MorphOS runs on QEmu
  • Paladin of the Pegasos
    Paladin of the Pegasos
    Yasu
    Posts: 1724 from 2012/3/22
    From: Stockholm, Sweden
    @acepeg

    Well, some AOS 4 users seems to think it's the best thing since sliced bread. Though you rarely hear anyway say they actually use it on that configuration :-P
    AMIGA FORUM - Hela Sveriges Amigatidning!
    AMIGA FORUM - Sweden's Amiga Magazine!

    My MorphOS blog
  • »08.03.16 - 11:24
    Profile Visit Website
  • Order of the Butterfly
    Order of the Butterfly
    amyren
    Posts: 219 from 2010/5/15
    From: Norway
    Quote:

    acepeg wrote:
    For me os4.1 on winuae is useless....just proof concept for geek.


    ..and geeks are probably hard to find among users of any amiga-like system:)

    You are probably right about usability in its current state. But my point is that I dont think the ppc emulation speed is that slow.
    I havent tested it so much, but to get an idea of the speed I launced and old adf game (Volfied) with runinuae, and the sound was stutterng but the game itself was playable.
    And this is running on a 3.5 years old windows tablet. I would guess that a modern desktop pc is many times faster.
  • »08.03.16 - 11:25
    Profile
  • Order of the Butterfly
    Order of the Butterfly
    Posts: 165 from 2004/11/18
    The problem is that modern entry level pc's are not faster than ten years Old core2duo. It's another problem. I agree that we are geek in fact. But what i want to explain is that nothing serious can be done with émulation. Most of classic users prefer native Amiga to winuae even it's 100times faster. So i think this is the same for Os4 users and Morphos one also. I don't say that it's bad for Morphos. For Os4 it's à différent case since they have only overpriced hardware disponible.
  • »08.03.16 - 18:24
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    Intuition
    Posts: 1110 from 2013/5/24
    From: Nederland
    This all went quiet. Did the developer give up?
    1.67GHz 15" PowerBook G4, 1GB RAM, 128MB Radeon 9700M Pro, 64GB SSD, MorphOS 3.15

    2.7GHz DP G5, 4GB RAM, 512MB Radeon X1950 Pro, 500GB SSHD, MorphOS 3.9
  • »02.05.16 - 11:14
    Profile
  • Just looking around
    dark_knight
    Posts: 11 from 2012/8/12
    There's a short testing report for qemu 2.7-rc1 on Emaculation. No significant progress has been made.
  • »07.08.16 - 18:19
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12058 from 2003/5/22
    From: Germany
    > There's a short testing report for qemu 2.7-rc1 on Emaculation.
    > No significant progress has been made.

    Thanks for reporting. The current state seems to be summarized as follows:

    "the MorphOS writers made several mistakes in their boot code using real addresses where they should have used virtual addresses. Not that this was noticed on real Apple OF since the MMU is disabled by default, however it breaks OpenBIOS which currently only runs in virtual mode."
    http://www.emaculation.com/forum/viewtopic.php?p=53511#p53511

    "MorphOS relies on the system booting up with the MMU off at first, then turning it on during the booting process. OpenBIOS has the MMU enabled all the time, which hinders MorphOS booting up."
    http://www.emaculation.com/forum/viewtopic.php?p=53948#p53949
  • »07.08.16 - 20:16
    Profile
  • Priest of the Order of the Butterfly
    Priest of the Order of the Butterfly
    KennyR
    Posts: 868 from 2003/3/4
    From: #AmigaZeux, Gu...
    Quote:

    Andreas_Wolf wrote:
    > There's a short testing report for qemu 2.7-rc1 on Emaculation.
    > No significant progress has been made.

    Thanks for reporting. The current state seems to be summarized as follows:

    "the MorphOS writers made several mistakes in their boot code using real addresses where they should have used virtual addresses. Not that this was noticed on real Apple OF since the MMU is disabled by default, however it breaks OpenBIOS which currently only runs in virtual mode."
    http://www.emaculation.com/forum/viewtopic.php?p=53511#p53511


    I'd bet they aren't mistakes but rather very much intentional.
  • »07.08.16 - 21:25
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12058 from 2003/5/22
    From: Germany
    >> https://github.com/mmuman/qemu/tree/sam460ex-WIP-rebasing
    >> http://www.amiga.org/forums/showthread.php?p=804695

    > That is interesting.

