execl benchmark is not 64-bit ready, and crashes when 0 is passed
    to execl(2) instead of NULL.

Fix: 

Since unixbench is unmaintained, I propose this patch be added
as a local FreeBSD ports patch.--LtKMN5XgQz8EQYR77CggR1kAe3S787F09uX5e9GcM7icvlVh
Content-Type: text/plain; name="file.shar"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment; filename="file.shar"

# This is a shell archive.  Save it in a file, remove anything before
# this line, and then unpack it by entering "sh file".  Note, it may
# create directories; files and directories will be owned by you and
# have default permissions.
#
# This archive contains:
#
#       unixbench_execl.patch
#
echo x - unixbench_execl.patch
sed 's/^X//' >unixbench_execl.patch << 'END-of-unixbench_execl.patch'
X--- unixbench-4.1.0/src/execl.c        Wed Jul 28 21:46:58 1999
X+++ unixbench-4.1.0-new/src/execl.c    Tue Feb 27 10:06:48 2007
X@@ -88,7 +88,7 @@
X               fprintf(stderr, "%lu loops\n", iter);
X               exit(0);
X               }
X-      execl(fullpath, fullpath, "0", dur_str, count_str, start_str, 0);
X+      execl(fullpath, fullpath, "0", dur_str, count_str, start_str, NULL);
X       printf("Exec failed at iteration %lu\n", iter);
X       perror("Reason");
X       exit(1);
END-of-unixbench_execl.patch
exit
How-To-Repeat: 	
    Run execl benchmark on amd64, observe the error message.