[ImageJ-devel] CPU usage

[Curtis Rueden]

Hi all,

Barry is out today but I will investigate this afternoon and hopefully have
> a fix soon.
>

I believe this issue is fixed in r3939. Please let me know if not.

-Curtis


On Wed, Sep 21, 2011 at 12:27 PM, Curtis Rueden <ctrueden at wisc.edu> wrote:

> Hi Grant,
>
>
> Anyone else seeing the CPU usage stuck at +12% ofter opening an image in
>> IJ2?  Or is it just me?
>>
> (Related: has anyone tried profiling the app?  I'm having trouble doing so
>> in NetBeans... going to try VirtualVM next.)
>>
>
> If you run ImageJ from the command line (see
> http://imagejdev.org/command-line) then press Ctrl+Pause (Ctrl+\ on
> non-Windows platforms) from the command prompt, you will get a full stack
> trace for all threads, which will be enlightening as to the cause.
>
> I just tested on my machine here and am seeing 100% CPU usage (of one CPU)
> on my system, after opening Blobs.
>
> The stack trace reveals that there is a recursive loop going on with the
> options synchronization. Relevant portion is:
>
>     at imagej.event.EventService.publish(EventService.java:72)
>     at imagej.event.Events.publish(Events.java:61)
>     at imagej.ext.options.OptionsPlugin.run(OptionsPlugin.java:116)
>     at imagej.ext.plugin.PluginModule.run(PluginModule.java:163)
>     at imagej.ext.options.OptionsService.setOption(OptionsService.java:178)
>     at imagej.ext.options.OptionsService.setOption(OptionsService.java:153)
>     at imagej.legacy.patches.PrefsMethods.set(PrefsMethods.java:90)
>     at ij.Prefs.set(Prefs.java:455)
>     at ij.Prefs.set(Prefs.java:461)
>     at
> imagej.legacy.OptionsSynchronizer.arrowOptions(OptionsSynchronizer.java:171)
>     at
> imagej.legacy.OptionsSynchronizer.updateIJ1SettingsFromIJ2(OptionsSynchronizer.java:82)
>     at
> imagej.legacy.LegacyService.updateIJ1Settings(LegacyService.java:143)
>     at imagej.legacy.LegacyService$2.onEvent(LegacyService.java:194)
>     at imagej.legacy.LegacyService$2.onEvent(LegacyService.java:190)
>     at
> org.bushe.swing.event.ThreadSafeEventService.publish(ThreadSafeEventService.java:971)
>     at
> org.bushe.swing.event.ThreadSafeEventService.publish(ThreadSafeEventService.java:904)
>     at imagej.event.EventService.publish(EventService.java:72)
>     at imagej.event.Events.publish(Events.java:61)
>     at imagej.ext.options.OptionsPlugin.run(OptionsPlugin.java:116)
>     at imagej.ext.plugin.PluginModule.run(PluginModule.java:163)
>     at imagej.ext.options.OptionsService.setOption(OptionsService.java:178)
>     at imagej.ext.options.OptionsService.setOption(OptionsService.java:153)
>     at imagej.legacy.patches.PrefsMethods.set(PrefsMethods.java:90)
>     at ij.Prefs.set(Prefs.java:455)
>     at ij.Prefs.set(Prefs.java:461)
>     at
> imagej.legacy.OptionsSynchronizer.arrowOptions(OptionsSynchronizer.java:171)
>     at
> imagej.legacy.OptionsSynchronizer.updateIJ1SettingsFromIJ2(OptionsSynchronizer.java:82)
>     at
> imagej.legacy.LegacyService.updateIJ1Settings(LegacyService.java:143)
>     at imagej.legacy.LegacyService$2.onEvent(LegacyService.java:194)
>     at imagej.legacy.LegacyService$2.onEvent(LegacyService.java:190)
>     at
> org.bushe.swing.event.ThreadSafeEventService.publish(ThreadSafeEventService.java:971)
>
> Barry is out today but I will investigate this afternoon and hopefully have
> a fix soon.
>
> -Curtis
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://imagej.net/pipermail/imagej-devel/attachments/20110921/7903d0cd/attachment.html>