Re: [rvm-research] Fixing size of metadata space?

Hi Nathan,

On 04/03/2015 04:55 PM, Nathan Ricci wrote:
>   Ah, I see, I didn't realized that they had to be chunk aligned. That 
> makes sense.
>
>   However, after fixing that, the metadata space doesn't actually 
> appear to be any larger, as reported by Plan.printPostStats()  (the 
> thing that prints in verbose mode) the metadata space is mostly empty. 
> Or is that reporting only the amount of stuff actually allocated, not 
> the size of the space?

I think it's the latter: it's printing the amount of stuff allocated, 
not the virtual memory bounds of the space.

I think the virtual memory bounds of the space get printed when Jikes 
RVM boots if you set GC verbosity high enough (>=3?). Note that in 
unmodified Jikes RVM, the metadata space is a discontiguous space, so it 
won't have dedicated virtual memory bounds.

Cheers,
Mike

>
>            --Nathan
>
> On Fri, Apr 3, 2015 at 12:05 PM, Michael Bond 
> <[email protected] <mailto:[email protected]>> wrote:
>
>     Hi Nathan,
>
>     It looks like the first problem is an assertion failure in
>     Space.<init>() at line 152. My line numbers don't match exactly,
>     but it seems like the error is probably occurring because of a
>     non-chunk-aligned size (10 MB):
>
>         if (extent.NE(chunkAlign(extent, false))) {
>           VM.assertions.fail(name + " requested non-aligned extent: "
>     + extent.toLong() + " bytes");
>         }
>
>     Cheers,
>     Mike
>
>
>     On 04/03/2015 11:55 AM, Nathan Ricci wrote:
>>     Hello,
>>
>>              I'm trying to compare two different collectors that use
>>     a different amount of meta-data, and as part of that I wanted to
>>     fix the size of the metadata space. It tried to do this by
>>     chaning the call where the metadata space is created in Plan, from:
>>
>>       metaDataSpace = new RawPageSpace("meta",
>>           VMRequest.create());
>>
>>     To:
>>        metaDataSpace = new RawPageSpace("meta",
>>           VMRequest.create(10));
>>
>>         In order to create a VMRequest of 10 MB, and give the
>>     metadata space an initial size of 10 MB. However, this causes a
>>     null pointer exeception during the build process (stack trace
>>     below).   Am I going about this the wrong way?
>>
>>              Thanks,
>>                   Nathan Ricci, Tufts University
>>
>>
>>           java.lang.ExceptionInInitializerError
>>             at
>>     org.jikesrvm.mm.mminterface.MemoryManager.getAllocationSite(MemoryManager.java:335)
>>             at
>>     org.jikesrvm.compilers.baseline.ia32.BaselineCompilerImpl.emit_resolved_new(BaselineCompilerImpl.java:3373)
>>             at
>>     org.jikesrvm.compilers.baseline.TemplateCompilerFramework.genCode(TemplateCompilerFramework.java:1593)
>>             at
>>     org.jikesrvm.compilers.baseline.BaselineCompiler.compile(BaselineCompiler.java:250)
>>             at
>>     org.jikesrvm.compilers.baseline.BaselineCompiledMethod.compile(BaselineCompiledMethod.java:165)
>>             at
>>     org.jikesrvm.compilers.baseline.BaselineCompiler.compile(BaselineCompiler.java:173)
>>             at
>>     org.jikesrvm.compilers.baseline.BaselineBootImageCompiler.compileMethod(BaselineBootImageCompiler.java:52)
>>             at
>>     org.jikesrvm.compilers.common.BootImageCompiler.compile(BootImageCompiler.java:75)
>>             at
>>     org.jikesrvm.compilers.common.BootImageCompiler.compile(BootImageCompiler.java:80)
>>             at
>>     org.jikesrvm.classloader.NormalMethod.genCode(NormalMethod.java:180)
>>             at
>>     org.jikesrvm.classloader.RVMMethod.compile(RVMMethod.java:729)
>>             at
>>     org.jikesrvm.classloader.RVMMethod.getCurrentEntryCodeArray(RVMMethod.java:713)
>>             at
>>     org.jikesrvm.classloader.RVMClass.instantiate(RVMClass.java:1436)
>>             at
>>     org.jikesrvm.classloader.RVMClass.instantiate(RVMClass.java:1413)
>>             at
>>     org.jikesrvm.tools.bootImageWriter.BootImageWorker.run(BootImageWorker.java:44)
>>             at
>>     java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1146)
>>             at
>>     java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
>>             at java.lang.Thread.run(Thread.java:701)
>>           Caused by: java.lang.NullPointerException
>>             at org.mmtk.policy.Space.reservedPages(Space.java:226)
>>             at org.mmtk.policy.Space.getPagesReserved(Space.java:501)
>>             at org.mmtk.policy.Space.printUsage(Space.java:640)
>>             at org.mmtk.policy.Space.printUsagePages(Space.java:519)
>>             at org.jikesrvm.mm.mmtk.Assert.fail(Assert.java:45)
>>             at org.mmtk.policy.Space.<init>(Space.java:152)
>>             at org.mmtk.policy.RawPageSpace.<init>(RawPageSpace.java:43)
>>             at org.mmtk.plan.Plan.<clinit>(Plan.java:162)
>>
>>
>>
>>
>>
>>     ------------------------------------------------------------------------------
>>     Dive into the World of Parallel Programming The Go Parallel Website, sponsored
>>     by Intel and developed in partnership with Slashdot Media, is your hub for all
>>     things parallel software development, from weekly thought leadership blogs to
>>     news, videos, case studies, tutorials and more. Take a look and join the
>>     conversation now.http://goparallel.sourceforge.net/
>>
>>
>>     _______________________________________________
>>     Jikesrvm-researchers mailing list
>>     [email protected]  <mailto:[email protected]>
>>     https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers
>
>
>     ------------------------------------------------------------------------------
>     Dive into the World of Parallel Programming The Go Parallel
>     Website, sponsored
>     by Intel and developed in partnership with Slashdot Media, is your
>     hub for all
>     things parallel software development, from weekly thought
>     leadership blogs to
>     news, videos, case studies, tutorials and more. Take a look and
>     join the
>     conversation now. http://goparallel.sourceforge.net/
>     _______________________________________________
>     Jikesrvm-researchers mailing list
>     [email protected]
>     <mailto:[email protected]>
>     https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers
>
>
>
>
> ------------------------------------------------------------------------------
> Dive into the World of Parallel Programming The Go Parallel Website, sponsored
> by Intel and developed in partnership with Slashdot Media, is your hub for all
> things parallel software development, from weekly thought leadership blogs to
> news, videos, case studies, tutorials and more. Take a look and join the
> conversation now. http://goparallel.sourceforge.net/
>
>
> _______________________________________________
> Jikesrvm-researchers mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers

------------------------------------------------------------------------------
BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
Develop your own process in accordance with the BPMN 2 standard
Learn Process modeling best practices with Bonita BPM through live exercises
http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_
source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF

_______________________________________________
Jikesrvm-researchers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jikesrvm-researchers