Prepared test (attached).
Thanks,
Dmytro
> Subject: Re: 6853676: OperatingSystemMXBean.TotalPhysicalMemorySize has
> incorrect value
> From: [email protected]
> Date: Thu, 30 Aug 2012 12:06:09 +0200
> CC: [email protected]; [email protected];
> [email protected]
> To: [email protected]
>
>
> On 30 aug 2012, at 11:31, David Holmes <[email protected]> wrote:
>
> > On 30/08/2012 5:50 PM, Staffan Larsen wrote:
> >> The patch looks good to me and I can sponsor the push to JDK8-TL.
> >> However, you need a review from an official Reviewer as well.
> >
> > Reviewed.
> >
> > Any way we can get a regression test too?
>
> That would be great - I wonder what the test would check, though? That the
> value returned is not -1? And hope that sometimes the test will be run on a
> machine with > 4GB ram.
>
> /Staffan
>
> >
> > David
> > -----
> >
> >> BTW, I suggest we remove the confusing comment in
> >> Java_com_sun_management_OperatingSystem_getTotalPhysicalMemorySize()
> >> while we are at it.
> >>
> >> /Staffan
> >>
> >> On 30 aug 2012, at 09:18, Dmytro Sheyko <[email protected]
> >> <mailto:[email protected]>> wrote:
> >>
> >>> Hi,
> >>>
> >>> Could you please review the patch and apply it if it's correct?
> >>> https://bugs.openjdk.java.net/show_bug.cgi?id=100077
> >>> http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6853676
> >>>
> >>> Thanks,
> >>> Dmytro
> >>>
> >>> > Date: Wed, 2 Feb 2011 20:09:59 +1000
> >>> > From:[email protected] <mailto:[email protected]>
> >>> > To:[email protected] <mailto:[email protected]>
> >>> > CC:[email protected] <mailto:[email protected]>;
> >>> [email protected]
> >>> <mailto:[email protected]>;[email protected]
> >>> <mailto:[email protected]>;[email protected]
> >>> <mailto:[email protected]>;[email protected]
> >>> <mailto:[email protected]>
> >>> > Subject: Re: 6853676: OperatingSystemMXBean.TotalPhysicalMemorySize
> >>> has incorrect value
> >>> >
> >>> > Alan Bateman said the following on 02/02/11 20:05:
> >>> > > David Holmes wrote:
> >>> > >> It looks like this was actually fixed under 6840305 back in July
> >>> 2009:
> >>> > >>
> >>> > >> http://hg.openjdk.java.net/jdk7/hotspot-rt/hotspot/rev/8c79517a9300
> >>> > >>
> >>> > >> This CR was not updated however.
> >>> > >>
> >>> > >> Does the problem still exist?
> >>> > >>
> >>> > >> David Holmes
> >>> > > I think this is separate and 6853676 is about
> >>> > > com.sun.management.OperatingSystemMXBean. The code for that is in jdk
> >>> > > repo in src/windows/native/com/sun/management. It should be using
> >>> > > GlobalMemoryStatusEx rather than GlobalMemoryStatus.
> >>> >
> >>> > Thanks Alan, the comments in 6853676 led me astray.
> >>> >
> >>> > As a P4 it looks like this has just slipped through the cracks.
> >>> >
> >>> > David
> >>> >
> >>> >
> >>> > > Dmytro - to your question, serviceability-dev is the right place to
> >>> > > bring it.
> >>> > >
> >>> > > -Alan
> >>
>
import java.lang.reflect.Method;
import java.util.Arrays;
import java.util.ArrayList;
import java.util.List;
import java.lang.management.ManagementFactory;
import com.sun.management.OperatingSystemMXBean;
/*
* @bug 6853676
* @summary On computers with more than 4 GB of memory, the GlobalMemoryStatus
function can return incorrect information, reporting a value of 1 to indicate
an overflow.
*/
public class MemoryStatusOverflow {
static final long MEMORYSTATUS_OVERFLOW = (1L << 32) - 1; // (DWORD) -1
public static void main(String... args) throws Exception {
OperatingSystemMXBean bean = (OperatingSystemMXBean)
ManagementFactory.getOperatingSystemMXBean();
List<String> failed = new ArrayList<String>();
for (String prop : Arrays.asList("getTotalSwapSpaceSize",
"getFreeSwapSpaceSize", "getTotalPhysicalMemorySize",
"getFreePhysicalMemorySize")) {
Method getter =
OperatingSystemMXBean.class.getMethod(prop);
long value = (Long) getter.invoke(bean);
if (value == MEMORYSTATUS_OVERFLOW) {
failed.add(prop);
}
}
if (!failed.isEmpty()) {
throw new AssertionError(failed);
}
System.out.println("Test passed.");
}
}
