| Re: [hatari-devel] Some patch suggestions |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/hatari-devel Archives
]
- To: hatari-devel@xxxxxxxxxxxxxxxxxxx
- Subject: Re: [hatari-devel] Some patch suggestions
- From: Thomas Huth <th.huth@xxxxxxxxx>
- Date: Mon, 23 Apr 2018 19:29:49 +0200
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/simple; d=posteo.de; s=2017; t=1524504591; bh=rm6g6b8Gn1vn8WAySVGjKZffhUJdFRwjzF9yq9M5TOc=; h=Date:From:To:Subject:From; b=Db5aiK1PPaLa/LCrPx59DOrLZZUSEt+MQON/4TnNcTzb+3n1ieagaZyM4vgy6Uv9r mbHoFrN9Bpq4aHnX3x1gynE7fUCY7MLxZT1cvD9R1mM5TiVkqe9zAfZwkZ3ADRCsYi reP4tkp1tdyYIG51+cx2RnxJ0Q0gAAiRWuntrUXBw48ZD76MLL5hJOgp3o6qI448F4 pe/BYEpznqh/hIKB83Ct6My9fvdHLP6id9wYx7Oy0+iLsH1tqvQxnKHEBatsAGN78q TTx/WjBdmWoAWUCy/9m9jr+juE7hCyVmTUdP6V7826a79J2wAGgGkUCuQ4b4iirEjN +ns5hSdK9NTlQ==
Am Sun, 22 Apr 2018 12:10:06 +0200
schrieb Thorsten Otto <admin@xxxxxxxxxxx>:
> while looking at the recently added test support, i have some
> suggestions which should be changed.
>
> The first patch makes sure that a0 is zero when the program is called
> (which is the GEMDOS convention for calling ordinary programs,
> otherwise the startup routine might think it is an accessory, and
> expects the value of the Basepage there).
>
> The second aborts a test program when calling unsupported GEMDOS
> functions, since the handler in the cartridge would otherwise try to
> jump to the original routine, which isn't there. It also adds the
> names of the MiNT functions for this purpose.
>
> The third uses the PC that was pushed by the exception for traces
> instead of the current PC. The current PC in that function will
> always be the return address from the cartridge, and is rather
> useless.
Thanks, patches committed now (with some small modifications).
Thomas