Re: [Gems-users] Miss Latencies

["soohong p kim" <spkim@xxxxxxxxxx>]

Dan,

 

What’s the difference between “miss_latency_L2Cache” and “miss_latency_L2Miss”?

And as Marco asked initially, what does _wCC suffix mean?

 

I appreciate if you can confirm that these are hooked up in the release…

 

Soohong

 

From: gems-users-bounces@xxxxxxxxxxx [mailto:gems-users-bounces@xxxxxxxxxxx] On Behalf Of Mike Marty
Sent: Friday, April 18, 2008 11:54 PM
To: Gems Users
Subject: Re: [Gems-users] Miss Latencies

 

those are enabled by passing a parameter to hitCallback from the L1 controller to the Sequencer.  The individual protocol can track who responds to which request.  I find these very valuable, but I don't recall if I hooked these up for any protocol in the release.

On Fri, Apr 18, 2008 at 6:05 AM, Marco Solinas <marco.solinas@xxxxxxxxxxxx> wrote:

Hi list,

I'm a little confused by the miss latency values in the ruby stats file.

There are more values for miss latency, each labeled with different
names: miss_latency, miss_latency_LD, miss_latency_ST,
miss_latency_IFETCH, miss_latency_ATOMIC, miss_latency_L1Cache,
miss_latency_L2Cache, miss_latency_Directory, miss_latency_L1Cache_wCC,
miss_latency_L2Cache_wCC, miss_latency_L2Miss.
I can't understand the meaning of the most part of them (for example,
that are L1Cache_wCC and L2Cache_wCC?, what's the difference between
L1Cache and L1Cache_wCC, or between L2Cache and L2Cache_wCC? what does
miss_latency_Directory represent? and L2Miss?).

Can anyone please help me?
Regards
Marco

_______________________________________________
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.