Date: | Thu, 28 Aug 2008 07:36:31 -0500 |
---|---|
From: | "Dan Gibson" <degibson@xxxxxxxx> |
Subject: | Re: [Gems-users] about the simulation result |
Your target has two processors. One of them (#2) is probably in the OS idle loop -- therefore, it is hitting in the cache on almost every memory access, and it is able to execute the idle loop a lot faster than the other program, which is taking more cache misses. This also accounts for why processor #2 has executed so many more instructions than #1.
On Thu, Aug 28, 2008 at 1:31 AM, llccdd1985 <llccdd1985@xxxxxxx> wrote:
-- http://www.cs.wisc.edu/~gibson [esc]:wq! |
[← Prev in Thread] | Current Thread | [Next in Thread→] |
---|---|---|
|
Previous by Date: | [Gems-users] Regarding Supervier Mode bit in Simics 3 and gems 2.1, sjafri |
---|---|
Next by Date: | [Gems-users] Redundant states in MOESI_SMP_directory cache implementation, Morten Sleth Rasmussen |
Previous by Thread: | Re: [Gems-users] about the simulation result, Xu Han |
Next by Thread: | [Gems-users] Are direct-mapped caches disabled for a reason?, Philip Garcia |
Indexes: | [Date] [Thread] |