| Date: | Tue, 18 Jul 2006 16:18:21 -0500 |
|---|---|
| From: | Mike Marty <mikem@xxxxxxxxxxx> |
| Subject: | Re: [Gems-users] MOESI_SMP_hammer question |
We added an optimization to the AMD Hammer protocol so that data is not
returned twice on any sharing miss. This requires meta-state to the
memory controller. It isn't really a directory.
--Mike Perry Wang wrote: Hi, I’m new to Ruby and couldn’t find much info in the list archive so please bear with me. I was under the impression that opteron based systems (< 8P) implement broadcast based MOESI, so I’m a bit confused about why there’s a directory implemented here MOESI_SMP_hammer-dir.sm Could this be modeling after larger systems using, say, HORUS chip? thanks in advance _______________________________________________ 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. |
| [← Prev in Thread] | Current Thread | [Next in Thread→] |
|---|---|---|
| ||
| Previous by Date: | [Gems-users] MOESI_SMP_hammer question, Perry Wang |
|---|---|
| Next by Date: | Re: [Gems-users] MOESI_SMP_hammer question, Mike Marty |
| Previous by Thread: | [Gems-users] MOESI_SMP_hammer question, Perry Wang |
| Next by Thread: | Re: [Gems-users] MOESI_SMP_hammer question, Mike Marty |
| Indexes: | [Date] [Thread] |