Zoran Jaksic <zorjak <at> gmail.com> writes:
> > Hi guys, > I have one small question. If anyone could help me I would be very grateful. Please answer if you have any hint. > > I run blackscholes benchmark from the image you provided. I used script to generate checkpoint and I used simmedium input set. I used moesi default configuration for the machine. Only thing that I changed in the code is some small part that counts the number of valid lines of every cache and prints them in log file. however when I run this simulation until the end it seems that more than 80% of the cache lines are invalid all the time (in L3) and every L2. L2 are 2MB and L3 8MB but more than 1.5 billion instructions are commited, so it should be enough? Similar situations is with many other parsec benchmarks. Do you have any clue what is happening here. Is this normal, or is there any way to warm up cache before simulation? > > Thank you very much in forward > > Best regards > Zoran > > > > > <div><div dir="ltr">Hi guys,<div><br></div> > <div>I have one small question. If anyone could help me I would be very grateful. Please answer if you have any hint.</div> > <div><br></div> > <div>I run blackscholes benchmark from the image you provided. I used script to generate checkpoint and I used simmedium input set. I used moesi default configuration for the machine. Only thing that I changed in the code is some small part that counts the number of valid lines of every cache and prints them in log file. however when I run this simulation until the end it seems that more than 80% of the cache lines are invalid all the time (in L3) and every L2. L2 are 2MB and L3 8MB but more than 1.5 billion instructions are commited, so it should be enough? Similar situations is with many other parsec benchmarks. Do you have any clue what is happening here. Is this normal, or is there any way to warm up cache before simulation?</div> > <div><br></div> > <div>Thank you very much in forward<br> > </div> > <div>Best regards</div> > <div>Zoran<br clear="all"><div><br></div> > <br> > </div> > </div></div> > I think PARSEC benchmark suite, especially the blackscholes benchmark has very small working set size to properly exercise your entire cache capacity. Are you using the simlarge input set ? And how many cores are you using ? I think the working set increases with number of threads as well. _______________________________________________ http://www.marss86.org Marss86-Devel mailing list [email protected] https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
