[Mono-list] Finding leaks from root pinned objects

Chris Jurney jurney at gmail.com
Fri Jul 13 20:26:09 UTC 2012

Thanks... I'm pretty sure it's a pinned gc handle.  I guess my next step is
to either modify the heapshot UI to let me see the data I need, or try to
find a way to run time inspect the heap to get some data on these
allocations.  Any mono pro tips on ways to get at these allocations would
be appreciated.

On Fri, Jul 13, 2012 at 12:32 PM, Rodrigo Kumpera <kumpera at gmail.com> wrote:

> Misc Roots include internal runtime structures and gc handles.
> You can run the profiler manually on console and ask for both, but I
> believe you would not be able to see both at the same time on the UI.
> On Thu, Jul 12, 2012 at 11:07 PM, Primitive <jurney at gmail.com> wrote:
>> I've got a leak in my mono app, and I'm having trouble tracking it down.
>>  I'm
>> hoping I'm just missing something in heapshot that will show me the data I
>> need.  I can see from the summary that I'm leaking pinned references from
>> "Misc Root" to System.Byte[] using heapshot.  I can also see a log of all
>> my
>> System.Byte[] allocations using the logging profiler to track allocations
>> and callstacks.  What I can't see is which specific allocations are pinned
>> and referenced by root.  Almost all the root referenced Byte[]s are leaks,
>> so being able to see a list of the callstacks that generated them would
>> show
>> me the problem.  Is there a way to get a combination of these two features
>> to see what I need?
>> Thanks for any help.
>> --
>> View this message in context:
>> http://mono.1490590.n4.nabble.com/Finding-leaks-from-root-pinned-objects-tp4650390.html
>> Sent from the Mono - General mailing list archive at Nabble.com.
>> _______________________________________________
>> Mono-list maillist  -  Mono-list at lists.ximian.com
>> http://lists.ximian.com/mailman/listinfo/mono-list
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ximian.com/pipermail/mono-list/attachments/20120713/46c663ce/attachment-0001.html>

More information about the Mono-list mailing list