On 29 August 2014 20:00, Bastian Koppelmann
<kbast...@mail.uni-paderborn.de> wrote:
> Hi Peter,
>
>
> On 08/29/2014 03:30 PM, Peter Maydell wrote:
>> Don't use pflash_cfi01_register() in new code, it gives you a
>> weird not-like-hardware flash device that we only have for
>> backwards compatibility with existing board models. Instantiate
>> and configure the flash device directly (compare vexpress.c's
>> ve_pflash_cfi01_register(), but use the correct config for the
>> hardware you're modelling, which might not be two parallel
>> 16 bit wide flash chips).
>
>
> This board does not exists. It's purpose is just to be able to run TriCore
> binaries. So I rather just get rid of the flash drive.
OK, if you're not using it that's easy.
>>> +static QEMUMachine ttb_machine = {
>>> + .name = "tricore_testboard",
>>> + .desc = "Just for testing",
>>> + .init = tricoreboard_init,
>>> + .is_default = 1,
>>> +};
>>
>> The .desc text here appears in the user-visible help output, so
>> can we have something slightly more useful to them, please?
>>
>> $ ./build/all/tricore-softmmu/qemu-system-tricore -M help
>> Supported machines are:
>> tricore_testboard Just for testing (default)
>> none empty machine
>
>
> Same as above. The description shows exactly the board's purpose. It is just
> for testing and not a real board. How about I change it to "a minimal
> TriCore board"?
Something like that would be OK, I just want it a bit more
descriptive and less flippant.
> I see your point about the default. For now I would leave it as is, since
> there are not other boards and qemu would give an error, when launched
> without -M.
The thing about this is that it's hard to change the default
later on, because it would mean that you'd break any
existing command lines that didn't specify the machine
but were expecting the test board. So personally I think
it's best not to specify a default if there isn't one that
makes sense.
thanks
-- PMM
