Performance issues

--------------ms000708000007010201010901
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit

I'm running 5.8.8 on 64-bit linux, and I have a nasty performance issue
that does not happen with the same perl program on 32-bit linux.

A small sample that exhibits the problem is:

use strict;

use Time::HiRes qw(time);
use HTTP::Request;

my %l;

    my $now = time;
    for (my $i = 0; $i < 30000; $i++) {
        my $dt = HTTP::Request->new(GET => "http://localhost/");

        $l{$i} = $dt;
        if ($i % 1000 == 999) {
            print time - $now,"\n";
            $now = time;
        }
    }

The first few thousand iterations go quite rapidly (68ms for the first
thousand, 94 ms for the second). By the twentieth line of output, it is
take 2 seconds per thousand....  On a 32 bit platform, the time remains
(roughly) the same for each block of a thousand.

It doesn't seem to depend on the exact object that I create. My
suspicion is that the memory allocation is the issue, but I don't know.
Any thoughts on how to track this down and/or fix it?

Thanks

Philip

--------------ms000708000007010201010901
Content-Type: application/x-pkcs7-signature; name="smime.p7s"
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="smime.p7s"
Content-Description: S/MIME Cryptographic Signature
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--------------ms000708000007010201010901--