Re: [hatari-devel] WinUAE and 030 cache hits/misses?

[ Thread Index | Date Index | More Archives ]


The patch seems OK to me.

I've done a quick test with it for cache/non cache timings, but there are still things to fix before this.
It didn't change many things as I first expected.

Still some work to do ;)

Thanks Eero,

Best regards


Le 24/01/2013 00:20, Eero Tamminen a écrit :

On maanantai 21 tammikuu 2013, laurent.sallafranque@xxxxxxx wrote:
If you add it, I can quickly improve the cycles counting by taking into
account the hit/miss flag. I'm sure this would increase the quality of
the approximative cycles computing I'm curently doing (and maybe let
some timing critic programs run a bit better), especially the programs
that deals with the DSP for music.
Attached is a preliminary patch for testing this, but I noticed some
wierd things with it:

* There can be 0-3 cache misses *per* instruction (prefetch), I'm assuming
   this to mean that several instructions that were not in cache can get
   fetched with the same prefetch.  Comments on that?

* With WinUAE core, CPU instruction cycle counts are zero when debugger
   is called.   With old UAE core they're fine.  Are debugger calls in
   WinUAE in similar places as they are in old UAE CPU core?

	- Eero

Mail converted by MHonArc 2.6.19+