On Thu, Dec 19, 2013 at 9:15 AM, Andreas Färber <afaer...@suse.de> wrote:
> Am 17.12.2013 02:40, schrieb Peter Crosthwaite:
>> As per current QOM conventions.
>>
>> Signed-off-by: Peter Crosthwaite <peter.crosthwa...@xilinx.com>
>> ---
>>
>> hw/char/cadence_uart.c | 2 ++
>> 1 file changed, 2 insertions(+)
>>
>> diff --git a/hw/char/cadence_uart.c b/hw/char/cadence_uart.c
>> index f18db53..2f19a53 100644
>> --- a/hw/char/cadence_uart.c
>> +++ b/hw/char/cadence_uart.c
>> @@ -110,7 +110,9 @@
>> #define CADENCE_UART(obj) OBJECT_CHECK(UartState, (obj), TYPE_CADENCE_UART)
>>
>> typedef struct {
>> + /* <private> */
>> SysBusDevice parent_obj;
>> + /* <public> */
>
> Note that it's /*< ... >*/ if I'm not mistaken.
Thanks, will fix v2.
>
> However since this is outside include/ it won't get parsed by the
> patches flying around. This being Cadence IP that is likely to be reused
> on other SoCs, I do guess we'll end up with an
> include/hw/char/cadence_uart.h at some point.
>
Well ideally all peripherals are headerified for the sake of
consistency. I dont think we should be making judgement calls on
coding style based on target usages.
I'm patching old style files as I notice issue just in general
browsing/developing hopefully to one day bring them into line.
Regards,
Peter
> Cheers,
> Andreas
>
>>
>> MemoryRegion iomem;
>> uint32_t r[R_MAX];
>
> --
> SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
> GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg
>
