If I set the FAST_PATH to true, then the L1 and L2 misses are the same. But if I don't, then there is much higher number of misses. Which one is the correct way to measure L1 and L2 misses?
Thanks,
Dave
----- Original Message ----
From: Dave Z. <zhu_dave@xxxxxxxxx>
To: gems <gems-users@xxxxxxxxxxx>
Sent: Wednesday, January 23, 2008 9:12:06 AM
Subject: [Gems-users] profiler
Hello,
Is it possible to keep track of L1 and L2 misses separately in
protocols where the L1 and L2 cache controllers are integrated? Could you
please show me some pointers?
Thanks,
Dave
____________________________________________________________________________________
Looking for last minute shopping deals?
Find them fast with Yahoo! Search.
http://tools.search.yahoo.com/newsearch/category.php?category=shopping
_______________________________________________
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. http://tools.search.yahoo.com/newsearch/category.php?category=shopping
|