Re^2: Faster creation of Arrays in XS?

This is magnitudes slower:

#!perl
use 5.006;

use strict;
use warnings;

use Benchmark qw(:all) ;
use Data::Dumper;

my $diffs = join(' ', (map { "$_:$_" } (0..49)));

    timethese( 50_000, {
       'split' => sub {
            my $LCS = [];
            for my $diff ( split ' ', $diffs ) {
                my( $x, $y ) = split ':', $diff;
                #push @$LCS, [$x,$y];
            }
        },
    });
############
~/github/LCS-XS/xt$ perl split.t 
Benchmark: timing 50000 iterations of split...
     split:  2 wallclock secs ( 2.04 usr +  0.00 sys =  2.04 CPU) @ 24
+509.80/s (n=50000)
[download]

24 kHz without push, with push 11 kHz.

Comment on Re^2: Faster creation of Arrays in XS? Download Code

Replies are listed 'Best First'.
Re^3: Faster creation of Arrays in XS? by BrowserUk (Patriarch) on Jun 21, 2015 at 21:20 UTC
The idea was to build the string in XS and give that to the caller; not build a string and then build the AoAs from the string; which of course will be slower as your still building the AoAs, but also building and the breaking down the string. Ie. Build the string instead of the AoAs in XS; and give the caller the string instead of the AoAs; thus avoiding the building of the AoAs entirely. Be sure to preallocate the string to (even much) bigger than will be required and then use set CUR to trim to the final size. 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. I'm with torvalds on this Agile (and TDD) debunked I told'em LLVM was the way to go. But did they listen!	[reply]

Replies are listed 'Best First'.

Re^3: Faster creation of Arrays in XS?
by BrowserUk (Patriarch) on Jun 21, 2015 at 21:20 UTC

The idea was to build the string in XS and give that to the caller; not build a string and then build the AoAs from the string; which of course will be slower as your still building the AoAs, but also building and the breaking down the string.

Ie. Build the string instead of the AoAs in XS; and give the caller the string instead of the AoAs; thus avoiding the building of the AoAs entirely.

Be sure to preallocate the string to (even much) bigger than will be required and then use set CUR to trim to the final size.

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.

I'm with torvalds on this Agile (and TDD) debunked I told'em LLVM was the way to go. But did they listen!

[reply]