1. Forum
  2. USENET
  3. comp.sys.atari.st

  • GNU/GCC 3.3.6

    From Francois LE COAT@lecoat@atari.org to comp.sys.atari.st on Thu Mar 11 22:00:04 2021
    From Newsgroup: comp.sys.atari.st

    Hi,
    What about proposing a complete ATARI development system, based on
    GNU/GCC 3.3.6 ? The GNU/GCC community has almost abandoned the m68k
    processor target. We will probably have to migrate to llvm/clang
    system if we want to develop for ATARI compatible computers, further ... Personally I can't use GNU/GCC more recent version than 3.3.6, to
    build my ATARI software. And I'm probably not an isolated case ...
    We need a free reference ATARI system development kit, that would
    allow building all software, including historical devels like mine :-)
    The ATARI devels, and system software are not evolving in a good way !
    Because it is based on evolution of GNU/GCC, that neglects m68k target.
    Best regards,
    --
    Frantois LE COAT
    Author of Eureka 2.12 (2D Graph Describer, 3D Modeller) http://eureka.atari.org/
    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Francois LE COAT@lecoat@atari.org to comp.sys.atari.st on Sun Jun 5 15:45:15 2022
    From Newsgroup: comp.sys.atari.st

    Hi,
    Is EmuTOS portable on CPU x86 and the ATARI VCS platform, where it
    is originating, since there is GEM on the Amstrad PC 1512 thanks
    to Digital Research ? The ATARI VCS is a well defined architecture,
    that allows to boot many different operating systems. Why not EmuTOS ?
    Thanks,
    Francois LE COAT writes:
    What about proposing a complete ATARI development system, based on
    GNU/GCC 3.3.6 ? The GNU/GCC community has almost abandoned the m68k
    processor target. We will probably have to migrate to llvm/clang
    system if we want to develop for ATARI compatible computers, further ...

    Personally I can't use GNU/GCC more recent version than 3.3.6, to
    build my ATARI software. And I'm probably not an isolated case ...

    We need a free reference ATARI system development kit, that would
    allow building all software, including historical devels like mine :-)

    The ATARI devels, and system software are not evolving in a good way ! Because it is based on evolution of GNU/GCC, that neglects m68k target.Best regards,
    --
    Frantois LE COAT
    Author of Eureka 2.12 (2D Graph Describer, 3D Modeller) http://eureka.atari.org/
    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From =?UTF-8?Q?Vincent_Rivi=c3=a8re?=@vincent.riviere@freesbee.fr to comp.sys.atari.st on Mon Jun 6 08:57:15 2022
    From Newsgroup: comp.sys.atari.st

    On 05/06/2022 |a 15:45, Francois LE COAT wrote:
    Is EmuTOS portable on CPU x86 and the ATARI VCS platform, where it
    is originating, since there is GEM on the Amstrad PC 1512 thanks
    to Digital Research ? The ATARI VCS is a well defined architecture,
    that allows to boot many different operating systems. Why not EmuTOS ?

    Indeed, Digital Research GEM sources were originally written for x86. EmuTOS used them on 68000, and filled the gaps with a new BIOS for Atari ST, and a lot of adjustments. Current EmuTOS sources only support 68000/ColdFire processors. Porting it to a new CPU would require to rewrite the assembler sources, and define a new API for user programs. Such OS would not run 68000 Atari programs, unless some emulation layer is provided.

    So this is unlikely to happen.
    --
    Vincent Rivi|?re
    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Francois LE COAT@lecoat@atari.org to comp.sys.atari.st on Tue Jun 7 16:00:22 2022
    From Newsgroup: comp.sys.atari.st

    Hi,
    Vincent Rivi*re writes:
    Francois LE COAT wrote:
    Is EmuTOS portable on CPU x86 and the ATARI VCS platform, where it
    is originating, since there is GEM on the Amstrad PC 1512 thanks
    to Digital Research ? The ATARI VCS is a well defined architecture,
    that allows to boot many different operating systems. Why not EmuTOS ?>
    Indeed, Digital Research GEM sources were originally written for x86.
    EmuTOS used them on 68000, and filled the gaps with a new BIOS for Atari
    ST, and a lot of adjustments. Current EmuTOS sources only support 68000/ColdFire processors. Porting it to a new CPU would require to
    rewrite the assembler sources, and define a new API for user programs. > Such OS would not run 68000 Atari programs, unless some emulation layer
    is provided.

    So this is unlikely to happen.
    I'm a EmuTOS user, and I'm not involved in its developments... But
    isn't it the goal of EmuTOS, to be available on as many hardware
    configurations as it is possible ? It is being ported to the ARM CPU
    <https://github.com/kelihlodversson/pTOS>
    and it would be appreciated on newest ATARI console indeed. It only
    contains 12% of assembly language according to the GitHub WEB site.
    There's many different x86 hardware, and many different ARM hardware,
    but the ATARI VCS is a unique one, and I suspect it would be simpler
    porting EmuTOS to a well defined architecture like the VCS. I'm sure
    it would be welcome by the ATARI new developers and new users =)
    Don't you think so ?
    Best regards,
    --
    Frantois LE COAT
    Author of Eureka 2.12 (2D Graph Describer, 3D Modeller) http://eureka.atari.org/
    --- Synchronet 3.21d-Linux NewsLink 1.2