[pgpointcloud] Wrong serialized size of a PcPoint ?

[Sandro Santilli]

On Wed, Apr 08, 2015 at 01:48:23PM +0200, Rémi Cura wrote:
> Wow,
> I wonder how you got the idea to do this check!
> ending with 19 bytes may have very bad consequences for perf if it is not a
> 8 byte multiple.

I actually don't think it affects performance so much, as PostgreSQL does
its padding too. But I hadn't reached that point yet.

Another thing to check is whether the padding is only at the end of the
serialized stream or if it really follows the first 8 bytes of headers.
The former case would be easy to change/fix.

--strk;