Hello, I installed gem5 on a fresh installation of CentOS 6.3 64bit using the following commands: sudo yum install mercurial scons swig gcc m4 python python-devel gcc-c++ zlib-devel hg clone http://repo.gem5.org/gem5 cd gem5/ scons build/ARM/gem5.opt -j2 scons build/X86/gem5.opt -j2 build/X86/gem5.opt configs/example/se.py -c tests/test-progs/hello/bin/x86/linux/hello build/ARM/gem5.opt configs/example/se.py -c tests/test-progs/hello/bin/arm/linux/hello I was able to successfully run the "hello world" test, but the default network test described on http://gem5.org/Interconnection_Network gives the following error (for both the ARM and X86): [bpayne(a)virtualcentos gem5]$ build/ARM/gem5.opt configs/example/ruby_random_test.py --num-cpus=16 --num-dirs=16 --topology=Mesh --mesh-rows=4 warning: No header file specified for SimObject: Switch gem5 Simulator System. http://gem5.org gem5 is copyrighted software; use the --copyright option for details. gem5 compiled Nov 15 2012 12:48:12 gem5 started Nov 16 2012 13:57:27 gem5 executing on virtualcentos command line: build/ARM/gem5.opt configs/example/ruby_random_test.py --num-cpus=16 --num-dirs=16 --topology=Mesh --mesh-rows=4 Global frequency set at 1000000000 ticks per second Warning: rounding error > tolerance 0.072760 rounded to 0 Warning: rounding error > tolerance 0.072760 rounded to 0 info: Entering event queue @ 0. Starting simulation... hack: be nice to actually delete the event here Exiting @ tick 22391 because Ruby Tester completed [bpayne(a)virtualcentos gem5]$ Note: the "run_random_test.py" listed on the wiki doesn't exist in the example directory. Let me know if I should update the documentation on the wiki to reflect the current file name "ruby_random_test.py" Summary: I'm not sure what is causing this "Warning: rounding error > tolerance" error or what to do about it. Other people have seen it but no resolution was listed. I see that the tolerance is set in gem5/build/<ARCH>/python/m5/ticks.py Do I just make the tolerance larger? Thank you, Ben Payne Laboratory for Physical Sciences http://www.lps.umd.edu/