Date: | Fri, 7 Dec 2007 22:09:25 -0800 (PST) |
---|---|
From: | "Dave Z." <zhu_dave@xxxxxxxxx> |
Subject: | Re: [Gems-users] Same L1D cache miss and L2 cache miss in MESI_CMP_filter_directory |
Hi, I don't know the specific details of the protocol you're working on, but if the *-L1cache.sm uses profile_miss rather than profile_L1Cache_miss and you set REMOVE_SINGLE_CYCLE_DCACHE_FAST_PATH to true, then you might get the same number of misses for both L1D and L2. At least, that was the case for me... Hope it helps, Dave ----- Original Message ---- From: Arnab Sinha <sinha@xxxxxxxxxxxxx> To: gems-users@xxxxxxxxxxx Sent: Friday, December 7, 2007 8:29:31 AM Subject: [Gems-users] Same L1D cache miss and L2 cache miss in MESI_CMP_filter_directory Hi,
I am simulating some SPLASH benchmarks (in LogTM framework) in GEMS 2.0 with the coherence protocol ʽMESI_CMP_filter_directoryʼ. In most of the benchmarks, I am getting same number of L1 data cache misses and L2 misses. Is it because that I set REMOVE_SINGLE_CYCLE_DCACHE_FAST_PATH: true?
Can anybody suggest?
Thanks, Arnab -----Inline Attachment Follows----- _______________________________________________ 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. Looking for last minute shopping deals? Find them fast with Yahoo! Search. |
[← Prev in Thread] | Current Thread | [Next in Thread→] |
---|---|---|
|
Previous by Date: | Re: [Gems-users] MOESI_CMP_directory, Dave Z. |
---|---|
Next by Date: | [Gems-users] Generating a new line address in Ruby, Tushar Krishna (tkrishna@xxxxxxxxxxxxx) |
Previous by Thread: | [Gems-users] Same L1D cache miss and L2 cache miss in MESI_CMP_filter_directory, Arnab Sinha |
Next by Thread: | [Gems-users] surgeclient error when creating checkpoints, wangzuo |
Indexes: | [Date] [Thread] |