From @masak

<masak> rakudo​: sub foo { @​_[0] = "bah" }; my $a = "!"; foo($a); say $a
<p6eval> rakudo 7ef386​: !␤
<TimToady> i'd've expected it to blow up
* masak submits rakudobug
<masak> TimToady​: so in the default signature, @​_ isn't writeable?
<TimToady> no, it's different from P5 in that
<masak> ok.
<TimToady> but most P5 programs aren't modifying their parameters
<TimToady> and we didn't want to pay the might-be-lvalue penalty anymore
<TimToady> that was decided in about 2003 or so
<masak> nod.

From @kyleha

This is an automatically generated mail to inform you that tests are now available in t/spec/S06-signature/unspecified.t

commit feadec2cec905db9b25b782725463e449c7a0006
Author​: kyle <kyle@​c213334d-75ef-0310-aa23-eaa082d1ae64>
Date​: Mon Jan 11 21​:15​:44 2010 +0000

  [t/spec] Test for RT 71112​: Cannot assign to readonly variable.
 
  git-svn-id​: http://svn.pugscode.org/pugs@&#8203;29524 c213334d-75ef-0310-aa23-eaa082d1ae64

diff --git a/t/spec/S06-signature/unspecified.t b/t/spec/S06-signature/unspecified.t
index 6bf0ed9..d9133f6 100644
--- a/t/spec/S06-signature/unspecified.t
+++ b/t/spec/S06-signature/unspecified.t
@@ -1,7 +1,7 @@
 use v6;
 use Test;
 
-plan 15;
+plan *;
 
 # L<S06/Perl5ish subroutine declarations/You can declare a sub without
 # parameter list>
@@ -41,4 +41,15 @@ is both( 'x', :delta<echo>, 'foxtrot' ), 'foxtrotecho',
 is both(), '',
    'sub using both named and position params works with no params';
 
+# RT 71112
+{
+    sub rt71112 { @_[0] = 'bug' }
+    my $tender = 'sanity';
+    #?rakudo todo 'RT 71112: Cannot assign to readonly variable.'
+    dies_ok { rt71112( $tender ) }, 'Sub that tries to modify @_ dies';
+    is $tender, 'sanity', 'The variable passed is unchanged.';
+}
+
+done_testing;
+
 # vim: ft=perl6

The RT System itself - Status changed from 'new' to 'open'

From @bbkr

NOM​: still broken (allows to modify @​_)

bbkr​:nom bbkr$ ./perl6 -e 'sub foo { @​_[0] = "bah" }; my $a = "!";
foo($a); say $a'
!

From @jnthn

On Mon Dec 07 21​:36​:46 2009, masak wrote​:

<masak> rakudo​: sub foo { @​_[0] = "bah" }; my $a = "!"; foo($a); say $a
<p6eval> rakudo 7ef386​: !␤
<TimToady> i'd've expected it to blow up
* masak submits rakudobug
<masak> TimToady​: so in the default signature, @​_ isn't writeable?
<TimToady> no, it's different from P5 in that
<masak> ok.
<TimToady> but most P5 programs aren't modifying their parameters
<TimToady> and we didn't want to pay the might-be-lvalue penalty anymore
<TimToady> that was decided in about 2003 or so
<masak> nod.

The ruling is that, since *@​_ would have "is copy" semantics, so the default @​_ should. Reference​:

http://irclog.perlgeek.de/perl6/2015-11-12#i_11527317

Codified the ruling in the spectests in S06-signature/unspecified.t.

/jnthn

@jnthn - Status changed from 'open' to 'resolved'