I think demerphq mentioned it already, but index() returns 0 if the 0-th character matches. It returns -1 if nothing in the string matches ( unless someone has played with $[ ).

I also had a couple of other thoughts. Surely the Inline::C solutions are faster, but I'm still trying to wring performance out of pure Perl. 1. Why not use $s2 through the reference as well? and 2. What about use bytes (); and then explicitly using bytes::substr and bytes::index instead of doing the import?

This code tests both of those together, for a whopping improvement over my solutions using the import and just using $s1 by reference. 

sub mrm_6 {
    # from mrn_5, testing bytes::misc explicitly instead of importing
    # also in-place using of $s2
    my ( $s1, $s2 ) = @_;
    use bytes ();
    my $pos = 0;
    while ( -1 < ( $pos = bytes::index( $$s1, '\0', $pos ) ) ) {
        bytes::substr( $$s1, $pos, 1, bytes::substr( $$s2, $pos, 1 ) )
+;
    }
}

[download]

The results are impressive for such simple changes:

Strawberry Perl 5.8.8 on WinXP, AthlonXP 2400+, 1Gig 

                    Rate ikegami_s
ikegami_s         37.0/s        --
avar               188/s      409%
avar2              192/s      418%
avar2_pos          282/s      663%
ikegami_tr         342/s      825%
moritz             783/s     2018%
avar2_pos_inplace 1640/s     4338%
mrm_3             2436/s     6491%
mrm_4             2449/s     6524%
mrm_5             2459/s     6553%
mrm_1             2517/s     6709%
mrm_6             3372/s     9023%

[download]
cygperl 5.8.6 on the same machine as above 
                    Rate ikegami_s
ikegami_s         37.9/s        --
avar               285/s      650%
avar2              310/s      718%
ikegami_tr         316/s      733%
avar2_pos          709/s     1769%
moritz            1002/s     2543%
mrm_5             3631/s     9474%
mrm_3             3841/s    10027%
mrm_4             3913/s    10219%
mrm_1             4044/s    10562%
avar2_pos_inplace 4393/s    11484%
mrm_6             6237/s    16345%

[download]
perl 5.8.7 on Mandriva Linux 2006 Athlon 1000, 512MB RAM 
                    Rate ikegami_s  
ikegami_s         17.2/s        --  
avar               168/s      876%
avar2              187/s      983%
ikegami_tr         205/s     1091%
avar2_pos          307/s     1684%
moritz             620/s     3499%
mrm_4             1088/s     6218%
avar2_pos_inplace 1184/s     6775%
mrm_3             1184/s     6775%
mrm_5             1224/s     7006%
mrm_1             1240/s     7101%
mrm_6             1921/s    11052%

[download]
... and perhaps a sign of good things to come: perl 5.9.5 on the above-mentioned Linux box 
                    Rate ikegami_s
ikegami_s         13.3/s        --
avar               172/s     1189%
ikegami_tr         176/s     1220%
avar2              182/s     1267%
avar2_pos          258/s     1837%
moritz             412/s     2987%
avar2_pos_inplace  776/s     5722%
mrm_3              780/s     5749%
mrm_1              785/s     5788%
mrm_5              798/s     5882%
mrm_4              806/s     5942%
mrm_6             2683/s    20026%

[download]
ActivePerl 5.8.0 on the Windows box didn't fare so well under this one

I killed the benchmark for ActivePerl at 7 minutes CPU time, over 23 million page faults, and over 175MB of memory usage. I guess there's probably a bug in bytes or in perl in that build that's causing the thrashing.

In reply to Re^4: Challenge: CPU-optimized byte-wise or-equals (for a meter of beer) by mr_mischief
in thread Challenge: CPU-optimized byte-wise or-equals (for a meter of beer) by dragonchild

Title:
Use:  <p> text here (a paragraph) </p>
and:  <code> code here </code>
to format your post, it's "PerlMonks-approved HTML":


  • Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!
  • Titles consisting of a single word are discouraged, and in most cases are disallowed outright.
  • Read Where should I post X? if you're not absolutely sure you're posting in the right place.
  • Please read these before you post! —
  • Posts may use any of the Perl Monks Approved HTML tags:
    a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr
  • You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)
            For:     Use:
    & &amp;
    < &lt;
    > &gt;
    [ &#91;
    ] &#93;
  • Link using PerlMonks shortcuts! What shortcuts can I use for linking?
  • See Writeup Formatting Tips and other pages linked from there for more info.