Hi.. On Fri, Oct 22, 2010 at 04:50, Andev <debiandev@xxxxxxxxx> wrote: > On Thu, Oct 21, 2010 at 4:23 PM, Sitsofe Wheeler <sitsofe@xxxxxxxxx> wrote: >> Are you searching for an improvement in determinism? > > Yes, if that was so, the slowdown is still 100x. > > In my test case, I removed all the entries from /proc/mtrr and also > disabled L1/L2/L3 caches. > When I run a benchmark which runs for 4 sec with cache, it takes > almost 3000 secs without cache. > > I think I am messing up the /proc/mtrr file. IMHO it is the "in cache" speed which is folded unlinearly. Due to L1/L2/L3 nature, it might be that the whole code is already in the same cache line thus subsequent call doesn't fetch anything from RAM. This is assuming everything is (almost) perfectly cache aligned. BTW, are all the codes executed? or are we hitting 10/90 percent scenario here? -- regards, Mulyadi Santosa Freelance Linux trainer and consultant blog: the-hydra.blogspot.com training: mulyaditraining.blogspot.com -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
