[Coin-discuss] Performance of CoinMemcpyN()

Laszlo Ladanyi ladanyi at us.ibm.com
Sat Jul 30 10:32:19 EDT 2005


Yes, that could be done, and probably would be slightly faster. If you are
willing to write the specializations I'd be glad to add them to the
repository. 

Also, I'd be curious if anyone has tested whether explicit loop unrolling
produces faster code or calling memcpy.

--Laci

On Fri, 29 Jul 2005, Joseph Young wrote:

> 	Then, why not define a specialization that works with simple data 
> types such as ints, floats, chars, and doubles?
> 
> Sinc, Joseph Young
> 
> > Some things can't be copied with memcpy.
> > Not everything is POD.
> > Some have user-defined assignment operators.
> > Some have implementation-defined internal pointers.
> > My recollection is that virtual bases are handled that way.
> 
> > BTW what is the Coin module for?
> > Does it depend on anything?
> 
> _______________________________________________
> Coin-discuss mailing list
> Coin-discuss at list.coin-or.org
> http://list.coin-or.org/mailman/listinfo/coin-discuss
> 




More information about the Coin-discuss mailing list