D pointers

[Kuba Ober]

On Saturday 01 October 2005 11:16, Stephan Kulow wrote:
> Am Samstag, 1. Oktober 2005 17:00 schrieb Thiago Macieira:
> > The optimising compiler will very likely elect this->d to be cached in a
> > register, just like this itself is.
>
> Yeah, right. Glad we have soo many registers on i386.

It's a moot argument unless you run kde on a 66MHz i486 system.

On recent enough processors (PIII,PIV) the notion of (E)AX,BX,CX,DX registers 
is just a sad leftover from old times. The hardware is implementing virtual 
registers, and it's clever enough (most of the time) in doing so. Oftentimes 
on PIII/PIV there's no real performance difference between having something 
'in a register' to having it on stack. The CPU doesn't execute the x86 
assembly anymore, it executes said assembly translated to risc code that runs 
on real hardware that bears little resemblance to the original i386 design. 
When you have a nifty hardware-based, parallelized real-time code translator 
running under the hood, things change.

Cheers, Kuba