Re: [ros-dev] RE: [ion] 19742: - Implement RtlPrefectMemoryNonTemporal.Patch by Patrick Baggett <[e-mail]> and myself.

[Alex Ionescu]

Ge van Geldorp wrote:

> >  From:
> > ion(a)svn.reactos.com
> >
> > +
> > +@RtlPrefetchMemoryNonTemporal@8:
> > +	ret         /* Overwritten by ntoskrnl/ke/i386/kernel.c if SSE is
> >     
> >  
> supported (see Ki386SetProcessorFeatures() ) */
>
> Is this routine so time-critical that we want to resort to self-modifying
> code?
>   
>  
Yup, and there are many more cases where NT does this:
- Syscall vs INT2e
- Cmpxch8 vs cmpxchg x2 + spinlock
- XMMI Page Zeroing vs regular Page Zeroing
- Prefetching

And I'm probably missing more. There are critical APIs that would be 
very slow should the check be done each time they are called.

Best regards,
Alex Ionescu

>  GvG
>
>   
>