Re: [RFC v5 08/12] accel: extend AccelState and AccelClass to user-mode

[Claudio Fontana]

Hi Eduardo,

thanks for looking at this,

On 11/24/20 6:56 PM, Eduardo Habkost wrote:
> On Tue, Nov 24, 2020 at 05:22:06PM +0100, Claudio Fontana wrote:
>> Signed-off-by: Claudio Fontana <cfontana@suse.de>
>> ---
> [...]
>> @@ -908,8 +909,12 @@ int main(int argc, char **argv)
>>      }
>>  
>>      /* init tcg before creating CPUs and to get qemu_host_page_size */
>> -    tcg_exec_init(0);
>> +    {
>> +        AccelClass *ac = accel_find("tcg");
>>  
>> +        g_assert(ac != NULL);
>> +        ac->init_machine(NULL);
> 
> Most init_machine() methods will crash if you call them with a
> NULL argument.

not tcg though,


> 
> This looks like another reason for having a
>   void accel_init(AccelState*)
> function and a
>   void (*init)(AccelState*)
> method in AccelClass.
> 
> Then the whole code block above would be as trivial as:
> 
>   accel_init(current_accel());


but this does look like an attractive result,

thanks!

CLaudio
> 
> 
>> +    }
> [...]
>>
>