On Wed, 6 Jun 2012, GE ZHIGUO wrote:
Hi, Nilay How do you do I noticed that you tried to enable O3 CPU for Ruby if I am not wrong. May I know whether you have tested Ruby for Timing CPU before? I found that execution results for blackscholes in PARSEC on GEM5/Timing CPU/ALPHA is not correct. Can you help give any advices on this?
Please avoid sending personal emails on issues related to gem5. Send all the emails to the mailing list. Most of those who are actively involved in development, read all the emails.
In your previous email, you had stated that the result was incorrect irrespective of whether or not Ruby was in use. This may mean that the problem is with the cpu portion and not the memory portion. I would suggest that you binary search through the versions between the one that works and the one that does not to figure out the first version that does not work. If we know this version number, it is highly likely that we would be able to detect the error by just reading the patch itself. Or else we would have to collect the instruction traces from two versions and see where they diverge. But the first step is to figure out the earliest
version that does not work correctly. -- Nilay _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