    Some progress:

    "Here's a series to add emulation of the Sam460ex to QEMU. [...] This code is based on previous work by François Revol, cleaning it up and adding some missing parts to get it to start working at least for the Linux kernel and the original U-Boot firmware of the board (which is needed to boot the Amiga like OSes easily). [...] This is not complete and cannot run most of the interesting OSes yet as there are still missing parts and likely some bugs I could not fix [...]. [...] MorphOS starts booting but does not find PCI devices so it fails. Unfortunately I don't know how to set kernel parameters to enable debug logging (passing them via the bootloader like on Mac does not work and MorphOS people haven't reply my query about it) so I don't care for now until other OSes easier to debug can't boot. (Besides, MorphOS could also run on the Mac emulation and may get a new version soon which might change a few things so it's low priority now; maybe I look at it again when new version is available.) I could not test AmigaOS because I don't have that but I would not be surprised if it would also fail during boot for some reason but if someone has a copy and can try it let me know how it fails (try -serial stdio to hopefully get some logs)."
    http://lists.gnu.org/archive/html/qemu-ppc/2017-08/msg00112.html
    http://lists.gnu.org/archive/html/qemu-devel/2017-08/msg02415.html
  • »20.08.17 - 02:37
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12058 from 2003/5/22
    From: Germany
    Update:

    > The current state seems to be summarized as follows: [...]

    The current state (as of March 2017) of getting MorphOS to run inside QEMU's G4 Mac emulation ("mac99p"), which is not to be confused with the effort by the same author of emulating the Sam460 in QEMU for the purpose of running MorphOS and other operating systems (see previous comment), seems to be as follows:

    "the USB controller is found [...]. [...] MorphOS now has input but still lacks some way to output video (this still needs some work)."
    http://lists.gnu.org/archive/html/qemu-ppc/2016-11/msg00276.html

    "this fixes mouse under MorphOS as well."
    http://lists.gnu.org/archive/html/qemu-ppc/2017-03/msg00189.html
  • »20.08.17 - 08:41
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4977 from 2009/1/28
    From: Delaware, USA
    Do these applications rely on KVM for virtualization?
    "Never attribute to malice what can more readily explained by incompetence"
  • »23.08.17 - 10:34
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    Intuition
    Posts: 1110 from 2013/5/24
    From: Nederland
    Quote:

    Jim wrote:
    Do these applications rely on KVM for virtualization?


    Running on a PPC CPU o you could virrualise the CPU with KVM, on an Intel CPU it would obviously have to emulate the PPC CPU.
    1.67GHz 15" PowerBook G4, 1GB RAM, 128MB Radeon 9700M Pro, 64GB SSD, MorphOS 3.15

    2.7GHz DP G5, 4GB RAM, 512MB Radeon X1950 Pro, 500GB SSHD, MorphOS 3.9
  • »23.08.17 - 21:46
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12058 from 2003/5/22
    From: Germany
    Update:

    > Some progress: [...]

    "Current status for OSes I've tried:
    - AROS boots (after I've fixed some bugs in it [...]) but keyboard doesn't work [...] and time is going slow [...]
    - Linux: kernel boots but hangs during user space [...]
    - AmigaOS 4 seems to boot but display is not working so not usable [...]
    - MorphOS does not boot (it seems to either deliberately do stuff to prevent it from running on QEMU or it has bugs but developers are not communicative about it)
    "
    http://lists.gnu.org/archive/html/qemu-ppc/2018-02/msg00189.html
  • »15.02.18 - 23:05
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4977 from 2009/1/28
    From: Delaware, USA
    Quote:

    acepeg wrote:
    The problem is that modern entry level pc's are not faster than ten years Old core2duo. It's another problem. I agree that we are geek in fact. But what i want to explain is that nothing serious can be done with émulation. Most of classic users prefer native Amiga to winuae even it's 100times faster. So i think this is the same for Os4 users and Morphos one also. I don't say that it's bad for Morphos. For Os4 it's à différent case since they have only overpriced hardware disponible.


