Sorry I don't have an answer off the top of my head for this, but since no one else is speaking up, I'd suggest wading through the code in configs/common/Simulation.py and see if you can figure out what's going on, and then perhaps you can fix it to do what you want.
Steve On Tue, Jul 6, 2010 at 10:31 PM, Sage <[email protected]> wrote: > Hi, Gabe, > > Thanks a lot for your answer. Now I am hoping someone can answer my second > question as follows. > > It seems that the two commands "--checkpoint-restore=XXX --at-instruction > --warmup-insts=250000000 --max-inst=250000000" and "--checkpoint-restore=XXX > --at-instruction --max-inst=250000000" produce the same simulation results, > indicating that "-warmup-insts=250000000" does not take effect at all. I am > wondering how I can warm up caches after recovering from a checkpoint??? > However, using "--fast-forward" instead of "--checkpoint-restore" seems to > work fine, but I 'd prefer to use ""--checkpoint-restore" to save time. > > > Leonard > > > > > On Tue, Jul 6, 2010 at 10:26 PM, Gabriel Michael Black > <[email protected]> wrote: >> >> I'll answer your first question. All addresses outside of the CPU should >> be physical. The TLB objects defined by the ISA do the translation, and >> they're hooked in as internal components of the CPU. We've talked a little >> bit about whether those should be pulled out into the memory system, but >> that unfortunately introduces a lot of complications. >> >> Gabe >> >> Quoting Sage <[email protected]>: >> >>> Hi, everyone, >>> >>> I have two questions regarding the m5 simulator. >>> >>> The first question is whether or not the address seen by a cache module >>> (e.g., the parameter "addr" passed to the cache function "accessBlock") >>> is a >>> virtual address. If it is, when does the virtual->physical translation >>> take >>> place? >>> >>> The second question is that it seems to me that the two commands >>> "--checkpoint-restore=XXX --at-instruction --warmup-insts=250000000 >>> --max-inst=250000000" and "--checkpoint-restore=XXX --at-instruction >>> --max-inst=250000000" produce the same simulation results, indicating >>> that >>> "-warmup-insts=250000000" does not take effect at all. I am wondering how >>> I >>> can warm up caches after recovering from a checkpoint??? However, >>> using >>> "--fast-forward" instead of "--checkpoint-restore" seems to work fine, >>> but I >>> 'd prefer to use ""--checkpoint-restore" to save time. >>> >>> Thanks in advance! >>> >>> Leonard >>> -- >>> Give our ability to our work, but our genius to our life! >>> >> >> >> _______________________________________________ >> m5-users mailing list >> [email protected] >> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users > > > > -- > Give our ability to our work, but our genius to our life! > > _______________________________________________ > 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
