RFR: JDK-8171084: heapdump/JMapHeapCore fails with java.lang.RuntimeException: Heap segment size overflow

[David Holmes]

Hi Jini,

Seems reasonable. I don't fully understand all the details here but have 
one non trivial comment ...

On 20/01/2017 2:40 AM, Jini George wrote:
> Hi, Requesting for reviews for the following SA issue:
>
> https://bugs.openjdk.java.net/browse/JDK-8171084
> Webrev:<http://cr.openjdk.java.net/%7Ejgeorge/8171084/webrev.00/>http://cr.openjdk.java.net/~jgeorge/8171084/webrev.00/

src/jdk.hotspot.agent/share/classes/sun/jvm/hotspot/utilities/HeapHprofBinWriter.java

This change:

  870     protected void writeInstance(Instance instance, int length) 
throws IOException {

seems incorrect. It adds a length parameter that is unused and also 
creates an overload, rather than override of the inherited version. As a 
result this code:

612             writeInstance(instance);

will call the super version in AbstractHeapGraphWriter.


Minor nit:

379     private static final long MAX_U4_VALUE = 4294967295L;

would be clearer as:

379     private static final long MAX_U4_VALUE = 0xFFFFFFFFL;

---

test/serviceability/sa/LingeredAppWithLargeArray.java

Style nit:

  27     public int hugeArray[];

should be

  27     public int[] hugeArray;

but why public ??

I don't know how LingeredApp works, but this looks odd:

32     public static void main(String args[]) {
33         LingeredAppWithLargeArray appObject = new 
LingeredAppWithLargeArray();
34         LingeredApp.main(args);
35     }

as the appObject is never used. ??

---

test/serviceability/sa/TestHeapDumpForLargeArray.java

Why is the test excluded on Mac?

Thanks,
David
-----

> Thanks,
> Jini.