Quoting Nathan Binkert <[EMAIL PROTECTED]>: > > For example, for bzip2 the unified L2 miss rate error is only 0.2%, but > > for the same benchmark the L1 dcache miss rate error is almost 50%. gap > > has a dcache miss rate error of only 3%, while its L2 error is 25%. And > > this story continues for many of the other benchmarks > > Which miss rate are you looking at? overall_misses or > overall_mshr_misses? The former is all misses in the cache, but the > latter is only unique misses (i.e. if two misses end up sharing an mshr, > only one is counted.)
I've looked at both, but the numbers I mentioned are based on overall_miss_rate. I assume that this is closest to SimpleScalar's miss_rate? If I look at the errors compared to overall_mshr_miss_rate, for some benchmarks/caches the error is small, for others it is very big. So, same story unfortunately. Jos ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_idv37&alloc_id865&op=click _______________________________________________ m5sim-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/m5sim-users
