Hi, thanks for your answer According to you ,should I use sim_second for computer performance?
I need to compute performance by these equations: performance=1/execution time execution time=cpu clock cycles/clock rate cpu clock cycles =instruction count *CPI Can I use sim_second instead of execution time on these equations ? or How can I compute CPI and instruction or cpu clock cycles ,If I don't value of execution time? > > > > > > > ________________________________ > From: gabriel.busnot--- via gem5-users <gem5-users@gem5.org> > Sent: Wednesday, May 5, 2021 7:42:32 PM > To: gem5-users@gem5.org <gem5-users@gem5.org> > Cc: gabriel.bus...@arteris.com <gabriel.bus...@arteris.com> > Subject: [gem5-users] Re: Interface Message Buffers between the Network > and the Ruby protocol > > Hi, > > This is based on the virtual network ID specified at the top of the > machine declaration with the MessageBuffer* declaration. > E.g.: MessageBuffer * L1RequestToL2Cache, network="From", > virtual_network="0", vnet_type="request"; > > Gabriel > _______________________________________________ > gem5-users mailing list -- gem5-users@gem5.org > To unsubscribe send an email to gem5-users-le...@gem5.org > %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s > -------------- next part -------------- > A message part incompatible with plain text digests has been removed ... > Name: not available > Type: text/html > Size: 2649 bytes > Desc: not available > > ------------------------------ > > Message: 3 > Date: Thu, 6 May 2021 12:02:45 +0200 > From: Arthur Perais <arthur.per...@univ-grenoble-alpes.fr> > Subject: [gem5-users] Re: statistics on gem5 > To: gem5-users@gem5.org > Message-ID: > <31cf9342-4281-ab3f-e1f0-cb7e38402...@univ-grenoble-alpes.fr> > Content-Type: multipart/alternative; > boundary="------------C22CA727C300F6514C0C7383" > > > On 5/6/21 10:47 AM, FARIDEH ZIAEE via gem5-users wrote: > > Hi, > > I have the following questions about simulation statistics: > > > > final_tick 9724380737064 > > host_inst_rate 79246 > > host_mem_usage 1185244 > > host_op_rate 87406 > > host_seconds 68211.69 > > host_tick_rate 129205119 > > sim_freq 1000000000000 > > sim_insts 5405536979 > > sim_ops 5962076897 > > sim_seconds 8.813299 > > sim_ticks 8813298942669 > > > > What is the difference between host_seconds and sim_second? > host_seconds is real world time spent simulating on your workstation, > sim_second is how much time was simulated : it took 68211 seconds to > simulate 8.8 seconds > > Which one is used to run time(execution time) between host_second and > > sim_second? > Not sure I understand but any performance metric or the simulated system > will use sim_second. host_second is only helpful to gauge the speed at > which gem5 simulates (KIPS/MIPS) > > What is the difference between final_tick and sim_ticks? > > sim_ticks is how many ticks were simulated (so it should be equal to > sim_seconds * ticks_per_second). final_tick is the value of curTick() > when simulation exited. If you start from tick 0, then final_tick and > sim_ticks should be the same (I think), but if you restore from a > checkpoint they will differ for sure. > > > > What do these parameters(host_inst_rate , host_op_rate , > > host_tick_rate , sim_freq , sim_insts ) mean? > > > host_inst_rate is how many instructions gem5 simulates per second > (79KIPS). host_op_rate is the same but for uops (one instruction can be > expanded to multiple uops). host_tick_rate is the same but for ticks > (essentially, one unit of simulated time takes how many real world units > of time to be processed). > > sim_freq is clockspeed you are running at > > sim_insts is how many instructions were simulated (pretty self > explanatory imho). > > > Arthur > > > > > > > > > _______________________________________________ > > gem5-users mailing list -- gem5-users@gem5.org > > To unsubscribe send an email to gem5-users-le...@gem5.org > > %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s > -------------- next part -------------- > A message part incompatible with plain text digests has been removed ... > Name: not available > Type: text/html > Size: 5236 bytes > Desc: not available > > ------------------------------ > > Subject: Digest Footer > > _______________________________________________ > gem5-users mailing list -- gem5-users@gem5.org > To unsubscribe send an email to gem5-users-le...@gem5.org > %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s > > ------------------------------ > > End of gem5-users Digest, Vol 178, Issue 10 > ******************************************* >
_______________________________________________ gem5-users mailing list -- gem5-users@gem5.org To unsubscribe send an email to gem5-users-le...@gem5.org %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s