    Core i3 and i5 systems ARE faster than Core Duo systems (I own both), and a relatively cheap Ryzen based system is also superior.
    Emulation COULD provide adequate performance, IF it wasn't focused on low end cpus like the 603e.
    And while I own legacy Amiga hardware, I am increasingly defaulting to emulation and FPGA platforms.

    So...obviously I disagree with the majority of your stated opinions.

    Also, a MorphOS X64 port could benefit from the use of something like QEMU in order to run our legacy ABox environment and its applications.

    The future is not the retention of a diminishing based of legacy hardware, its moving on to bigger and better systems.
    "Never attribute to malice what can more readily explained by incompetence"
  • »16.02.18 - 16:11
    Profile
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Andreas_Wolf
    Posts: 12058 from 2003/5/22
    From: Germany
    > Emulation COULD provide adequate performance, IF it wasn't focused
    > on low end cpus like the 603e.

    The more complex the emulated CPU, the slower the emulation.
  • »16.02.18 - 18:46
    Profile
  • Just looking around
    dark_knight
    Posts: 11 from 2012/8/12
    The lack of available documentation seems to be the biggest hurdle for emulation. An efficient emulation environment can be developed if you have access to the source. However, it's difficult to get hold of this hardware documentation without signing contracts that limit your ability to re-implement it, so developers must work it out for themselves and spend much of their time writing workarounds to cater for various scenarios.
  • »16.02.18 - 23:05
    Profile
  • Jim
  • Yokemate of Keyboards
    Yokemate of Keyboards
    Jim
    Posts: 4977 from 2009/1/28
    From: Delaware, USA
    Quote:

    Andreas_Wolf wrote:
    > Emulation COULD provide adequate performance, IF it wasn't focused
    > on low end cpus like the 603e.

    The more complex the emulated CPU, the slower the emulation.


    Not necessarily true. Cpus that feature more advanced/powerful instructions can be emulated utilizing similar advanced instructions in the target processor. Emulation could be multi-threaded to some degree (particularly when emulating out of order cpus and cpus with speculative branching) allowing for some SMP support (even in situations where the emulated cpu is running non-SMP code).
    A really fast emulated 603e is still not going to be a match for a more complex emulated PPC cpu, even with the added overhead needed to emulate the more advanced component.
    And the limitations of the earlier component WILL negatively impact its ultimate capabilities.

    Adding more advanced features to early cpus is ALWAYS a bonus.
    To give you an example, we were initially considering added HD6309 instructions to an MC6809 FGPA project I have been working on.
    This would give us two additional 8 bit accumulators, an additional 16 bit accumulator and a new 32 bit accumulator, as well as new instructions like a divide instruction (to complement the 6809's multiplication instruction).
    That's still under consideration, but I'm also looking into simply using the libraries the FPGA itself is capable of executing.

    How does this relate to software emulation?
    Pretty directly.
    Efficiency and speed ARE affected by complexity, BUT how the software initially analyses the code it is to execute, and what options/paths it takes to accomplish the emulation are affected by both what it is running on (the capabilities of the host system) AND the capabilities of that which is to be emulated.

    To simplify, I could build one heck of a emulated 6809 system, with multiple cores, very high speed...greatly improved power over the older silicon, BUT I'd always be better off emulating a 68000 (or '020, '040, '060).
    Similarly, an emulated 603e is better than an emulated 68000.
    SO, it seems fairly obvious that a more powerful PPC WOULD lead not better emulation, not necessarily slower.

    [ Edited by Jim 17.02.2018 - 11:58 ]
    "Never attribute to malice what can more readily explained by incompetence"
  • »17.02.18 - 16:55
    Profile
  • Priest of the Order of the Butterfly
    Priest of the Order of the Butterfly
    polluks
    Posts: 778 from 2007/10/23
    From: Gelsenkirchen,...
    Quote:

    Andreas_Wolf schrieb:
    Further progress:
    ...

