On 02.04.2010, at 10:12, MacArthur, Ian (SELEX GALILEO, UK) wrote:

> 
>> those of you who read the commit logs will have been blasted 
>> with huge commits in the FLTK 3.0 branch. Yes, I am actively 
>> developing this again, and guess what, so far I believe I am 
>> pretty successful. 
> 
> Yes - I had noticed.
> I even wondered if I should grab a snapshot to try it, but decided to
> wait until the churn of check-ins stopped...

Ah, see, the current branch does not have "matt" attached in the SVN anymore. 
Please feel free to hack away (and I saw you already did).

> I think you missed revising some copyright dates though! 

Yes, most definitely ;-)

>> Overhead: FLTK1 and 2 code wil have one additional 
>> indirection per call. Every widget has a sister widget which 
>> is just one pointer in size. I am sure this is bearable for 
>> most systems.
> 
> Is it feasible that the compiler can resolve that indirection at compile
> time in some cases? I'm wondering if this additional indirection my be
> optimised away in some cases (so that it would actually be no worse at
> all...) though this seems unlikely for any reasonably complicated bit of
> code...

That would be cool! But who knows, except for the gcc core developer group... .


_______________________________________________
fltk-dev mailing list
[email protected]
http://lists.easysw.com/mailman/listinfo/fltk-dev

Reply via email to