[Gems-users] MSI_MOSI direct-mapped cache

[David Bonavila <david.bonavila@xxxxxxxxxxxxxxxxxx>]

Hi.

I am having problems configuring a direct-mapped cache with MSI_MOSI_CMP_directory protocol.

When I set L2_CACHE_ASSOC = 1, the simulation stops when only about 7,000 instructions have been executed, without any error or warning.

If I run the same script but with L2_CACHE_ASSOC = 2, it executes 200M instructions without any problem.

I have also tried the same script with MOESI_CMP_directory protocol and it works fine, so it seems to be a problem with MSI_MOSI_CMP_directory protocol and direct-mapped caches.

Has anyone been able to model a direct-mapped cache using MSI_MOSI_CMP_directory?

Here is the script which is *not* working properly:

read-configuration final-2p.check



magic-break-enable

@conf.con0.input = "cd /parsec/\n";



@conf.con0.input = "./bin/parsecmgmt -a run -i simsmall -x gcc-hooks -p bodytrack -n 2\n";




c

cpu-switch-time 1



instruction-fetch-mode instruction-fetch-trace
istc-disable



dstc-disable

load-module ruby



ruby0.setparam g_PROCS_PER_CHIP 2
ruby0.setparam g_NUM_PROCESSORS 2



ruby0.setparam NUMBER_OF_VIRTUAL_NETWORKS 5

ruby0.setparam_str g_REPLACEMENT_POLICY LRU



ruby0.setparam L1_CACHE_ASSOC 1
ruby0.setparam L1_CACHE_NUM_SETS_BITS 8



ruby0.setparam L2_CACHE_ASSOC 1
ruby0.setparam L2_CACHE_NUM_SETS_BITS 12



ruby0.init

load-module opal



opal0.init
opal0.sim-start "output_OBJ1-1_BODYTRACK_256KB_ASSOC1.opal"



opal0.sim-step 100000
opal0.stats



ruby0.dump-stats output_OBJ1-1_BODYTRACK_256KB_ASSOC1.ruby
q


Thank you. Best regards!
David