Re: [hatari-devel] The external disassembler is now really external |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/hatari-devel Archives
]
- To: Christian Zietz <czietz@xxxxxxx>
- Subject: Re: [hatari-devel] The external disassembler is now really external
- From: Thomas Huth <th.huth@xxxxxxxxx>
- Date: Sun, 28 Aug 2022 08:29:47 +0000
- Cc: hatari-devel@xxxxxxxxxxxxxxxxxxx
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/simple; d=posteo.de; s=2017; t=1661675395; bh=+/gLhc5zpcduLQzPbBkLGxL5qe6JkFYTQW7vEpE3TO0=; h=Date:From:To:Cc:Subject:From; b=KIZRSJ+MPnnzLxECWCQlCFn0yPV6iExzMd+JrydKAtEkJfVD45l0S7dwBk/1suuBw f0ikEiM2y4XMHdDGsfas6Dblch/Y+7y+WhGK6U9mYOUAFXN1xiActDb+9egr3Ia1sm /kTsp4q7E/JgVAmnETR2snyo870rx/Ucw6wu1J1AoYxw84My9Csueg7jgklm3/PmOa vs3QaPvVL5CScSweIOSGjsJ0sqYl4AjiV2dgKM5YJhSNWSqx0BSRxRY8RjbT9nPgV+ KvI4ZhdeXO4EluHq3favn5CAshspu0o0I9Q5pIynouUTDYXX6HZN0MhBg2jQfcUrMv TwAhni0PxPHXg==
Am Wed, 24 Aug 2022 20:25:28 +0200
schrieb Christian Zietz <czietz@xxxxxxx>:
> Thomas Huth schrieb:
>
> >> - Most importantly: 68030 MMU opcodes are incorrectly disassembled as
> >> *FPU* instructions.
> >
> > Uh, that's bad :-( Could you maybe a bug ticket at
> > https://github.com/capstone-engine/capstone/issues ?
> > (otherwise, let me know, then I'll do it)
>
> I would appreciate if you did it. Because there might be questions like
> "how is Hatari integrating Capstone", which I could not answer.
Done now, FNOP here:
https://github.com/capstone-engine/capstone/issues/1916
and MMU instructions here:
https://github.com/capstone-engine/capstone/issues/1917
> > PS: Capstone comes with a nice command line tool for disassembly, so it's
> > even possible to verify this on the command line:
> >
> > $ cstool m68k40 0xf0394c0000e4995a
> > 0 f0 39 4c 00 fmove.x invalid, fp0
> >
> > ... that should be a pmove instead.
>
> I'm not sure this is the best way of demonstrating the bug to the
> Capstone developers. The 68040 does not have PMOVE. It has some other
> PMMU instructions, but these differ from the 68030.
Right! Unfortunately, the cstool does not seem to have a
mode for the 68030 yet, so I've tried to come up with a 040-specific
instruction in the ticket for "cstool" and reported the problem with the
68030 with some dumps from the Hatari disassembler instead.
Thomas
PS: https://github.com/capstone-engine/capstone/issues/1668 seems to be
interesting for m68k, too ... if you prefer "bra.s" instead of "bra.b",
please leave a comment there.