[OpenJDK 2D-Dev] [OpenJDK Rasterizer] Path2d needRoom very slow for huge paths

Laurent Bourgès bourges.laurent at gmail.com
Fri Apr 10 19:08:12 UTC 2015


Jim,

Please consider other changes also !

I added the trim method after the discussion. I can remove it if it makes
you approve the patch quicker.

Laurent
Le 10 avr. 2015 20:51, "Jim Graham" <james.graham at oracle.com> a écrit :

> Hi Laurent,
>
> Adding a new method is not as simple as just sending a webrev.  We'll need
> to do some other internal processes to get approval for that...
>
>                 ...jim
>
> On 4/10/15 8:07 AM, Laurent Bourgès wrote:
>
>> Jim,
>>
>> Here is the new webrev:
>> http://cr.openjdk.java.net/~lbourges/path2D/Path2D_needRoom.1/
>>
>> Changes:
>> - needRoom() applies your pseudo-code; see expandPointTypes() and
>> expandCoords()
>> - added a new public trimToSize() method to Path2D implemented by both
>> Path2D.Float and Path2D.Double classes
>>
>> Cheers,
>> Laurent
>>
>> 2015-04-08 22:53 GMT+02:00 Jim Graham <james.graham at oracle.com
>> <mailto:james.graham at oracle.com>>:
>>
>>     Hi Laurent,
>>
>>     I'd probably do:
>>
>>     int newsizemin = oldcount + newitems;
>>     if (newsizemin < oldcount) {
>>          // hard overflow failure - we can't even accommodate
>>          // new items without overflowing
>>          return failure, throw exception?
>>     }
>>     int newsize = <growth algorithm computation>;
>>     if (newsize < newsizemin) {
>>          // overflow in growth algorithm computation
>>          newsize = newsizemin;
>>          ... OR ...
>>          newsize = MAX_INT;
>>     }
>>     while (true) {
>>          try {
>>              allocate newsize;
>>              break;  (or return?)
>>          } catch (OOME e) {
>>              if (newsize == newsizemin) {
>>                  throw e;
>>              }
>>          }
>>          newsize = newsizemin + (newsize - newsizemin) / 2;
>>     }
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/2d-dev/attachments/20150410/21573dfe/attachment.html>


More information about the 2d-dev mailing list