P96 3.1.2: Crash when using new graphics.library in AmigaOS 3.2.1

Caution: Non registered users only see threads and messages in the currently selected language, which is determined by their browser. Please create an account and log in to see all content by default. This is a limitation of the forum software.


Also users that are not logged in can not create new threads. This is a, unfortunately needed, counter measure against spam. Please create an account and log in to start new threads.

Don't Panic. Please wash hands.
  • I'm a little puzzled by this one, as I have three Amigas, all running AmigaOS 3.2.1 with P96 3.1.2, and only one of them repros. But:


    On my A2000 Rev6, if any P96 monitors are in Devs:Monitors, my machine crashes within 30 seconds of LoadMonDrvs returning. If I comment that out in startup-sequence, the machine crashes when I manually double-click the "Generic" monitor or my ZZ9000's monitor file. It also crashes immediately when opening the P96 preferences app.


    If I remove the new graphics.library from the LoadModule directive in startup-sequence, the crash goes away. Is there a bug in P96, or a bug in the new graphics.library? Or both? How might I help narrow it down? I'm not sure where to send this bug, especially since it only repros on one out of my three machines with P96 installed. And one of the two machines that's not crashing, another A2000, has a nearly identical copy of the same boot volume that crashes on the other machine.

  • It's got a Magnum 40/4 68040 accelerator with 64MB of RAM on the accelerator, plus an ACE2B. I've only seen it crash when both the new graphics.library and P96 are activated at the same time. Otherwise it's stable. Bad memory seems unlikely, no? I'd love to just give you a crash dump to open in a debugger, but I've been told in the past by others that's not possible. I'm not sure how to debug a crash like this shooting in the dark.


    The two working machines are both 030s. One has a 2630 and a BigRAM2630 (you were just helping me with this one crashing because of the BigRAM), and the other is a stock 16MHz A3000.

  • The machine successfully got through twenty passes of AmigaTestKit's RAM test on the system's fast memory, so between that and the specificity of the crash, I think I'm personally satisfied it's not bad RAM.

  • Okay. Is there any sort of diagnostic mode on the P96 side I could enable to possibly catch any information about the crash? A call stack, a faulting task name—anything? I don’t know how to do debugging on the Amiga and only have experience on Windows, where I could just be looking at a memory dump in a debugger.

  • Please try P96 V3.2.1, released just an hour ago. Although the changes list does not show anything fixed that sounds like the crash you reported, we also don't have any other case that is remotely similar to yours. Please try a clean install of P96 V3.2.1 (the exact same version number of P96 and Amiga OS is a pure coincidence!).

  • I'm only mentioning "clean install" because a customer from Germany has fixed things with that. To be honest, I can't say how exactly he did it, but I assume that if "uninstall" is an option you can choose, it should work.

  • Sorry, I can't - I'd have to go through the install script the same way as anyone else (or run Snoopdos doing an install), but this really does not fit my schedule.

  • It looks like the installer does delete everything.


    The reason this doesn't repro on my other A2000 turns out to, in fact, be the different accelerator. The issue is triggered by a confluence of three things:

    1. Updating graphics.library with AmigaOS 3.2.1
    2. P96 being activated
    3. CSA's autoboot driver for the Magnum 40/4 SCSI controller, which gets stored in the RDB

    If I take the exact same disk and boot from a different SCSI controller, the crash goes away.

  • The last reply was more than 365 days ago, this thread is most likely obsolete. It is recommended to create a new thread instead.