Please refer to this link:
http://www.digipedia.pl/usenet/thread/11777/15880/

Seems GEM5-dev introduced some problems in running blackscholes. I have not 
tested the newest GEM5-Dev yet.



I tried on gem5-stable and gem5-dev, and compare the results:



Ran parsec on GEM5 Full System simulation according to the following link:

http://www.cs.utexas.edu/~parsec_m5/



The execution command used in Full System linux:

./blackscholes 4 /parsec/install/inputs/blackscholes/in_16.txt 
/parsec/install/inputs/blackscholes/prices_simdev.txt





GEM5-stable:

1. Normal FS (MI protocol ruby?):

build/ALPHA_FS/gem5.opt configs/example/ruby_fs.py --ruby --timing

Execution results: Correct.



2. Normal FS(classic memory?):

build/ALPHA_FS/gem5.opt configs/example/fs.py --timing

Execution results: Correct



3. ALPHA_FS_MOESI_CMP_token:

scons -j4 build/ALPHA_FS_MOESI_CMP_token/gem5.opt RUBY=true

build/ALPHA_FS_MOESI_CMP_token/gem5.opt configs/example/ruby_fs.py --ruby 
--timing

Execution results: Wrong.



4. ALPHA_FS_MOESI_hammer:

scons -j4 build/ALPHA_FS_MOESI_hammer/gem5.opt RUBY=true

build/ALPHA_FS_MOESI_hammer/gem5.opt configs/example/ruby_fs.py --ruby --timing

Execution results: Wrong





GEM5-dev:

1. Normal FS timing CPU:

scons -j4 build/ALPHA_FS/gem5.opt

build/ALPHA_FS/gem5.opt configs/example/fs.py --cpu-type=timing

Execution results: Wrong



2. Normal FS atomic CPU:

build/ALPHA_FS/gem5.opt configs/example/fs.py --cpu-type=atomic

Execution results: Wrong



3. Normal FS O3 CPU:

build/ALPHA_FS/gem5.opt configs/example/fs.py --cpu-type=detailed --caches

Execution results: Wrong



4. ALPHA_FS_MOESI_hammer:

scons -j4 build/ALPHA_FS_MOESI_hammer/gem5.opt RUBY=true

build/ALPHA_FS_MOESI_hammer/gem5.opt configs/example/ruby_fs.py --ruby 
--cpu-type=timing

Execution results: Wrong


From: [email protected] [mailto:[email protected]] On 
Behalf Of Steve Reinhardt
Sent: Wednesday, September 12, 2012 10:59 PM
To: gem5 users mailing list
Subject: Re: [gem5-users] GEM5 outputs are not right for PARSEC blackscholes

Are you saying that the classic memory model gives the right results, but Ruby 
does not?  Which Ruby protocol are you using?  It could be that the Alpha LL/SC 
synchronization operations are not implemented properly in some of the Ruby 
protocols.

Steve
On Tue, Sep 11, 2012 at 8:00 PM, GE ZHIGUO 
<[email protected]<mailto:[email protected]>> wrote:
Hi, I have the same problem.
You can run the benchmark on desktop to get the correct results. Or you run the 
benchmark using class memory model
In stead of ruby model.

Regards,
Zhiguo

From: [email protected]<mailto:[email protected]> 
[mailto:[email protected]<mailto:[email protected]>] On 
Behalf Of Mostafa Mahmoud Hassan
Sent: Monday, September 10, 2012 9:22 AM
To: [email protected]<mailto:[email protected]>
Subject: [gem5-users] GEM5 outputs are not right for PARSEC blackscholes

Dear All,

I am simulating PARSEC on ALPHA in FS mode followed the instructions available 
here
http://www.cs.utexas.edu/~parsec_m5/<https://mail.eng.cu.edu.eg/owa/redir.aspx?C=838e06d80af04332891850b3e24d1c35&URL=http%3a%2f%2fwww.cs.utexas.edu%2f%257Eparsec_m5%2f>

but I am facing the same problem wdescribed previously on this thread
http://www.digipedia.pl/usenet/thread/11777/15880/<https://mail.eng.cu.edu.eg/owa/redir.aspx?C=838e06d80af04332891850b3e24d1c35&URL=http%3a%2f%2fwww.digipedia.pl%2fusenet%2fthread%2f11777%2f15880%2f>

that the output does not match the correct output (anyway, how could you know 
the correct output ??)

I have tried "EXAMPLE" coherence protocol and MESI_CMP_directory


Your quick help is appreciated

Sincerely,
Mostafa

_______________________________________________
gem5-users mailing list
[email protected]<mailto:[email protected]>
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to