Unless things have changed, the combined L1/L2 controller use exclusive
caches, not inclusive.
...Greg
Derek Hower wrote:
Inclusivity/exclusivity is a coherence-level issue. Thus, it is
determined solely by the behavior of the SLICC generated protocol. To
the best of my knowledge, all released protocols in GEMS are inclusive.
-Derek
On Wed, May 13, 2009 at 10:22 AM, Edward Lee <edwl202@xxxxxxxxx
<mailto:edwl202@xxxxxxxxx>> wrote:
Hi all,
I am trying to add another cache besides L1/L2 to my target system
like a victim cache or an RAC. So, I am wondering how to determine
exclusivity of the caches in GEMS? For example,
MOESI_SMP_directory protocol has a single SLICC controller for
L1/L2, does this mean that the L2 is inclusive? Where in the GEMS
code is this determied?
Thanks for any help.
Regards,
Ed
_______________________________________________
Gems-users mailing list
Gems-users@xxxxxxxxxxx <mailto:Gems-users@xxxxxxxxxxx>
https://lists.cs.wisc.edu/mailman/listinfo/gems-users
Use Google to search the GEMS Users mailing list by adding
"site:https://lists.cs.wisc.edu/archive/gems-users/" to your search.
------------------------------------------------------------------------
_______________________________________________
Gems-users mailing list
Gems-users@xxxxxxxxxxx
https://lists.cs.wisc.edu/mailman/listinfo/gems-users
Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/" to your search.
|