Looks like you can do this quicker with map:

Perl Code:

#! /usr/bin/perl

use strict ;
use warnings ;
$|++ ;

use Benchmark qw( :all ) ;

# Construct the test string.
our @arr ;
for ( 1..100_000 ) { push @arr, "foo, bar, zoot, blarg,\n" }

cmpthese( 10_000_000, { 'Mine' => \&mine, 'Yours' => \&yours } ) ;

sub mine
{
    local @arr ;
    return map { split /,\s*|\n/ } @arr ;
}

sub yours { # Roughly.
    local @arr ;
    my @word_list ;
    while ( @arr ) {
        chomp;
        push @word_list, split ", ";
    }
    \@word_list;
}
[download]

Results:

Benchmark: timing 10000000 iterations of Mine, Yours...
      Mine:  9 wallclock secs ( 8.82 usr +  0.02 sys =  8.84 CPU) @ 11
+31221.72/s (n=10000000)
     Yours: 25 wallclock secs (21.95 usr +  0.04 sys = 21.99 CPU) @ 45
+4752.16/s (n=10000000)
           Rate Yours  Mine
Yours  454752/s    --  -60%
Mine  1131222/s  149%    --
[download]

_______________
DamnDirtyApe

Those who know that they are profound strive for clarity. Those who
would like to seem profound to the crowd strive for obscurity.
            --Friedrich Nietzsche