Re: [hatari-devel] Suspected Hatari 2.3.0 regression (prefetch?)

[Nicolas Pomarède <npomarede@xxxxxxxxxxxx>]

Le 16/12/2020 à 19:03, Laurent Sallafranque a écrit :
> This brings me a little question : is the extra CPU cycles time used to 
> cross the MMU tables by the 68030 taken into account by Hatari ?

Hi

in some parts I think that yes, but as the MMU / prefetch is using some 
kind of pipeline to try to run memory accesses in parallel as much as 
possible and this inner work is not completely known, you're likely to 
have difference in cycle counting for some cases in the end (the caches 
make it even more complex)

Nicolas