http://qs321.pair.com?node_id=703316


in reply to Re^3: what would you like to see in perl5.12?
in thread what would you like to see in perl5.12?

Oh, you meant runtime performance! Sorry, I completely failed to recognize that. I thought you meant the clunkiness of Perl 5 subroutine argument processing: 

      my ($foo, $bar, $baz);          # oops, forgot " = @_"
      my ($foo, $bar, $baz) = shift;  # oops, used shift() instead of 
+"@_"
      my $single_arg = @_;            # oops, will usually be 1

[download]

Perl6::Declare lets you declare subroutine/methods using Perl 6 signatures, which are lovely. I heard this turns out to actually be faster than Perl 5 calling conventions, but I can't say I fully understand why this is so.

Replies are listed 'Best First'.
Re^5: what would you like to see in perl5.12?
by BrowserUk (Patriarch) on Aug 09, 2008 at 20:57 UTC
    Oh, you meant runtime performance! Sorry, I completely failed to recognize that.

    Really? How very strange.

    I heard this turns out to actually be faster than Perl 5 calling conventions,

    That's worth a look. Is it packaged for install anywhere, or only available from the repositories?

    Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
    "Science is about questioning the status quo. Questioning authority".
    In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]
      The parse failure was wholly mine... Perl6::Signature and Devel::Declare are on CPAN, but you may need a newer version of the latter. Like I said, experimental.
[reply]

        Sorry if I missed the correspondence between Perl6::Declare and Devel::Declare & Perl6::Signatures. If indeed there is any?

        Perl6::Declare lets you declare subroutine/methods using Perl 6 signatures, which are lovely. I heard this turns out to actually be faster than Perl 5 calling conventions,

        Devel::Declare talks about "declarator magic" and Perl6::Signature contains a Parse::RecDescent-based parser for signatures.".

        If anything that uses P::RD turns out to be faster than my( $x, $y, $z ) = @_;, it will indeed be "magic". Oh! and I'll eat my hat! (I'll have to buy a hat to eat, but I'm sure I can find something tasty at a thrift shop:)

        To clarify, what I meant by my request for less overhead, was a reduction in the internal (C-level) overhead associated with calling a perl subroutine, that results in this kind of comparison of similar subroutines coded in Perl and another bytecode compiled, interpreted language: Java.

        Are Perl's function calls slow?

        The archetypal test of function call performance is the Ackermann Function. Go here to see the list of other languages that out-perform Perl for function call performance. Many of those are fully pre-compiled languages. Many are not.

        For comparison purposes, look at straight forward implementations in Perl, with the same in Java.

        Perl: 

                
    P:\test\MJD>ack1 9
    Ack(3,9): 4093
    78.843 83.671 0 0

        [download]

        Java: 

        P:\test\MJD>timethis "c:javac ackermann.java && java ackermann 9"

TimeThis :  Command Line :  c:javac ackermann.java && java ackermann 9
TimeThis :    Start Time :  Fri Sep 02 22:01:28 2005

Ack(3,9): 4093

TimeThis :  Command Line :  c:javac ackermann.java && java ackermann 9
TimeThis :    Start Time :  Fri Sep 02 22:01:28 2005
TimeThis :      End Time :  Fri Sep 02 22:01:29 2005
TimeThis :  Elapsed Time :  00:00:01.031

        [download]

        I've included the compilation and runtime for Java to even the score a little. JIT was not enabled.

        So 83.6 seconds for Perl, and 1.031 seconds for Java!

        Conclusion: Perl's function calls are slow.

        Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
        "Science is about questioning the status quo. Questioning authority".
        In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]
[d/l]
[select]

In Section Meditations