Anonymous Monk,
Unfortunately, you didn't provide the stuff that manipulates $val - part of the "slowness" could be there. When you want to evaluate how a tweak has impacted performance - look into Benchmark. The thing to remember here is to go through many iterations to remove "flukes", vary your data as code behaves differently based off input, and try to test on a system at rest so it won't be influenced by other running programs. There is also Devel::DProf.

You also haven't indicated what $val is. In your comment, you say that it generates 4 arrays. If it is a very large scalar, it may be useful to pass the argument as a reference (see References Tutorial & References quick reference) instead of copying the whole thing which is slow.

Additionally, instead of returning the entire hash, you may just want to return a reference to the hash. That would change the my %hash = build_hash($val); to something like my $hash_ref = build_hash($val); You could also not use a return value at all, but build the hash inside the subroutine:

#!/usr/bin/perl -w
use strict;
my %hash;
my $val = "foo bar";
build_hash(\%hash,$val);

sub build_hash {
    my ($hash_ref, $val) = @_;
    my @vals = split /\s+/ , $val;
    foreach (@vals) {
        $hash_ref->{$_} = 1;
    }
}
[download]

I am sure there is more optimization that could be done, but this should be a start. The other answers should become obvious after reading the references tutorials I provided.

Cheers - L~R