On 9 Oct 2009, at 23:45, Fred Kiefer wrote:
David Chisnall schrieb:
Another example is that a lot of classes call
NSDeallocateObject(self) rather than [super dealloc], which has a
negligible performance impact but breaks any category that tries to
replace NSObject's dealloc method.
I never understood why we do this.
A lot of what looks like premature optimisation nowadays was
benchmarked or profiled and found to be worthwhile on slower CPUs ten
years ago.
I can't believe that more than a few dozen cases of this particular
issue exist anymore in GNUstep (perhaps I'll try grepping and fixing).
I definitely think it should go (I hated the fact that it was totally
pervasive in Cocotron when I looked at their code).
_______________________________________________
Gnustep-dev mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/gnustep-dev
