qemu-arm
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [RFC PATCH 15/35] hw/i386/xen/xen-hvm: Emit warning when old code is


From: Philippe Mathieu-Daudé
Subject: Re: [RFC PATCH 15/35] hw/i386/xen/xen-hvm: Emit warning when old code is used
Date: Mon, 8 Jun 2020 19:37:33 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.5.0

On 6/8/20 6:54 PM, Paul Durrant wrote:
>> -----Original Message-----
>> From: Philippe Mathieu-Daudé <philmd@redhat.com>
>>
>> This code hasn't been QOM'ified yet. Warn the user.
> 
> "Based on today's IRC chat, this is a trivial RFC series
> to anotate pre-qdev/QOM devices so developers using them
> without knowing they are not QOM'ified yet can realize
> it and convert them if they have time."
> 
> So, how should this be coded then? The XenIOState doesn't really qualify as a 
> 'device', does it?

There is a pending question whether Machines are Devices or not.

Xen is a tricky case, it is created as a Q35 machine overloaded with Xen
features.

>> @@ -1401,6 +1401,8 @@ void xen_hvm_init(PCMachineState *pcms, MemoryRegion 
>> **ram_memory)
>>      xen_pfn_t ioreq_pfn;
>>      XenIOState *state;
>>
>> +    qdev_warn_deprecated_function_used();
>> +
>>      state = g_malloc0(sizeof (XenIOState));

XenIOState is indeed not a device, it is allocated once, we won't gain
anything making it a QOM object... so this patch is incorrect.

Sorry for the noise :S

Regards,

Phil.

>>
>>      state->xce_handle = xenevtchn_open(NULL, 0);
>> --
>> 2.21.3
> 
> 




reply via email to

[Prev in Thread] Current Thread [Next in Thread]