Doesn't even compile:

Bareword found where operator expected at derp.pl line 7, near "// to"
    (Missing operator before to?)
"my" variable $str2 masks earlier declaration in same scope at derp.pl
+ line 8.
"my" variable $str1 masks earlier declaration in same statement at der
+p.pl line 9.
"my" variable $str2 masks earlier declaration in same statement at der
+p.pl line 9.
syntax error at derp.pl line 7, near "// to auto"
Execution of derp.pl aborted due to compilation errors.
[download]

You pontificate about the language about testing, you don't even know how to comment perl code. This is not the first time you've done this.

... (Perl >= 5.16) ...

Ignoring the fact that I don't understand the meaning of the quoted bit, and sidestepping the foreign comment operator, it compiles and does what ovedpo15 seems to want. On the whole, a bucketful out of the boat, so an upvote from me. (Now please, please don't negate this with a dozen bucketsful back into the boat!)

I know you're just trying to help; I think it's water on a duck. What ensures the downvote from me is that he could edit the node to fix the comment and mark it with (Update: fixed comment character). It's trivial to improve the monastery with these tiny bits of effort. It's what us “self-important pricks” do on a daily basis.

... water on a duck.

Sadly, I'm starting to agree.

---

Give a man a fish:  <%-{-{-{-<

Regex will be slower than the above solution.
I would code a regex version like this:

#!/usr/bin/perl
use strict; 
use warnings;
my $str1 = "helloworld";
my $str2 = "hello";

if  ($str1 ne $str2 and $str1 =~ /\Q$str2\E/) 
{ 
    print "yes, $str2 is a substring of $str1\n"; 
}

# Note: the \Q and \E escapes are not needed in
# this exact situation. These escape characters mean
# to ignore any chars in $str2 that might otherwise
# mean something to the regex engine.

# In this simple case the \E is not needed, but 
# I would recommend a \Q...\E pair.

__END__
Prints:
yes, hello is a substring of helloworld
[download]