[Mono-dev] SEGV during nursery scanning

[Vlad Brezae]

Hey Neale,

       How are you getting this crash ? Is it reliable ? Do you know what commit seems to trigger it ?

       You can try running with MONO_GC_DEBUG=nursery-canaries for additional logging.

Vlad

From: Mono-devel-list <mono-devel-list-bounces at lists.dot.net> on behalf of Neale Ferguson <neale at sinenomine.net>
Date: Wednesday, 12 July 2017 at 00:35
To: Mono-Devel <mono-devel-list at lists.ximian.com>
Subject: [Mono-dev] SEGV during nursery scanning

I recently started getting the following SEGV. It appears the stable field is offset by 4 bytes:


#3  <signal handler called>

#4  sgen_client_par_object_get_size (vtable=0x8013150000000000, o=0x20001002060) at sgen-gc.c:3752

#5  0x000000008028a4c4 in sgen_safe_object_get_size (obj=0x20001002060) at ../../mono/sgen/sgen-gc.h:756

#6  pin_objects_from_nursery_pin_queue (ctx=..., do_scan_objects=0) at sgen-gc.c:634

#7  pin_objects_in_nursery (do_scan_objects=do_scan_objects at entry=0, ctx=<error reading variable: value has been optimized out>) at sgen-gc.c:715

#8  0x00000000802911ea in collect_nursery (reason=reason at entry=0x803b9c04 "Nursery full", is_overflow=is_overflow at entry=0, unpin_queue=0x0) at sgen-gc.c:1719

#9  0x0000000080291d6c in sgen_perform_collection (requested_size=requested_size at entry=4096, generation_to_collect=<optimized out>, reason=0x803b9c04 "Nursery full", wait_to_finish=wait_to_finish at entry=0, stw=stw at entry=1) at sgen-gc.c:2504


The only changes I have made is removing the PROFILING code from mini-s390x.c.


Neale
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.dot.net/pipermail/mono-devel-list/attachments/20170713/83b07c45/attachment.html>