Btw. I'm actually not sure that the problem really is an out of memory
exception. Yes, the log says so, but I think it's caused by an other
unhandled exception, and then something is catching that exception and
throws an out of memory exception. Don't know why, but that is a hunch
that I get.



On 5 Aug, 09:09, Bob Kerns <r...@acm.org> wrote:
> OK. I'm not 100% certain that the command line will take precedence
> over the eclipse.ini. You might want to edit the -Xmx512m there.
>
> Except that with a drive letter like L:, I'm surmising that you're
> running it from a network share? This may cause some significant
> performance problems, and is not something I would recommend. That
> wouldn't cause you to run out of memory, though. But you might want to
> experiment with a local copy.
>
> Private working set is not relevant to running out of memory. "working
> set" merely indicates, out of all the memory allocated in this
> process, how much is actually in RAM? It is relevant to competing for
> RAM with other processes, and thus paging slowdowns.
>
> By "Commit size is now 381.500 kB", you mean 381.5 MB? For this one
> process? (Just checking). That is perhaps about right for a setting of
> -Xmx512m, with the GC factored in, etc. It is not consistent with
> running out of a > 1GB allocation. (And I'm not sure the 32-bit
> version of Java can actually handle a heap of that size; I don't
> recall the limit, but I think you're not that far from it at best. I
> do recall it changing from release to release and finding a server
> stopped working as a result...) So one possibility is that the Eclipse
> launcher may be dropping that setting without telling you, just
> because it's too big.
>
> One thing you can do, if you haven't already: In Eclipse, go to
> Windows / Preferences... / General and check Show Heap Status. This
> puts a field in the status bar that shows Java's idea of how much Java
> heap it has allocated, and the current limit (i.e. the upper number
> will grow from -XmsXXXm to -XmxYYYm). This is an easy way to see
> Eclipse's GC status.
>
> On Aug 4, 10:37 pm, Kaj Bjurman <kaj.bjur...@gmail.com> wrote:
>
>
>
> > Eclipse is started using a shortcut:
>
> > Galileo
> > L:\eclipse-jee-galileo-win32\eclipse\eclipse.exe -vmargs -Xmx1324m
>
> > Ganymede
> > L:\eclipse\eclipse.exe -vmargs -Xmx1324m
>
> > (I'm having the same problem in both)
>
> > eclipse.ini says:
>
> > --launcher.library
> > plugins/org.eclipse.equinox.launcher.win32.win32.x86_1.0.200.v20090519
> > -startup
> > plugins/org.eclipse.equinox.launcher_1.0.201.R35x_v20090715.jar
> > -product
> > org.eclipse.epp.package.jee.product
> > --launcher.XXMaxPermSize
> > 256M
> > -showsplash
> > org.eclipse.platform
> > --launcher.XXMaxPermSize
> > 256m
> > -vmargs
> > -Dosgi.requiredJavaVersion=1.5
> > -Xms40m
> > -Xmx512m
>
> > The java version that is in path:
>
> > java version "1.6.0_13"
> > Java(TM) SE Runtime Environment (build 1.6.0_13-b03)
> > Java HotSpot(TM) Client VM (build 11.3-b02, mixed mode, sharing)
>
> > I only checked the process in task manager, but I also made the
> > columns for handles, threads and user objects visible. The number of
> > threads was only about 40 which surprised me. (An fresh started
> > Eclipse on my system has about 28 threads)
>
> > Memory (Private working set) was about 400.000 kB. I did not check
> > commit size, I tonigh when I got time to provoke Eclipse again. The
> > paging size looked normal.
>
> > I now saw that that Eclipse only used revision 5 of the Android SDK
> > Tools. I started a fresh Eclipse, opened up Android SDK and AVD
> > Manager, selected to install all available packages, and then bumbed
> > into the same error again (so it took less than 2 minutes)
>
> > Commit size is now 381.500 kB, Private working set 327.048 kB, Threads
> > 26, Handles 506
>
> > On 5 Aug, 03:44, Bob Kerns <r...@acm.org> wrote:
>
> > > OK, this is good information. You are, in fact, running out of memory
> > > -- somehow.
>
> > > I've see this, but seldom as performance starts to suck for me long
> > > before.
>
> > > How exactly are you allocating memory for Java? How are you starting
> > > Eclipse?
>
> > > And when you were checking how much memory Eclipse was using, how did
> > > you check that, exactly? It sounds like you used the task manager --
> > > what is the exact title of the column you used? Task manager is not
> > > ideal for looking at this, but the only column that is actually
> > > relevant out of the 7 memory columns available is "Memory - Commit
> > > size".  But the relationship between that and Java's GC'd heap memory
> > > allocation is tenuous. About the only thing you can say is that Memory
> > > - Commit size" will be greater than the amount of memory Java is
> > > actively using in its heap.
>
> > > Did you check your paging file space? That's another way you can run
> > > out of memory.
>
> > > And finally, it may report this error when you've run out of native
> > > threads.  That is, in fact, WHERE you are running out of memory,
> > > though it may not be the specific cause. Check the Threads column.
>
> > > With Galileo, and Rev 6 of the Android SDK tools, I don't see a thread
> > > leak. In fact, my thread count was initially 47, went up to 52 briefly
> > > while auto-building my project, and then as I started looking at
> > > layouts, it has gone down to 37, and I don't seem to find anything
