Thanks. I knew I was doing something wrong. Here's another attempt.

#!/usr/bin/perl -w
use strict;

use Benchmark;

my $file = $0;
open IN, $file or die "$file: $!\n";

sub joinit { seek(IN, 0, 0); my $content = join '', <IN>; }

sub dollarslash { seek(IN, 0, 0); my $content = do { local $/; <IN> } 
+}

timethese(100_000, {join => \&joinit,
                    slash => \&dollarslash} );
[download]

Which gives these (much more believable) results:

Benchmark: timing 100000 iterations of join, slash...
      join: 12 wallclock secs ( 9.98 usr +  1.33 sys = 11.31 CPU)
     slash:  6 wallclock secs ( 3.79 usr +  1.53 sys =  5.32 CPU)
[download]

"Perl makes the fun jobs fun
and the boring jobs bearable" - me