On 08/20/10 21:21, Kishore Kumar Pusukuri wrote:
Hi,
I am able to measure TLB miss-rate of a multi-threaded application
running on my multi-core AMD Opteron machine by reading performance
monitoring event counters using cpustat utility. However, I would like
to measure the amount of time spent on TLB misses? Specifically, I am
looking a way like the utility trapstat functions. Please share your ideas.
TLB miss costs (e.g. page table walk) costs can be estimated by writing
code that misses in both cache and tlb, and comparing steady state
performance to that of otherwise similar code that misses only in cache.
Keep in mind that different size pages may take different amounts of
time, and that modern processors do an amazing job of predicting
access patterns - so attempts to miss in cache will prob. need some
random component.
You may find http://icl.cs.utk.edu/projectsfiles/hpcc/RandomAccess/
to be interesting reading...
- Bart
--
Bart Smaalders Solaris Kernel Performance
bart.smaald...@oracle.com http://blogs.sun.com/barts
"You will contribute more with mercurial than with thunderbird."
_______________________________________________
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
