Hi,
Some months ago, I asked a question about "Starting simulation..."
messages. At that time, I understand the logic, however there are some
new questions:

1- Using X86_SE and only one core, I run:

   -F 1000000 -d --caches --maxtick 10000000 --l2cache -b bzip2_chicken

In the output I see:

Global frequency set at 1000000000000 ticks per second
0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000
Switch at instruction count:1000000
info: Entering event queue @ 0.  Starting simulation...
warn: instruction 'fnstcw_Mw' unimplemented
warn: instruction 'fldcw_Mw' unimplemented
spec_init
Loading Input Data
Duplicating 652307 bytes
hack: be nice to actually delete the event here
Switched CPUS @ tick 670909000
info: Entering event queue @ 670909000.  Starting simulation...
info: Entering event queue @ 671024000.  Starting simulation...
info: Entering event queue @ 671046000.  Starting simulation...
Changing memory mode to timing
switching cpus
**** REAL SIMULATION ****
info: Entering event queue @ 671048000.  Starting simulation...
Exiting @ tick 681048000 because simulate() limit reached


I expect, "-F 1,000,000" triggers  "Starting simulation..." @1,000,000
and "--maxtick 10,000,000 " triggers another "Starting simulation..."
@1,100,000

What I see in the output is:
four "Starting simulation..." messages in the fast forward time (which
I don't understand) and one "Starting simulation..." when it passes
fatsfwd and enters real simulation (which is normal).





2- The understanding is more difficult when use more than one core:
    -F 1000000 -d --caches --maxtick 10000000 --l2cache -b
bzip2_chicken,bzip2_liberty

I have set np=2 in cmp.py. In the output I see:
Global frequency set at 1000000000000 ticks per second
0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000
0: system.remote_gdb.listener: listening for remote gdb #1 on port 7001
Switch at instruction count:1000000
info: Entering event queue @ 0.  Starting simulation...
warn: instruction 'fnstcw_Mw' unimplemented
warn: instruction 'fldcw_Mw' unimplemented
spec_init
spec_init
Loading Input Data
Loading Input Data
Duplicating 652307 bytes
Duplicating 289631 bytes
hack: be nice to actually delete the event here
Switched CPUS @ tick 996602000
info: Entering event queue @ 996602000.  Starting simulation...
info: Entering event queue @ 996672000.  Starting simulation...
info: Entering event queue @ 996704000.  Starting simulation...
Changing memory mode to timing
switching cpus
**** REAL SIMULATION ****
info: Entering event queue @ 996706000.  Starting simulation...
Exiting @ tick 1006706000 because simulate() limit reached


As you can see, number of "Starting simulation..." messages equl to
uni processor.

Can someome clarify that?
Thanks,
--
// Naderan *Mahmood;
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to