Hi,
Am 18.06.2013 05:28, schrieb Peter Crosthwaite:
> On Tue, Jun 18, 2013 at 11:58 AM, Andreas Färber <afaer...@suse.de> wrote:
>> Split qdev initfn into instance_init and realize functions.
>> Change one occurrence of "klass" while at it.
>>
>> Signed-off-by: Andreas Färber <afaer...@suse.de>
>
> Reviewed-by: Peter Crosthwaite <peter.crosthwa...@xilinx.com>
>
>> ---
>> hw/intc/openpic.c | 34 +++++++++++++++++++---------------
>> 1 file changed, 19 insertions(+), 15 deletions(-)
>>
>> diff --git a/hw/intc/openpic.c b/hw/intc/openpic.c
>> index 875c6b8..2d6b05c 100644
>> --- a/hw/intc/openpic.c
>> +++ b/hw/intc/openpic.c
>> @@ -1531,8 +1531,16 @@ static void map_list(OpenPICState *opp, const MemReg
>> *list, int *count)
>> }
>> }
>>
>> -static int openpic_init(SysBusDevice *dev)
>> +static void openpic_init(Object *obj)
>> {
>> + OpenPICState *opp = OPENPIC(obj);
>> +
>> + memory_region_init(&opp->mem, "openpic", 0x40000);
>> +}
>> +
>> +static void openpic_realize(DeviceState *dev, Error **errp)
>> +{
>> + SysBusDevice *d = SYS_BUS_DEVICE(dev);
>
> FWIW, i have been using "sbd" for this variable name in similar
> conversions (sdhci, xilinx_spips, axidma, axienet and a few friends).
> There are also a few other precedents out there such as arm_gic.
So far we don't seem to have a consistent convention. I've seen busdev,
sysbusdev, d; also pcidev vs. pci_dev vs. d for PCIDevice etc.
sbd is fine with me, too. But since we're not yet consistent in using oc
rather than klass either (including in your super class RFC), do you see
a strong need to respin? Or can we just follow-up with a sed across the
tree at some point once there is agreement on the naming?
We should collect naming conventions into the QOMConventions Wiki page.
Regards,
Andreas
--
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg
