Avalanche / SharedCLibrary crash

Just tested the latest hi vector 5.23.
Avalanche crashes with the following error
Internal error. trap whilst in trap handler.Internal error
AoDT at FC171EF4
pc = FC171EEC
regs = 248A4

*where
Address &FC171EF4 is at offset &00031A5C in module ‘SharedCLibrary’
*showregs 248a4
Syntax: *ShowRegs
*showregs
Register dump (stored at &2001DF30) is:
R0 = 00026178 R1 = 00026174 R2 = 00000004 R3 = 00000010
R4 = 80000002 R5 = 00000001 R6 = 00000010 R7 = FC1732F0
R8 = FC1732F0 R9 = 0002482C R10 = 000258C0 R11 = 00000010
R12 = 03EB153C R13 = 00026158 R14 = FC1718FC R15 = FC171EF4
Mode USR32 flags set: nzCvqjggggeaift PSR = 20000010

It did this with an earlier version of the hi vector 5.23 but not the last low vector version.
Server on the other end is Fedora 23 & 24.

Rebooted all machines & this is a consistant error.
Afterwards the Avalanche window just shows a black screen & click in it crashes Avalanche. Nothing in particular shows in the Log.

Is there a way you can backtrace a error in ‘CLib’ to a jmp function?
ie offset &3297c in SharedCLib

Until very recently Avalanche would crash/bomb-out here when vertically scrolling a window seemingly too energetically. The error in the WIMP log was:

Internal error, trap while in trap handler: Internal error: abort on data transfer at &FC19BB6C, pc = FC19BB64: registers at 000248A4

With the current ROM (08-Jun-17), similarly-energetic attempts to repeat this problem have so far failed!

Edit: The problem is more difficult to provoke!

An excellent outcome, but what fairly recent changes might have improved matters?