[gem5-users] Hanged on two poitns 1. Queens Beenchmark used in tutorial doesn't show out put and 2. what is differnce between two flags

2012-10-09 Thread mir shan
Hi communityfirst I am sorry to Bother you. May be question are simple but I couldn't find, I am spending my whole Time from last two weeks to understand GEM5. first Thanks of Andreas for previous solution.  1. Issue  here is my terminal snap. talpur@talpur-OptiPlex-380:~/GEM5$ ./build/ARM/gem5

[gem5-users] Running bbench on gem5

2012-10-09 Thread Amit Singh
hi... I am trying to run bbench on gem5 simulating ARM. But following error comes. Tell me, now what to do... command line: build/ARM/m5.fast --outdir=bbench2 configs/example/fs.py -b bbench-gb --kernel=vmlinux.smp.mouse.arm --frame-capture --caches --cpu-type=arm_detailed --l2cache --cacheline_

Re: [gem5-users] Running bbench on gem5

2012-10-09 Thread Anthony Gutierrez
A number of users have observed errors with the VNC code. You can try disabling the VNC stuff in gem5 or you can try to debug that error and find out why it's happening. -Tony On Oct 9, 2012 7:25 AM, "Amit Singh" wrote: > hi... > I am trying to run bbench on gem5 simulating ARM. But following e

Re: [gem5-users] Assertion failures with java and NPB benchmarks w/ ARM ISA

2012-10-09 Thread Manu Awasthi
> Hi Manu, > A gdb backtrace. >   > Thanks, > Ali Ali, Attached is the gdb backtrace for DaCaPo/lusearch #0 0x003630e30285 in raise () from /lib64/libc.so.6 #1 0x003630e31d30 in abort () from /lib64/libc.so.6 #2 0x003630e29706 in __assert_fail () from /lib64/libc.so.6 #3 0x000

[gem5-users] 2nd Issue is resolved

2012-10-09 Thread mir shan
Dear Community members my 2 issue is resolved from below link. http://docs.python.org/library/optparse.html still waiting from community response fro 1st and 3rd issue. regards Mir --- On Tue, 9/10/12, mir shan wrote: From: mir shan Subject: Hanged on two poitns 1. Queens Beenchmark used in t

Re: [gem5-users] 2nd Issue is resolved

2012-10-09 Thread Tao Zhang
For issue 1, how long does your Queen program run (in second, walltime). It’s most likely the program is running correctly. (remember the simulation is very slow) For issue 3, excerpt the following sentence from website “The stable repository is updated about once every 6 months, while the d

Re: [gem5-users] replacing simple memory with DRAMSim2 (patch released)

2012-10-09 Thread Rizwana Begum
Hi Tao, I am trying to integrate DRAMSim2 with Gem5 for ARM. I tried applying below patch (http://www.mail-archive.com/gem5-users@gem5.org/msg05088.html), however I am running into build errors, probably because of ARM platform. Can I apply your patch directly to ARM platform? FYI, I tried with bo

Re: [gem5-users] dump stats for every N cycles

2012-10-09 Thread shervin hajiamini
Hi Sascha, I added "from m5.internal.stats import periodicStatDump as statDump" to fs.py, however while building gem5, it gives me the following error: can not import name periodicStatDump I can not find the definition of "periodStatDump" function in any of the python files. I would appreciate

Re: [gem5-users] 2nd Issue is resolved

2012-10-09 Thread mir shan
Thanks Tao.In first issue It doesn't show any output like showing in tutiral video , it was in same mode as I menstioned in my email upto 2 hours. should I chage its time as you mention ((in second, walltime)) but where i find this method I mean in which file. regardsMir --- On Tue, 9/10/12,

Re: [gem5-users] Assertion failures with java and NPB benchmarks w/ ARM ISA

2012-10-09 Thread Andreas Hansson
Hi Manu, Would it be possible to enable the CachePort, Cache and Bus debug flags for the last 10 ticks? Thanks, Andreas > > >On 10/9/12 8:47 AM, "Manu Awasthi" wrote: > >>> Hi Manu, >>> A gdb backtrace. >>> >>> Thanks, >>> Ali >> >>Ali, >>Attached is the gdb backtrace for DaCaPo/lusearch >

[gem5-users] Running Multocore Scheduler on Gem5?

