while ($mmap =~ m/^([^\n]*$pat[^\n]\n)/omg)
{
  print $1;
}
[download]

I found performance was abysmal: 20 seconds to search a 400K file (/usr/share/dict/words) for "foo". Some investigation (and perl -Dr) revealed that the regex engine was finding "foo" in the string very quickly, then was trying every single line preceding it to see if it matched against the remaining regex. I rewrote my simple pattern to:

while ($mmap =~ m/$pat/omg)
{
  my $pos = pos $mmap;
  # Find the beginning and end of this line
  my $first = rindex($mmap,"\n",$pos)+1;
  my $last  = index($mmap,"\n",$pos)+1;
  print substr($mmap,$first,$last-$first);
}
[download]

My question is: is there anything I could have told Perl to give it a hint about how to perform the match, such as "Hey Perl, the ^ is going to be pretty close to $pat, so just search backwards from wherever you find it"?

Here's a full copy of the original program.

#!/usr/bin/perl

use warnings;
use strict;

use Sys::Mmap;

our $pat = shift;

foreach my $a (@ARGV)
{
  my $mmap;
  if (! -f $a)
  {
    warn "'$a' is not a regular file, skipping\n";
    next;
  }
  open(F, "< $a")
    or die "Couldn't open '$a': $!\n";
  mmap($mmap, 0, PROT_READ, MAP_SHARED, F)
    or die "mmap error: $!\n";
  while ($mmap =~ m/^([^\n]*$pat[^\n]\n)/omg)
  {
    print $1;
  }
  munmap($mmap)
    or die "mmunmap error: $!\n";
  close(F)
    or die "Couldn't close '$a': $!\n";
}
[download]