Hey,
Op 13-mrt-10, om 21:03 heeft Korey Sewell het volgende geschreven: > > Because I want to model the branch penalty of an in-order core this > will probably not affect the total number of inpredicted branches, nor > the state of the branch predictor. > I'm a little confused here. Are you are saying that you want to know > many cycles do mispredicted branches contribute to overall execution? > That's correct. And you can do that by using the InOrderCPU and looking at the produced statistics. However, I want to get the number of mispredicted branches by using the SimpleCPU model since it's much faster. In the mean time I digged into the code a little deeper. Since we can get the PC from SimpleThread (which is an object in the SimpleCPU), we can use this as an argument for the functions in src/cpu/pred. These functions will get me the statistics I need. Max _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
