Changes between Version 3 and Version 4 of April 2012 MeetingRethinkingRendering


Ignore:
Timestamp:
Apr 19, 2012 11:59:29 AM (9 years ago)
Author:
dpranke@chromium.org
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • April 2012 MeetingRethinkingRendering

    v3 v4  
    6363  * jchaffraix: dunno, need to redesign counters? possible could be done during pre layout
    6464
     65* render trees use raw pointers are over the place
     66  * could we do something like weak ptrs instead so that collection is well defined (as long as it doesn't impact performance)
     67  * what happened to the experiment of holding references to dom nodes?
     68    * jamesr: 5% perf impact (follow up w/ ojan?)
     69    * rniwa: you could hold on to a lot of nodes until script invocation ends
     70      * rniwa: needs a lot more investigation
     71
     72* rniwa: jamesr, are you adding assertions / invariants to the code?
     73  * jamesr: did some, but there were too many exceptions to be able to turn things on
     74  * jamesr: whenever an invariant failed, it always lead to a security bug, it seemed
     75  * jamesr: we need to document what the invariants should be (e.g., anonymous objects shouldn't have a layer)
     76  * jchaffraix: perhaps this is part of "create a new object"
     77 
     78* jchaffraix: the whole problem of the render tree is corner cases - you always forget one (or 10%)
     79    * rniwa: e.g., people always forget editing / designMode
     80  * jchaffraix: responsibility is also spread out, e.g. overflow
     81  * smfr: a lot of this has been done for optimization
     82  * jchaffraix: should we be more strict about who is allowed to know about who?
     83    * smfr: probably, but we're very performance-sensitive
     84    * eseidel: we might be too sensitive here
    6585
    6686
     
    6989
    7090
     91
     92