Re: [Gems-users] Problem setting up runs with SNUCA and DNUCA config

["Mishali Naik" <mishalinaik@xxxxxxxxx>]

Only one of them have assert failure. Using the first module I am
simulating NUCA with SNUCA (I started this run first, it doesn't have
any assertion failures).  Then I simulated NUCA w/ DNUCA using second
module, this gives assertion failure.
I am using gen-scripts for my runs. I have attached the script. And
here are the ruby config params:

   run_sim_command("ruby0.setparam_str g_PRINT_TOPOLOGY true")
   run_sim_command("ruby0.setparam_str g_DYNAMIC_TIMEOUT_ENABLED false")
   run_sim_command("ruby0.setparam_str g_adaptive_routing false")
   run_sim_command("ruby0.setparam_str
REMOVE_SINGLE_CYCLE_DCACHE_FAST_PATH true")
   run_sim_command("ruby0.setparam NUMBER_OF_VIRTUAL_NETWORKS 7")
   run_sim_command("ruby0.setparam_str g_CACHE_DESIGN NUCACOL")
   run_sim_command("ruby0.setparam g_MEMORY_SIZE_BYTES 4294967296")
   run_sim_command("ruby0.setparam L1_CACHE_NUM_SETS_BITS 7")
   run_sim_command("ruby0.setparam L2_CACHE_NUM_SETS_BITS 15")
   run_sim_command("ruby0.setparam L2_CACHE_ASSOC 8")
   run_sim_command("ruby0.setparam L1_CACHE_ASSOC 4")
   run_sim_command("ruby0.setparam g_endpoint_bandwidth 1000")
   run_sim_command("ruby0.setparam g_NUM_DNUCA_BANK_SETS 4")
   run_sim_command("ruby0.setparam_str g_NUCA_PREDICTOR_CONFIG DNUCA")
   run_sim_command("ruby0.setparam_str PERFECT_DNUCA_SEARCH true")
   run_sim_command("ruby0.setparam_str
COLLECTOR_HANDLES_OFF_CHIP_REQUESTS true")






On 10/10/06, Dan Gibson <degibson@xxxxxxxx> wrote:
> Which one (or both) of the executions have the assert failures?
> Send the startup script to the list.  If the script is monstrously long,
> send the configuration of Ruby only. I'll need to see the configuration
> options from the failed execution.
>
> Regards,
> Dan
>
> Mishali Naik wrote:
> > No in different Simics process. This is what I am doing:
> >
> > One simics process : I am using module MOESI_CMP_NUCA, I set ruby
> > parameters w/ different scripts
> >
> > Another simics process: I am using module MOESI_CMP_NUCA1. I set ruby
> > parameters w/ different scripts.
> >
> > Thanks.
> >
> >
> > On 10/10/06, Dan Gibson <degibson@xxxxxxxx> wrote:
> >> Are you trying to use both modules within the same Simics process?
> >>
> >> Regards,
> >> Dan
> >>
> >> Mishali Naik wrote:
> >> > Hi,
> >> >
> >> > I compiled different modules, but the simulation stops by giving
> >> > errors for the ruby config parameters:
> >> >
> >> >
> >> > failed assertion 'L2bank < RubyConfig::numberOfL2CachePerChip()/2' at
> >> > fn MachineID map_L1CacheMachId_to_L2CacheLocalBank(const Address&,
> >> > const MachineID&) in slicc_interface/RubySlicc_ComponentMapping.h:171
> >> > failed assertion 'L2bank < RubyConfig::numberOfL2CachePerChip()/2' at
> >> > fn MachineID map_L1CacheMachId_to_L2CacheLocalBank(const Address&,
> >> > const MachineID&) in slicc_interface/RubySlicc_ComponentMapping.h:171
> >> > At this point you might want to attach a debug to the running and get
> >> > to the
> >> > crash site; otherwise press enter to continue
> >> > PID: 29724
> >> >
> >> > I got similar error when I was using just one module. It would be a
> >> > great help if someone could point me to the where the problem might
> >> > be.
> >> > thanks a lot,
> >> > Mishali
> >> >
> >> >
> >> >
> >> > On 10/9/06, Dan Gibson <degibson@xxxxxxxx> wrote:
> >> >> Compile separate modules:
> >> >> two make's with different DESTINATION=
> >> >>
> >> >> Different configuration options:
> >> >> Different scripts to run the simulations, eg:
> >> >> ruby0.setparam... etc.
> >> >>
> >> >> Mike Marty wrote:
> >> >> > With 2 separate simulation jobs, sure.
> >> >> >
> >> >> > Either compile separate modules, or use the same module but specify
> >> >> > different config options in Simics before starting simulation
> >> >> >
> >> >> >
> >> >> >> -----Original Message-----
> >> >> >> From: gems-users-bounces@xxxxxxxxxxx [mailto:gems-users-
> >> >> >> bounces@xxxxxxxxxxx] On Behalf Of Mishali Naik
> >> >> >> Sent: Monday, October 09, 2006 11:35 AM
> >> >> >> To: Gems Users
> >> >> >> Subject: Re: [Gems-users] Problem setting up runs with SNUCA and
> >> >> DNUCA
> >> >> >> config
> >> >> >>
> >> >> >> I would like to have one run with SNUCA config, and another run
> >> with
> >> >> >> DNUCA config (in parallel). Is it possible to have this set up?
> >> >> >>
> >> >> >> On 10/9/06, Mike Marty <mikem@xxxxxxxxxxx> wrote:
> >> >> >>
> >> >> >>> How don't get the question.  You want results for 2 protocols
> >> >> (SNUCA and
> >> >> >>> DNUCA) at the same time?  You need to pick one config.
> >> >> >>>
> >> >> >>> Mike
> >> >> >>>
> >> >> >>>
> >> >> >>>
> >> >> >>>> Hi All,
> >> >> >>>>          I have been trying to set up runs with SNUCA and DNUCA
> >> >> >>>> predictors simultaneously, but I haven't had any luck so far. At
> >> >> one
> >> >> >>>> time, I can have runs with SNUCA configuration only or DNUCA
> >> >> >>>> configuration only but not both. Is it because of the ruby
> >> config
> >> >> >>>>
> >> >> >> parameters?
> >> >> >>
> >> >> >>>> I am wondering if there is a work around this? I really would
> >> >> >>>> appreciate any kind of help.
> >> >> >>>>
> >> >> >>>> thanks,
> >> >> >>>> Mishali
> >> >> >>>> _______________________________________________
> >> >> >>>> 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.
> >> >> >>
> >> >> >>> _______________________________________________
> >> >> >>> 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.
> >> >> >>
> >> >> >>>
> >> >> >> _______________________________________________
> >> >> >> 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.
> >> >> >>
> >> >> >
> >> >> >
> >> >> > _______________________________________________
> >> >> > 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!
> >> >>
> >> >> _______________________________________________
> >> >> 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.
> >> >>
> >> >>
> >> >
> >>
> >

Attachment: mfacet.py
Description: Binary data