It turns out it's just a minor configuration problem... it's unexpected to specify --l2cache without also giving the --caches flag (since what you end up with is actually a shared L1 cache), so the script missed that. I just pushed a patch that makes it work:
http://repo.m5sim.org/m5/rev/8c68656b8564 Steve On Mon, Sep 14, 2009 at 3:44 PM, Felix Loh <[email protected]> wrote: > Hello, > > I've tried running the following command: > $ build/ALPHA_FS/m5.opt -d ../orig_netperf/ configs/example/fs.py --l2cache > -b NetperfStream > > The thing is, it eventually terminates prematurely due to the "--l2cache" > flag: > > Shell output: > > M5 compiled Sep 14 2009 17:50:19 > M5 revision 94c016415053 6283 default tip > M5 started Sep 14 2009 18:00:01 > M5 executing on felixloh-desktop > command line: build/ALPHA_FS/m5.opt -d ../orig_netperf/ > configs/example/fs.py --l2cache -b NetperfStream > Global frequency set at 1000000000000 ticks per second > info: kernel located at: > /home/felixloh/research/m5/system_files/binaries/vmlinux > Listening for testsys connection on port 3456 > 0: testsys.tsunami.io.rtc: Real-time clock set to Thu Jan 1 00:00:00 > 2009 > info: kernel located at: > /home/felixloh/research/m5/system_files/binaries/vmlinux > Listening for drivesys connection on port 3457 > 0: drivesys.tsunami.io.rtc: Real-time clock set to Thu Jan 1 00:00:00 > 2009 > 0: testsys.remote_gdb.listener: listening for remote gdb #0 on port 7000 > 0: drivesys.remote_gdb.listener: listening for remote gdb #1 on port 7001 > **** REAL SIMULATION **** > info: Entering event queue @ 0. Starting simulation... > 2240256513000: testsys.cpu.break_event: break event panic triggered > panic: M5 panic instruction called at pc=0xfffffc000031add0. > @ cycle 2240353352500 > [execute:build/ALPHA_FS/arch/alpha/atomic_simple_cpu_exec.cc, line 10987] > Memory Usage: 401872 KBytes > For more information see: http://www.m5sim.org/panic/116b925c > Program aborted at cycle 2240353352500 > Aborted > > Conversely, when I left out this "--l2cache" flag in the command above, the > benchmark runs to completion. > > Do you have any suggestions on what I can do here? I've attached the > corresponding testsys.terminal output file as well. > > Thanks, > Felix > _______________________________________________ > m5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users > _______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