2012-10-09 Thread Victor Ling
Hi, I am doing the research on shared cache contention in multicore via scheduling. Does any have experience running a open source multicore scheduler on Gem5? Can you share your suggestion? I mean in terms of level of difficult. Use a real system would be easier but I don't have one handy. T

Re: [gem5-users] dump stats for every N cycles

2012-10-09 Thread Sascha Bischoff
Hi Shervin, Try updating your gem5. The patch which enables this only recently got committed to the repository (about 2 weeks ago). The definition is in src/sim/stat_control.cc, and the swig interface is in src/python/swig/stats.i. Sascha From: gem5-users-boun...@gem5.org [mailto:gem5-users-b

Re: [gem5-users] Network Message and Memory Message

2012-10-09 Thread tejasi pimpalkhute
Thanks, Nilay. I can know the ranks and banks from the physical address (datatype physical_address_t) which I get from MemoryMsg or RubyRequest but my main issue is how to extract that information from a single flit. As per the current code, I can only get a msg_ptr from it of type Message*. I am a

[gem5-users] How do you access data field of a garnet flit/packet?

2012-10-09 Thread Dean Michael Ancajas
Hi All, How do you access the data portion of a flit in garnet (if it's not a control flit)? The gem5 docu says that the data can only be accessed thru the Packet class. However, the NetworkInterface only deals with NetworkMessage class (which I think cannot be safe-casted to Packet). I am a bit

Re: [gem5-users] Running Multocore Scheduler on Gem5?

2012-10-09 Thread Steve Reinhardt
What kind of scheduler are you looking to support? One very direct approach is just to run in full system mode, and then you're running the Linux kernel scheduler. This is probably the world's most popular open source multicore scheduler ;-). Steve On Tue, Oct 9, 2012 at 10:16 AM, Victor Ling w

Re: [gem5-users] Running Multocore Scheduler on Gem5?

2012-10-09 Thread Victor Ling
I am looking for a scheduler that can reduce multicore shared cache contention (last level cache contention)  then reduce the thread missrate and improve throughput. Hope that scheduler can run on Gem5 so it is easier to change configurations. Thanks. Victor

Re: [gem5-users] How do you access data field of a garnet flit/packet?

2012-10-09 Thread Nilay
On Tue, October 9, 2012 7:27 pm, Dean Michael Ancajas wrote: > Hi All, >How do you access the data portion of a flit in garnet (if it's not a > control flit)? The gem5 docu says that the data can only be accessed thru > the Packet class. However, the NetworkInterface only deals with > NetworkMe

Re: [gem5-users] Network Message and Memory Message

2012-10-09 Thread Nilay
On Tue, October 9, 2012 6:16 pm, tejasi pimpalkhute wrote: > Thanks, Nilay. I can know the ranks and banks from the physical address > (datatype physical_address_t) which I get from MemoryMsg or RubyRequest > but > my main issue is how to extract that information from a single flit. As > per > the

[gem5-users] Sending information between CPUs

2012-10-09 Thread David Gloe
Hello, I would like to be able to send information (the PC and an integer value) between CPUs in a gem5 simulation. Instead of using interprocess communication, I would like to simulate the existence of an instantaneous or nearly instantaneous buffer which can be written to and read from with

Re: [gem5-users] Sending information between CPUs

2012-10-09 Thread Amin Farmahini
David, There are functions in gem5 to copy register file from a processor to another. Those functions might help you with what you are tying to do. I don't remember function names off the top of my head though. Thanks, Amin On Tue, Oct 9, 2012 at 10:14 PM, David Gloe wrote: > Hello, > > I woul