Re: [Gems-users] PerfectCacheMemory in MOESI_CMP_directory

[Huan Fang <huanf@xxxxxx>]

Thanks for the really quick reply. 

Yes, there is localDirectory.deallocate when L1 wants to write back, but 
you can allocate it anytime without any replacement,right?
> Yes, it is an infinitely growing cache. 
> It is used as a hack to implement an on-chip directory.  But if I 
> recall correctly, I use deallocate to remove something??
> --Mike
>
>
> On Tue, Apr 15, 2008 at 6:43 PM, Huan Fang <huanf@xxxxxx 
> <mailto:huanf@xxxxxx>> wrote:
>     Hello everyone,
>
>     I know there is DirEntry which is a PerfectCacheMemory in
>     MOESI_CMP_directory protocol. After having a look at the
>     PerfectCacheMemory.h, I found the ::cacheAvail returns always true and
>     ::cacheProbe is not implemented. Does that mean it's an *infinitely*
>     large cache?
>     I am curious about the intention to use this structure in MOESI,
>     because
>     it's not present in MSI_MOSI protocol. Maybe it's intented to use this
>     simple structure for improving performance, but is it meaningful for
>     study if it has infinite size? Please point out if I have any
>     misunderstandings.
>
>     Best Regards,
>     Huan
>     _______________________________________________
>     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.
>
>