> > > with layouts that bumps it up at all.
>
> > > If you see a different behavior, then we have a clue. (Otherwise, it's
> > > still a clue, but a weak one).
>
> > > On Aug 4, 2:00 pm, Kaj Bjurman <kaj.bjur...@gmail.com> wrote:
>
> > > > I just provoked my Eclipse again, just to see what the error message
> > > > is. I remembered it as out of handles, but that was a bit wrong. I
> > > > checked the process in the task manager when it was in the error
> > > > state. It isn't out of memory, and it's probably not out of handles.
> > > > The CPUs are also idle, so nothing is spinning.
>
> > > > The error from Eclipse is a dialog that says:
>
> > > > Multiple problems have occured (as title)
>
> > > > The error message is then, "unable to create new native thread".
>
> > > > Followed by a textfield that is displaying:
>
> > > > Collecting garbage
> > > > Collecting garbage
> > > > Unhandled event loop exception
>
> > > > unable to create new native thread
>
> > > > The eclipse error log can be seen below.
>
> > > > Eclipse was at that time taking about 400 MB of memory (of the 1300M
> > > > that I had assigned to the VM)
>
> > > > It took me about 10-15 minutes. All I did was editing one layout file,
> > > > switched between layout view, then edited the xml directly, changed
> > > > some properties through the property view, saved quite frequently,
> > > > even with compilation errors, and also tried keep editing while it was
> > > > saving.
>
> > > > This never happens when I write code in Eclipse.
>
> > > > !ENTRY org.eclipse.core.resources 8 2 2010-08-04 22:40:49.414
> > > > !MESSAGE Problems occurred when invoking code from plug-in:
> > > > "org.eclipse.core.resources".
> > > > !STACK 1
> > > > org.eclipse.core.runtime.CoreException: 'aapt' error. Pre Compiler
> > > > Build aborted.
> > > >         at
> > > > com.android.ide.eclipse.adt.internal.build.BaseBuilder.stopBuild(Unknown
> > > > Source)
> > > >         at
> > > > com.android.ide.eclipse.adt.internal.build.PreCompilerBuilder.build(Unknown
> > > > Source)
> > > >         at org.eclipse.core.internal.events.BuildManager
> > > > $2.run(BuildManager.java:627)
> > > >         at org.eclipse.core.runtime.SafeRunner.run(SafeRunner.java:42)
> > > >         at
> > > > org.eclipse.core.internal.events.BuildManager.basicBuild(BuildManager.java:
> > > > 170)
> > > >         at
> > > > org.eclipse.core.internal.events.BuildManager.basicBuild(BuildManager.java:
> > > > 201)
> > > >         at org.eclipse.core.internal.events.BuildManager
> > > > $1.run(BuildManager.java:253)
> > > >         at org.eclipse.core.runtime.SafeRunner.run(SafeRunner.java:42)
> > > >         at
> > > > org.eclipse.core.internal.events.BuildManager.basicBuild(BuildManager.java:
> > > > 256)
> > > >         at
> > > > org.eclipse.core.internal.events.BuildManager.basicBuildLoop(BuildManager.j
> > > >  ava:
> > > > 309)
> > > >         at
> > > > org.eclipse.core.internal.events.BuildManager.build(BuildManager.java:
> > > > 341)
> > > >         at
> > > > org.eclipse.core.internal.events.AutoBuildJob.doBuild(AutoBuildJob.java:
> > > > 140)
> > > >         at
> > > > org.eclipse.core.internal.events.AutoBuildJob.run(AutoBuildJob.java:
> > > > 238)
> > > >         at org.eclipse.core.internal.jobs.Worker.run(Worker.java:55)
> > > > !SUBENTRY 1 com.android.ide.eclipse.adt 8 0 2010-08-04 22:40:49.414
> > > > !MESSAGE 'aapt' error. Pre Compiler Build aborted.
> > > > !SUBENTRY 1 com.android.ide.eclipse.adt 8 0 2010-08-04 22:40:49.414
> > > > !MESSAGE 'aapt' error. Pre Compiler Build aborted.
>
> > > > !ENTRY org.eclipse.ui 4 0 2010-08-04 22:45:29.067
> > > > !MESSAGE Unhandled event loop exception
> > > > !STACK 0
> > > > java.lang.OutOfMemoryError: unable to create new native thread
> > > >         at java.lang.Thread.start0(Native Method)
> > > >         at java.lang.Thread.start(Unknown Source)
> > > >         at org.eclipse.jface.text.contentassist.AdditionalInfoController
> > > > $Timer.<init>(AdditionalInfoController.java:244)
> > > >         at org.eclipse.jface.text.contentassist.AdditionalInfoController
> > > > $10.<init>(AdditionalInfoController.java:457)
> > > >         at
> > > > org.eclipse.jface.text.contentassist.AdditionalInfoController.install(Addit
> > > >  ionalInfoController.java:
> > > > 457)
> > > >         at
> > > > org.eclipse.jface.text.contentassist.CompletionProposalPopup.displayProposa
> > > >  ls(CompletionProposalPopup.java:
> > > > 1198)
> > > >         at 
> > > > org.eclipse.jface.text.contentassist.CompletionProposalPopup.access
> > > > $23(CompletionProposalPopup.java:1157)
> > > >         at org.eclipse.jface.text.contentassist.CompletionProposalPopup
> > > > $2.run(CompletionProposalPopup.java:500)
> > > >         at 
> > > > org.eclipse.swt.custom.BusyIndicator.showWhile(BusyIndicator.java:
> > > > 70)
> > > >         at
> > > > org.eclipse.jface.text.contentassist.CompletionProposalPopup.showProposals(
> > > >  CompletionProposalPopup.java:
> > > > 482)
> > > >         at
>
> ...
>
> läs mer »

-- 
You received this message because you are subscribed to the Google
Groups "Android Developers" group.
To post to this group, send email to android-developers@googlegroups.com
To unsubscribe from this group, send email to
android-developers+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/android-developers?hl=en

Reply via email to