RFR(S): 8058827: XCode 6.0 (Clang) warning "operator new' should not return a null pointer unless..."

Vladimir Kozlov vladimir.kozlov at oracle.com
Sat Sep 20 18:00:37 UTC 2014


Changes are good, David.
We were only concern that we have failures in general. Yes, they are not related to your changes.
We need to investigate these 2 timeouts - I also observed them. But it is different issue.

Thanks,
Vladimir

On 9/20/14 9:57 AM, David Chase wrote:
> A couple of timeouts, otherwise ignored tests, with the new jtreg.
>
> TEST RESULT: Error. Program `/Users/dr2chase/work/jdk9/build/macosx-x86_64-normal-server-fastdebug/images/j2sdk-image/bin/java' interrupted! (timed out?)
> TEST RESULT: Error. Test ignored: 7154567
> TEST RESULT: Error. Test ignored: 8031978
> TEST RESULT: Error. Program `sh' interrupted! (timed out?)
> TEST RESULT: Error. Test ignored: 8046268
> TEST RESULT: Error. Test ignored: 8046268
> TEST RESULT: Error. Test ignored: 8046268
> TEST RESULT: Error. Test ignored: 8046268
> TEST RESULT: Error. Test ignored: 8046268
> TEST RESULT: Error. Test ignored: 8046268
> TEST RESULT: Error. Test ignored: 8046268
> TEST RESULT: Error. Test ignored: 8046268
>
> But my main goal (and the goal of the bug) is the compilation problem, not random test foolishness
> that is invariant across my changes.
>
> David
>
> On 2014-09-19, at 8:26 PM, David Chase <david.r.chase at oracle.com> wrote:
>
>> It’s errors from ignored tests, and turned out I was running a down-rev jtreg.
>> I can re-run with the proper jtreg, if that helps.
>>
>> David
>>
>> On 2014-09-19, at 6:03 PM, Christian Thalinger <christian.thalinger at oracle.com> wrote:
>>
>>>
>>> On Sep 19, 2014, at 12:30 PM, David Chase <david.r.chase at oracle.com> wrote:
>>>
>>>> bug: https://bugs.openjdk.java.net/browse/JDK-8058827
>>>> webrev: http://cr.openjdk.java.net/~drchase/8058827/webrev.00/
>>>>
>>>> Problem: I decided I would find trouble before it found me, and installed XCode 6 and tried a fastdebug build.
>>>> There appears to be one stopper, which is this warning compiling hotspot:
>>>>
>>>> /Users/dr2chase/work/jdk9/hotspot/src/share/vm/memory/allocation.cpp:698:3: error: 'operator new' should not return a null pointer unless it is declared 'throw()' [-Werror,-Wnew-returns-null]
>>>> return 0;
>>>> ^
>>>>
>>>> The file in question has a comment warning against careless meddling, so I did not touch the signature,
>>>> and instead rewrote the zero as a static.
>>>>
>>>> Testing:
>>>> compilation on Mavericks + XCode 6.0.1, compilation on Ubuntu 13.10 + Gcc 4.8, both fastdebug.
>>>>
>>>> I ran jtreg on hotspot/test/compiler and it looked basically functional (i.e., about as many failures as usual, about a dozen).
>>>
>>> This should not be the case.  Which are failing?
>>>
>>>>
>>>> David
>>
>


More information about the hotspot-runtime-dev mailing list