Re: RFR (S) 8212931 HeapMonitorStatIntervalTest.java fails due average calculation

Sat, 10 Nov 2018 15:26:36 -0800

Hi Jc,

Looks good.

One minor comment.

http://cr.openjdk.java.net/%7Ejcbeyler/8212931/webrev.00/test/hotspot/jtreg/serviceability/jvmti/HeapMonitor/MyPackage/HeapMonitor.java.frames.html

 The "average" in comments is not needed anymore:
 112   // Calculate the size of a 1-element array in order to assess average sampling interval
 113   // via the HeapMonitorStatIntervalTest.
 ...

 125     // Get the actual average size.
 126     _oneElementSize_ = getSize(frames);
 127     System.out.println("Element size is: " + oneElementSize);

Thanks,
Serguei


On 11/9/18 19:42, JC Beyler wrote:
Hi all,

Any chance I can get a review on this (technically two reviews :-))?

Thanks,
Jc

On Fri, Nov 2, 2018 at 3:55 PM JC Beyler <[email protected]> wrote:
Hi all,

Could I get a review for a bug in the calculation of the average size of an allocation? The solution is actually not to do the calculation at all. Instead, I just go in the cache and find an object with the right stacktrace and get its size (since all sizes should be equal because they are allocating every time 1-element arrays).

This removes the risk of a problem and simplifies the test.

Webrev: http://cr.openjdk.java.net/~jcbeyler/8212931/webrev.00/


--

Thanks,
Jc

Reply via email to