[RFC]: backport to fix PR655

[Xerxes Rånby]

On 2011-03-01 15:25, Denis Lila wrote:
> Hi.
>
>> I can reproduce this bug using the testcase at the end of:
>> http://bugs.sun.com/view_bug.do?bug_id=6934977
>>
>> But hmm.. the test still fail for me with this backport applied..
>> although in a different way. Do this backport need something that can
>> handle the thrown InternalError Exception to work?
> Thanks for testing this.
>
> The InternalError you're seeing is exactly what is supposed to happen.
> If you run the jtreg tests included in the patch (Basic.java and
> Truncate.java) you'll see that without the patch, Basic will fail
> with an IOException, and Truncate will fail with a SIGBUS. With the
> backport applied, neither of them will fail (because Truncate.java has
> some logic to ignore the Error).

Thank you for clarifying.
The included test-cases Basic and Truncate do pass on my setup.

> ----- Original Message -----
>> Hi!
>> On 2011-03-01 00:07, Denis Lila wrote:
>>> Hi.
>>>
>>> The attached patch would backport this changeset:
>>> http://hg.openjdk.java.net/jdk7/jdk7/jdk/rev/24741c4bf300
>>> which fixes S6934977. PR655 seems to be a duplicate of that
>>> (but I don't have a reproducer for PR655 so I can't say for
>>> certain; however, looking at the 2 bug descriptions, there
>>> is little doubt).
>>> Ok to push?
>>
Ok to push to HEAD and 1.10 branch.


>>> Thank you,
>>> Denis.
>>
Cheers
Xerxes