On Tue, Apr 08, 2008 at 11:39:25AM -0700, Chris Heath wrote:
The following test program gives this error:
Bizarre copy of ARRAY in sassign at /usr/lib/perl5/5.8.8/Carp/Heavy.pm line 45.
Tested on the 5.8.x branch and on perl-current.
use Carp;
sub do_carp {
print Carp::longmess;
}
sub call_with_args {
my ($arg_hash, $func) = @_;
$func->(@{$arg_hash->{'args'}});
}
my $h = {};
# Deleting the undef makes it all work again!
my $arg_hash = {'args' => [undef]};
call_with_args($arg_hash, sub {
$arg_hash->{'args'} = [];
do_carp(sub { $h; });
});
Thanks for this bug report. The error has been seen before, but not with this
concise a test case. I can reduce it down to this:
$ cat C.pm
package C;
sub longmess_real {
package DB;
() = caller(2);
use Devel::Peek; Dump \@DB::args;
foreach(@DB::args) {
my $a = $_;
}
}
1;
__END__
$ cat 52610
sub longmess {
require C;
goto &C::longmess_real;
}
sub do_carp {
longmess;
}
sub call_with_args {
my ($arg_hash, $func) = @_;
$func->(@{$arg_hash->{'args'}});
}
my $h = {};
# Deleting the undef makes it all work again!
my $arg_hash = {'args' => [undef]};
call_with_args($arg_hash, sub {
$arg_hash->{'args'} = [];
do_carp(sub { $h; });
});
__END__
$ ./perl -Ilib 52610
SV = IV(0x10081d9a0) at 0x10081d9a8
REFCNT = 1
FLAGS = (TEMP,ROK)
RV = 0x10081dbd0
SV = PVAV(0x1008059e8) at 0x10081dbd0
REFCNT = 2
FLAGS = ()
ARRAY = 0x10050b538
FILL = 0
MAX = 3
ARYLEN = 0x0
FLAGS = ()
Elt No. 0
SV = PVAV(0x100805808) at 0x10081d738
REFCNT = 1
FLAGS = ()
ARRAY = 0x10050b4c8
FILL = 1
MAX = 3
ARYLEN = 0x0
FLAGS = (REAL)
Elt No. 0
Elt No. 1
SV = PVNV(0x100802f18) at 0x10081d828
REFCNT = 2
FLAGS = (POK,FAKE,pPOK)
IV = 0
NV = 1.97626258336499e-323
PV = 0x10050d2f8 "$h"\0
CUR = 2
LEN = 8
Bizarre copy of ARRAY in sassign at C.pm line 9.
[the line with Devel::Peek is optional]
It looks like the address of something on Perl's stack ends up being re-used
as something (else) that points into a PAD. I'm not entirely sure why, and
it's something "special" with how caller and package DB interact.
2 messages in org.perl.perl5-portersRe: [perl #52610] Bizarre copy of ARR...
