RFR (XXS): 8029366: ShouldNotReachHere error when creating an array with component type of void

[Vladimir Kozlov]

On 12/6/13 11:38 AM, Christian Thalinger wrote:
> For small ones like this I don’t bother creating a webrev.  The patch is archived in the mailing list archive anyway.

Please, don't invent a new way of review. I would prefer to look on past 
webrevs only in one place.

Vladimir

>
> Thanks for the review.
>
> On Dec 6, 2013, at 10:41 AM, Vladimir Kozlov <vladimir.kozlov at oracle.com> wrote:
>
>> What happened with webrevs in cr.openjdk?
>>
>> Looks fine to me.
>>
>> Thanks,
>> Vladimir
>>
>> On 12/6/13 9:05 AM, Christian Thalinger wrote:
>>> https://bugs.openjdk.java.net/browse/JDK-8029366
>>>
>>> 8029366: ShouldNotReachHere error when creating an array with component type of void
>>>
>>> The code in LoadNode::klass_value_common tries to create a ciArrayKlass of type T_VOID which fails since there is no
>>> such thing as a void array.
>>>
>>> Primitive classes have null for a klass field. For the non-array case we return null but for the array case we don't and
>>> so we hit an assert in ciArrayKlass::make.
>>>
>>> The fix is to check for void types in the array case and return TypePtr::NULL_PTR.
>>>
>>> diff -r 55dd6e77b399 src/share/vm/opto/memnode.cpp
>>> --- a/src/share/vm/opto/memnode.cppFri Nov 22 15:26:37 2013 +0100
>>> +++ b/src/share/vm/opto/memnode.cppFri Dec 06 09:04:15 2013 -0800
>>> @@ -2071,6 +2071,11 @@ const Type *LoadNode::klass_value_common
>>>         if (t != NULL) {
>>>           // constant oop => constant klass
>>>           if (offset == java_lang_Class::array_klass_offset_in_bytes()) {
>>> +          if (t->is_void()) {
>>> +            // We cannot create a void array.  Since void is a primitive type return null
>>> +            // klass.  Users of this result need to do a null check on the returned klass.
>>> +            return TypePtr::NULL_PTR;
>>> +          }
>>>             return TypeKlassPtr::make(ciArrayKlass::make(t));
>>>           }
>>>           if (!t->is_klass()) {
>>> diff -r 55dd6e77b399 test/compiler/reflection/ArrayNewInstanceOfVoid.java
>>> --- /dev/nullThu Jan 01 00:00:00 1970 +0000
>>> +++ b/test/compiler/reflection/ArrayNewInstanceOfVoid.javaFri Dec 06 09:04:15 2013 -0800
>>> @@ -0,0 +1,44 @@
>>> +/*
>>> + * Copyright (c) 2013, Oracle and/or its affiliates. All rights reserved.
>>> + * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
>>> + *
>>> + * This code is free software; you can redistribute it and/or modify it
>>> + * under the terms of the GNU General Public License version 2 only, as
>>> + * published by the Free Software Foundation.
>>> + *
>>> + * This code is distributed in the hope that it will be useful, but WITHOUT
>>> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
>>> + * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
>>> + * version 2 for more details (a copy is included in the LICENSE file that
>>> + * accompanied this code).
>>> + *
>>> + * You should have received a copy of the GNU General Public License version
>>> + * 2 along with this work; if not, write to the Free Software Foundation,
>>> + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
>>> + *
>>> + * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
>>> + * or visit www.oracle.com <http://www.oracle.com> if you need additional information or have any
>>> + * questions.
>>> + */
>>> +
>>> +/*
>>> + * @test
>>> + * @bug 8029366
>>> + * @summary ShouldNotReachHere error when creating an array with component type of void
>>> + */
>>> +
>>> +public class ArrayNewInstanceOfVoid {
>>> +    public static void main(String[] args) {
>>> +        for (int i = 0; i < 100_000; i++) {
>>> +            test();
>>> +        }
>>> +    }
>>> +
>>> +    private static void test() {
>>> +        try {
>>> +            java.lang.reflect.Array.newInstance(void.class, 2);
>>> +        } catch (IllegalArgumentException e) {
>>> +            // expected
>>> +        }
>>> +    }
>>> +}
>