kern/50457: putter_ilist corruption

[martin%NetBSD.org@localhost]

>Number:         50457
>Category:       kern
>Synopsis:       putter_ilist corruption
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sat Nov 21 14:20:00 +0000 2015
>Originator:     Martin Husemann
>Release:        NetBSD 7.99.21
>Organization:
The NetBSD Foundation, Inc.
>Environment:
System: NetBSD thirdstage.duskware.de 7.99.21 NetBSD 7.99.21 (MODULAR) #414: Sat Nov 21 12:20:44 CET 2015 martin%thirdstage.duskware.de@localhost:/usr/src/sys/arch/sparc64/compile/MODULAR sparc64
Architecture: sparc64
Machine: sparc64
>Description:

Sometimes during file system tests (but not exactly reproducable) I see
strange effects from a userland process dying (for unrelated reasons; have
not been able to catch a crash dump of that yet).

This leaves the system in a semi-corrupted state: a rump mount is still
there, but no rump process. On next reboot, when trying to detach
putter, the kernel panics due to a memory fault in putter_detach, apparently
putter_ilist has been corrupted (somehow).

>How-To-Repeat:
No idea yet, have seen it several times but could not easily reproduce
it. Always happend durning a full atf test run for me.

>Fix:
n/a