Thanks Nate for reviewing all of our patches. To answer your question about the ruby tester. While both the ruby tester and memtester use false sharing to test coherence, they behave in different ways. To my understanding, the Memtester has all cpus write a cache block before checking it. Meanwhile the Ruby tester issues only four writes to a cache block, before checking it with a read. I believe both testers have value. The Memtester uses relatively large global checks, while the ruby tester uses smaller more random checks.
I was initially testing the coherence protocols only using the memtester, but I noticed that a few later obvious bugs where not being encountered. That is why I added the ruby tester. I don't think the ruby tester functionality can easily be merge with the memtester. It might be easier to merge the memtester with the ruby tester by configuring the number of checks. Brad -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of nathan binkert Sent: Monday, December 21, 2009 8:49 AM To: M5 Developer List Subject: Re: [m5-dev] More GEM5 Changes > I have more patches that I would like everyone to review. I haven't yet > received any feedback on my last set of patches, so I'm going to hold off on > pushing them before Christmas. For those of you who haven't had a chance > yet to review the previous set, you can review both sets at your leisure > over the next couple of weeks. I, like I'm sure many others, will be on > vacation thru 1/3. I'm actually reviewing your patches right now. > The other major change in this new set of patches is I've resurrected the > old Ruby tester and made it a cpu MemObject, similar to the memtester. The > Ruby tester still needs to test atomics, dma accesses, and work with M5 data > rather than the ruby SubBlock object. I'll make sure to this before I check > it in. What is the difference between the ruby tester and the memtester? Can we merge their functionality? _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
