[parsec-users] bodytrack and x264 inputs incomplete for running

Christian Bienia cbienia at CS.Princeton.EDU
Mon Jan 12 13:55:49 EST 2009


Hi Eddy,

 

Here's how you can fix it:

 

1.)    Go to the PARSEC Wiki and download the missing simdev input for x264:
http://wiki.cs.princeton.edu/index.php/PARSEC#Patches_and_Bugfixes

2.)    Edit the simlarge.runconf file in the parsec/ directory of x264 and
rename the variable run_args_simlarge to run_args.

3.)    Do the same with the simlarge.runconf file of the bodytrack program.

 

After these 3 steps all workloads should work without any problems.

 

- Chris

 

 

From: parsec-users-bounces at lists.cs.princeton.edu
[mailto:parsec-users-bounces at lists.cs.princeton.edu] On Behalf Of Eddy Zheng
Zhang
Sent: Monday, January 12, 2009 1:44 PM
To: parsec-users at lists.cs.princeton.edu
Subject: [parsec-users] bodytrack and x264 inputs incomplete for running

 

Hi,

I found that when using parsecmgmt to run bodytrack and x264 program, some
input options are missing. For bodytrack, "simlarge" is missing. For x264,
"simdev" and "simlarge" are missing.

When I type: parsecmgmt -a run -c gcc-hooks -p bodytrack -i simlarge, the
output is as follows:
................................
[PARSEC] Running 'time
/home/scratch/eddy/SynergisticCache/parsec-1.0/pkgs/apps/bodytrack/inst/x86_
64-linux.gcc-hooks/bin/bodytrack ':
[PARSEC] [---------- Beginning of output ----------]
PARSEC Benchmark Suite Version 1.0
[HOOKS] PARSEC Hooks Version 1.0
Error : Invalid number of arguments
Usage : Track (Dataset Path) (# of cameras) (# of frames to process)
              (# of particles) (# of annealing layers)
              [thread model] [# of threads] [write .bmp output (nonzero =
yes)]

        Thread model : 0 = Auto, 1 = Single thread, 2 = OpenMP, 3 = Posix
Threads.

real    0m0.021s
user    0m0.004s
sys     0m0.000s
[PARSEC] [----------    End of output    ----------]
[PARSEC] Done.

I checked the bodytrack/parsec/simlarge.runconf, it writes
"run_args_simlarge="sequenceB_4 4 4 4000 5 0 ${NTHREADS}", I think it is not
consistent with other benchmarks, which is usually "run_args="something
blahblah"". I changed run_args_simlarge to run_args, it works fine. Maybe
consider fix this for the next release.


Eddy

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cs.princeton.edu/pipermail/parsec-users/attachments/20090112/71c86410/attachment.html>


More information about the parsec-users mailing list