I have been looking into using M5 to run some full system disk IO tests by expanding the disk model, (I believe it currently using a flat timing model). M5 initially looks like a very good option since it was developed with IO in mind (I read "The M5 Simulator: Modeling Networked Systems") and it is open source. I have spent a fair amount of time looking though the documentation and source code (for the full system alpha) and it seems to me that the disk model could be made more accurate by replacing the flat access time with a realistic access time. My question is, do you think modeling the access time more accurately will allow M5 to properly model disks or are there other caveats in different parts of the system, like the IDE controller and PCI system, that will mess up my plans in ways I cannot foresee? Essentially, I am afraid there could be a timing issue that I didn't pick up on that would mess up my plan so I wanted to know that the experts thought.
Ben Parsons _______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
