Chris Forbes <chr...@ijw.co.nz> writes:

> Ideally, we wouldn't need init() and you could just have a
> copy-constructor do the right thing, I think -- currently, all the
> backend_reg fields get initialized 3 times:
>
> - backend_reg ctor
> - init() blasting everything to zero
> - this assignment
>
Yeah, I completely agree, but this is nothing new.  Changing that
probably belongs to a separate clean-up patch...

>
>
> On Tue, Dec 3, 2013 at 9:22 AM, Francisco Jerez <curroje...@riseup.net> wrote:
>> Chris Forbes <chr...@ijw.co.nz> writes:
>>
>>> +fs_reg::fs_reg(const backend_reg &reg)
>>> +{
>>> +   init();
>>> +   *static_cast<backend_reg *>(this) = reg;
>>>
>>> This is a bit gross...
>>
>> Would you prefer 'this->operator=(reg);'?
>>
>> Meh... :)

Attachment: pgpfPqIxbB0EH.pgp
Description: PGP signature

_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Reply via email to