And you should use PROTOCOL=MESI_CMP_filter_directory.
From:
gems-users-bounces@xxxxxxxxxxx [mailto:gems-users-bounces@xxxxxxxxxxx] On
Behalf Of Dan Gibson
Sent: Monday, July 20, 2009 6:38 AM
To: Gems Users
Subject: Re: [Gems-users] Ruby statistics issue
You need to run Simics (e.g.,
with cpu0.step-break and c) after loading and init-ing Ruby (see below
inlined).
Regards,
Dan
On Sun, Jul 19, 2009 at 1:52 PM, <holey003@xxxxxxx> wrote:
Hi,
I am running 'conswtch' transactional benchmark provided in GEMS. Simics is
configured as 4 core CMP. I have compiled ruby using the instruction given
on this webpage
http://www.cs.wisc.edu/gems/doc/gems-wiki/moin.cgi/QuickStart
i.e. using
command "make PROTOCOL=MOSI_SMP_bcast DESTINATION=MOSI_SMP_bcast".
But at
the end of simulation, all the ruby statistics are zero. Looks like
simulation run is disconnected from ruby. I have used following script for
the run.
##### Simulation script starts ######
magic-break-enable
read-configuration
"/scratch/tianwu/para_taint/htm/simics_space/simics-3.0.30/bin/anup_space/fourCPUtest_new.chk"
#con0.input "psrset -c 0 \n"
con0.input "psrset -c 1 \n"
con0.input "psrset -c 2 \n"
con0.input "psrset -c 3 \n"
con0.input "/host/benchmarks/conswtch/conswtch 3 3 \n"
c
cpu0.step-break 200000000
c
istc-disable
dstc-disable
stc-status
conf.sim.cpu_switch_time = 1
magic-break-enable
cpu0.step-break 100000
instruction-fetch-mode instruction-fetch-trace
instruction-fetch-mode
c
load-module ruby
ruby0.setparam g_PROCS_PER_CHIP 1
ruby0.setparam g_NUM_PROCESSORS 4
ruby0.init
cpu0.step-break 10000000
# Don't forget a 'c' here!
c
ruby0.dump-short-stats ./stats_short.txt
ruby0.dump-stats ./stats.txt
###### Script ends #######
Following is a compact data from stats I got. Everything is zero.
Ruby_current_time: 1
Ruby_start_time: 1
Ruby_cycles: 0
Total_misses: 0
total_misses: 0 [ 0 0 0 0 ]
user_misses: 0 [ 0 0 0 0 ]
supervisor_misses: 0 [ 0 0 0 0 ]
instruction_executed: 4 [ 1 1 1 1 ]
simics_cycles_executed: 4 [ 1 1 1 1 ]
cycles_per_instruction: 0 [ 0 0 0 0 ]
misses_per_thousand_instructions: 0 [ 0 0 0 0 ]
transactions_started: 0 [ 0 0 0 0 ]
transactions_ended: 0 [ 0 0 0 0 ]
instructions_per_transaction: 0 [ 0 0 0 0 ]
cycles_per_transaction: 0 [ 0 0 0 0 ]
misses_per_transaction: 0 [ 0 0 0 0 ]
What could go wrong in such case?
Thanks and regards,
Anup
_______________________________________________
Gems-users mailing list
Gems-users@xxxxxxxxxxx
https://lists.cs.wisc.edu/mailman/listinfo/gems-users
Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/"
to your search.
--
http://www.cs.wisc.edu/~gibson
[esc]:wq!