Malogro <malogro <at> gmail.com> writes:
> No problem. By the way, going on a tangent here: It might have been implied in my last e-mail that the number of L2 caches HAVE to be equal to the CPUs for the simulation to run, while that is NOT the case. The simulation will run fine with l2 caches >= 2 (only one won't run for some reason) > > > > I've never tested this to the end, but I'm confident it'd work just fine. > > Alexandre Yasuo YamamotoOn Fri, Feb 3, 2012 at 3:05 PM, ankit more <am434 <at> drexel.edu> wrote: > Alexandre - that fixed it.ThanksAnkit > On Fri, Feb 3, 2012 at 4:01 PM, Malogro <malogro <at> gmail.com> wrote: > > Declare the L2 caches to be the same number as the CPUs and Directories. That should solve the problem right away. Command to do so:--num- l2caches=whatyouwant > Alexandre Yasuo Yamamoto > > > On Fri, Feb 3, 2012 at 2:24 PM, Ankit More <am434 <at> drexel.edu> wrote: > > > > Tushar Krishna <tushar <at> csail.mit.edu> writes: > > > > Hi Xing, > > How many directories did you declare? > > A few things to note: > > The Mesh topology by default works with equal cpus (= l1 caches) and > > directories, connecting them all to one router each. > > If you use only 4 directories, MeshDirCorners might be a better topology > > as it connects each cache to a router, and connects a directory at each > > corner. > > > > Another thing to note: you also need to declare --mesh-rows ... > > > > The error you see occurs because the Mesh topology assumes that after > > connecting all caches and directories, the only node left is DMA. > > > > Working examples would be : > > --num-cpus=16 --num-dirs=16 --topology=Mesh --mesh-rows=4 > > --num-cpus=16 --num-dirs=4 --topology=MeshDirCorners --mesh-rows=4 > > > > cheers, > > Tushar > > > Hello All, > I have hot upon this problem again, and as Tushar has suggested I have set the > number of CPUs and number of directories to be equal. It still gives me the same > error as Xing. > The error info. is as below: > *********************** > Error: could not create topology Mesh > Traceback (most recent call last): > File "<string>", line 1, in<module> > File "/home/xxxx/Projects/gem5/src/python/m5/main.py", line 360, in main > exec filecode in scope > File "configs/example/ruby_fs.py", line 123, in<module> > system._dma_devices) > File "/home/xxxx/Projects/gem5/configs/ruby/Ruby.py", line 104, in create_system > % options.topology) > File "<string>", line 1, in<module> > File "/home/xxxx/Projects/gem5/src/mem/ruby/network/topologies/Mesh.py", > line 80, in makeTopology > assert(node.type == 'DMA_Controller') > AssertionError > It doesnt complain when I use the default MI_example protocol. Any idea what > this could be about? > Thanks > Ankit > _______________________________________________ > gem5-users mailing listgem5-users <at> gem5.orghttp://m5sim.org/cgi- bin/mailman/listinfo/gem5-users > > > > > > > _______________________________________________ > gem5-users mailing listgem5-users <at> gem5.orghttp://m5sim.org/cgi- bin/mailman/listinfo/gem5-users > > > > > _______________________________________________ > gem5-users mailing listgem5-users <at> gem5.orghttp://m5sim.org/cgi- bin/mailman/listinfo/gem5-users > > > > > > > _______________________________________________ > gem5-users mailing list > gem5-users <at> gem5.org > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users Hello everyone, I am new to Gem5. While performing some of the simulations i get some common errors which i can resolve by looking at the mailing list. My question is that how do you guys get to know all the answers by experience or am i not reading the documentation properly? _______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
