I use SE. For the following two commands: build/X86_SE/m5.debug --trace-start=20000000 --debug-flags=HWPrefetch configs/example/cmp.py -F 200000000 --maxtick 100000000000
and build/X86_SE/m5.debug --trace-start=50000000 --debug-flags=HWPrefetch configs/example/cmp.py -F 200000000 --maxtick 100000000000 you can see that the messages start with this tick number: 54083000: system.l2-pf: Found a pf candidate addr: 0xf6480, inserting into prefetch queue with de lay 1000 time 54093000 54083000: system.l2-pf: Found a pf candidate addr: 0xf64c0, inserting into prefetch queue with de lay 1000 time 54093000 .... questions: 1- the number in '-F' shows number of instruction. Is there any relation between this and number of ticks? 2- As you can see with two different values for --trace-start, the messages start from a specific tick number. why? Thanks On 11/13/11, Dibakar Gope <[email protected]> wrote: > set --trace-start to boot completion time (u have to find out a approximate > value for that), that will allow u to begin tracing from a particular > simulation > tick > > On 11/13/11, Mahmood Naderan wrote: >> Hi, >> when I use --debug-flags=Cache, the messages are written from the >> beginning of the program (I mean while it is in fast forward mode). How >> can I tell m5.debug to print messages when it enter --detail or --timing? >> >> >> >> >> >> Thanks, >> -- >> // Naderan *Mahmood; >> >> >> >> >> _______________________________________________ >> gem5-users mailing list >> [email protected] >> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > > _______________________________________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > -- -- // Naderan *Mahmood; _______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
