($a==34) {}
    elsif ($a==35) {}
    elsif ($a==36) {}
}

sub test3 {
    return if $_[0] > 9;
    my $a = $_[0];

    if ($a==1) {}
    elsif ($a==2) {}
    elsif ($a==3) {}
    elsif ($a==4) {}
    elsif ($a==5) {}
    elsif ($a==6) {}
    elsif ($a==7) {}
    elsif ($a==8) {}
    elsif ($a==9) {}
    elsif ($a==10) {}
    elsif ($a==11) {}
    elsif ($a==12) {}
    elsif ($a==13) {}
    elsif ($a==14) {}
    elsif ($a==15) {}
    elsif ($a==16) {}
    elsif ($a==17) {}
    elsif ($a==18) {}
    elsif ($a==19) {}
    elsif ($a==20) {}
    elsif ($a==21) {}
    elsif ($a==22) {}
    elsif ($a==23) {}
    elsif ($a==24) {}
    elsif ($a==25) {}
    elsif ($a==26) {}
    elsif ($a==27) {}
    elsif ($a==28) {}
    elsif ($a==29) {}
    elsif ($a==30) {}
    elsif ($a==31) {}
    elsif ($a==32) {}
    elsif ($a==33) {}
    elsif ($a==34) {}
    elsif ($a==35) {}
    elsif ($a==36) {}
}

my @dispatch = (
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
    sub {},
);

sub test4 {
    return if $_[0] > 9;
    $dispatch[ $_[0] ]->( $_[0] );
}

my $start = time;
test1( $_ ) for 1 .. 1e6;
printf "test1 took: %.5f seconds\n", time - $start;

$start = time;
test2( $_ ) for 1 .. 1e6;
printf "test2 took: %.5f seconds\n", time - $start;

$start = time;
test3( $_ ) for 1 .. 1e6;
printf "test3 took: %.5f seconds\n", time - $start;

$start = time;
test4( $_ ) for 1 .. 1e6;
printf "test4 took: %.5f seconds\n", time - $start;

__END__
C:\test>junk37
test1 took: 1.70500 seconds
test2 took: 0.30789 seconds
test3 took: 0.29240 seconds
test4 took: 0.29207 seconds

[download]
With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
"Science is about questioning the status quo. Questioning authority".
In the absence of evidence, opinion is indistinguishable from prejudice.

The start of some sanity?

Replies are listed 'Best First'.
Re^4: Question about curious performance of if...elsif block
by markseger (Beadle) on Jan 16, 2012 at 19:34 UTC
    ok, one more time. try embedding my subruotine in a loop reading /proc/pid/status and measure it. THAT's my point. In a small program the timings are as expected. My question is why the unexpected performance when reading /proc?

    Did you try my second script on your system? Try that without the return statement.

    -mark

[reply]
      Did you try my second script on your system?

      No. Because it would not run here.

      ok, one more time. try embedding my subruotine in a loop reading /proc/pid/status and measure it. THAT's my point.
      Then it is fairly obvious that the reason for the differences you are concerned about have nothing to do with that subroutine and every thing to do with the code that is calling it. Ie. All your concentration on the subroutine is a complete red herring.

      It seems fairly clear to me that you are misinterpreting the data you are reading. What timing do you see if you comment out the subroutine call entirely?

      But don't bother with "ok, one more time.". Cos I almost certainly wouldn't understand it.

      With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
      Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
      "Science is about questioning the status quo. Questioning authority".
      In the absence of evidence, opinion is indistinguishable from prejudice.

      The start of some sanity?
[reply]
        > No. Because it would not run here. ahh, so I guess you're not running linux.

        no problems. here's a run with the subroutine call commented out: 

        
[root@poker collectl]# time ./test.pl
real    0m14.464s
user    0m5.089s
sys     0m9.087s

        [download]
        now with the call executing BUT the return statement executing first thing in the subroutine: 
        [root@poker collectl]# time ./test.pl
real    0m19.508s
user    0m10.461s
sys     0m8.853s

        [download]
        so clearly the overhead in calling the subroutine 1M times is about 5-1/2 seconds. Now I'll comment out the early return in the subroutine in the subroutine and run it again: 
        [root@poker collectl]# time ./test.pl
real    0m25.648s
user    0m16.350s
sys     0m9.123s

        [download]
        This says to me the if...elsif... takes about 6 seconds and I can't fathom why! Could there be some weird interaction going on between perl/linux and the /proc filesystem?

        As I said in my base note I'm not an innards kind of a guy but it feels like something that is not going to show up in every perl script. Rather if feels like there is some kind of weird interraction.

        and speaking of environment this is rhel5.3 and perl 5.8.8. I just tried the code again on rhel6.2 in a VM with perl 5.10, and the differences between the last 2 runs are: 

        real    0m7.359s
user    0m2.840s
sys     0m4.481s

real    0m9.782s
user    0m5.082s
sys     0m4.615s

real    0m11.927s
user    0m7.266s
sys     0m4.604s

        [download]
        so in this case both the subroutine calls and the if...elsif... block are only taking a little over 2 seconds each. This says to me there is something more efficient going on with 5.10.

        Perhaps this is less of a big deal as 2 extra seconds is getting less of a concern, but I still an curious about what exactly is going on.

        -mark

[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom