Re: port-m68k/4265 (crashme freezes NetBSD on an hp433)

[Izumi Tsutsui <tsutsui%ceres.dti.ne.jp@localhost>]

The following reply was made to PR port-m68k/4265; it has been noted by GNATS.

From: Izumi Tsutsui <tsutsui%ceres.dti.ne.jp@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: mlelstv%serpens.de@localhost, tsutsui%ceres.dti.ne.jp@localhost
Subject: Re: port-m68k/4265 (crashme freezes NetBSD on an hp433)
Date: Sun, 4 Sep 2022 21:56:42 +0900

 >  tsutsui%NetBSD.org@localhost writes:
 >  
 >  >The following test code also triggers the problem.
 >  >---
 >  >int
 >  >main(int argc, char *argv[])
 >  >{
 >  > asm(".short 0xf244");
 >  > asm(".short 0xf000");
 >  >}
 >  
 >  I've tried this on a 68040 emulator (uae), where it first
 >  didn't fail at all as the emulator ignored the upper bits
 >  of the extension word.
 >  
 >  After adding a check for the extension word I get a normal
 >  core dump.
 
 On my LUNA-II, when the freeze is triggered, it doesn't respond
 even an NMI switch on it.
 
 I have a feeling that this is caused by (undocumented) processor
 internal behavior around the F-line exception. In that case
 emulators could not reproduce these behaviors, as the UAE doesn't
 reproduce XC68LC040 FPU_EMULATE troubles.
 (Note it works on real MC68LC040).
 
 >  The trap routines in NetBSD/amiga and NetBSD/hp300 differ.
 
 Yes, we should merge more locore.s routines (like busaddrerr.s)
 but currently I cannot find major differences that can cause
 visible bahaviors among them.
 
 ---
 Izumi Tsutsui