I see that the response and request latencies are used. Is the response
latency equivalent to one bank's access latency and the request latency
equivalent to the latency going through the cache controller?
Also, is the L1 cache always 1 cycle. I tried to find the fast path
parameter to turn it off, but I can't seem to find the exact parameter.
thx,
Tom
On Thu, 5 Oct 2006, Mike Marty wrote:
Various protocol controllers use different parameters. See what latency
parameters are used in the MOESI_CMP_directory-L2cache.sm file.
Latency is also determined by the interconnect.
--Mike
How are cache latencies defined when using both ruby and opal?
The values inside the opal config file seems to be ignored, and there are
several latency values when I look inside ruby's stats:
CACHE_RESPONSE_LATENCY: 12
L2_RESPONSE_LATENCY: 6
L1_RESPONSE_LATENCY: 3
L1_REQUEST_LATENCY: 2
L2_REQUEST_LATENCY: 4
L2_RECYCLE_LATENCY: 5
I am using the MOESI_CMP protocol with PT2PT links, and I wondering how I
can modify the latencies when accessing local L1 and L2 caches?
I've looked through the archive, but I am still not clear how to change
the latencies.
Tom
_______________________________________________
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.
_______________________________________________
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.
|