Bug 40610 - SIGSEGV when creating 52.000.000 objects in Parallel.For
Summary: SIGSEGV when creating 52.000.000 objects in Parallel.For
Alias: None
Product: Runtime
Classification: Mono
Component: GC ()
Version: 4.4.0 (C7)
Hardware: Macintosh Mac OS
: --- normal
Target Milestone: ---
Assignee: Bugzilla
Depends on:
Reported: 2016-04-21 13:32 UTC by Christian Hüning
Modified: 2016-10-28 09:12 UTC (History)
4 users (show)

Is this bug a regression?: ---
Last known good build:

Stacktrace (135.19 KB, text/plain)
2016-04-21 13:32 UTC, Christian Hüning
Repro Code (1.19 KB, application/octet-stream)
2016-04-21 13:35 UTC, Christian Hüning

Notice (2018-05-24): bugzilla.xamarin.com is now in read-only mode.

Please join us on Visual Studio Developer Community and in the Xamarin and Mono organizations on GitHub to continue tracking issues. Bugzilla will remain available for reference in read-only mode. We will continue to work on open Bugzilla bugs, copy them to the new locations as needed for follow-up, and add the new items under Related Links.

Our sincere thanks to everyone who has contributed on this bug tracker over the years. Thanks also for your understanding as we make these adjustments and improvements for the future.

Please create a new report on GitHub or Developer Community with your current version information, steps to reproduce, and relevant error messages or log files if you are hitting an issue that looks similar to this resolved bug and you do not yet see a matching new report.

Related Links:

Description Christian Hüning 2016-04-21 13:32:55 UTC
Created attachment 15801 [details]

I am hunting down problems with the mono sgen GC in large memory applications (> 40 GB). I finally managed to create a stacktrace, which includes the debug symbols. the stacktrace is attached, as well as the repro code. 
I executed with Mono 4.4.0 on OSX.
Comment 1 Christian Hüning 2016-04-21 13:35:14 UTC
Created attachment 15802 [details]
Repro Code
Comment 2 Vlad Brezae 2016-10-27 23:59:32 UTC
Hey Christian,

       I've tried running the test case on my osx (with 16gb RAM) and it didn't crash, even though the test case allocates up to 22 GB of heap. Are you able to get this crash reliably ? What is the configuration of your machine ? By the stack trace it looks like we are just running out of memory, situation in which we don't have many options so we just crash.

Comment 3 Christian Hüning 2016-10-28 05:16:44 UTC
Hey Vlad,

If I remember correctly that one was running inside a Docker Container inside a VM on a NUMA based system with 96 GB of memory. So when it runs out of memory utilizing less than 96 GB I thought it would be Mono's fault unless I learned that QEMU in combination with the Kernel version I was using and the NUMA architecture was unable to use all 96 GB of memory. So I think you may close this issue. Thanks!

Comment 4 Vlad Brezae 2016-10-28 09:12:54 UTC
Thank you