    Aha, day X is coming closer.
    Pegasos II G4: MorphOS 3.9, Zalman M220W · iMac G5 12,1 17", MorphOS 3.18
    Power Mac G3: OSX 10.3 · PowerBook 5,8: OSX 10.5, MorphOS 3.18
  • »29.06.18 - 23:50
    Profile
  • Paladin of the Pegasos
    Paladin of the Pegasos
    Intuition
    Posts: 1110 from 2013/5/24
    From: Nederland
    Quote:

    Raf_MegaByte wrote:
    Quote:



    This screenshot tells almost everything. In detail - bigfoot made some patches for qemu to make it emulate the Pegasos2. And no, MorphOS 1.4 won't run on it due to some bug in OF emulation.

    [ Edited by scf on 2008/4/30 14:30 ]


    Last time I used virtual machines like QEMU or PearPC on Intel X86, these were uncapable to emulate a PPC clocked more than 75 MHz...

    A lightweight OS like MorphOS (or even AmigaOS) could run in such slooooow speed, but it is unusable for many uses, like playing DVD videos that require speed clocked minimum at 500/600 MHz.

    If the PPC virtual machine was not being DRAMATICALLY enhanched and Clock Speed raised up, then the emulated machine it is almost unusable.

    IMHO nice proof of concept, but nothing else.

    [ Edited by Raf_MegaByte on 2008/5/1 12:53 ]

    [ Edited by Raf_MegaByte on 2008/5/1 13:03 ]



    QEMU PPC-JIT on Threadripper vs QEMU KVM on a G5.

    https://image.ibb.co/koER1d/download_20180624_183327.png
    1.67GHz 15" PowerBook G4, 1GB RAM, 128MB Radeon 9700M Pro, 64GB SSD, MorphOS 3.15

    2.7GHz DP G5, 4GB RAM, 512MB Radeon X1950 Pro, 500GB SSHD, MorphOS 3.9
  • »30.06.18 - 18:45
    Profile
  • Leo
  • Order of the Butterfly
    Order of the Butterfly
    Leo
    Posts: 417 from 2003/8/18
    It boots :)

    morphos-qemu.png
    Nothing hurts a project more than developers not taking the time to let their community know what is going on.
  • »04.07.18 - 12:43
    Profile Visit Website
  • Priest of the Order of the Butterfly
    Priest of the Order of the Butterfly
    polluks
    Posts: 778 from 2007/10/23
    From: Gelsenkirchen,...
    Quote:

    Leo schrieb:
    It boots :)

    morphos-qemu.png

    Aha, please execute Code:
    c:cpu
    Pegasos II G4: MorphOS 3.9, Zalman M220W · iMac G5 12,1 17", MorphOS 3.18
    Power Mac G3: OSX 10.3 · PowerBook 5,8: OSX 10.5, MorphOS 3.18
  • »04.07.18 - 21:19
    Profile
  • Leo
  • Order of the Butterfly
    Order of the Butterfly
    Leo
    Posts: 417 from 2003/8/18
    Quote:

    polluks wrote:
    Quote:

    Leo schrieb:
    It boots :)

    morphos-qemu.png

    Aha, please execute Code:
    c:cpu





    Ht1xC4a.png
    Nothing hurts a project more than developers not taking the time to let their community know what is going on.
  • »05.07.18 - 08:48
    Profile Visit Website
  • Leo
  • Order of the Butterfly
    Order of the Butterfly
    Leo
    Posts: 417 from 2003/8/18
    Quote:

    Papiosaur wrote:
    Excellent news !

    Thanks Leo for the snapshot.

    It usable ?




    Not really, not yet:

    - it crashes frequently (I guess its due to qemu emulation)
    - it's slow (it takes at least 2min to open OWB & several minutes to show installation wizard)
    - it's 2D only and I guess there's no sound

    But still, I think it's a good thing to have the ability to run it on any platform. For example I can now finally test cross compiled apps I created. Before that I couldn't do it unless I decided to buy a machine just for that.

    [ Edited by Leo 05.07.2018 - 13:01 ]
    Nothing hurts a project more than developers not taking the time to let their community know what is going on.
  • »05.07.18 - 12:59
    Profile Visit Website