Re: [gem5-users] Adding an Ethernet interface to the ARM platform

On Sun, Jun 3, 2012 at 10:25 PM, Nilay wrote: > On Fri, June 1, 2012 11:43 am, Anirudh Sivaraman wrote: >> On Thu, May 31, 2012 at 6:31 PM, Nilay Vaish wrote: >>> On Thu, 31 May 2012, Anirudh Sivaraman wrote: >>> This is related to my earlier post, but I ll post this as a separate threa

Re: [gem5-users] Error "Can't find a path to system files" in System Full mode set up

Hello Eng.Bojun, Thank you so much for your reply. No,i haven't changed the SysPath.py. Actually i am confused in this point because i downloaded from " http://gem5.org/Download"; the full system for X86 : - Full System Files-- The kern

Re: [gem5-users] Android kernel stalls

On Sun, Jun 3, 2012 at 10:21 PM, Ali Saidi wrote: >> >> Sorry about switching between top and bottom posts, gmail defaults to top. >> >> Originally I thought I was building the 38 kernel; however, because I >> did a "git checkout -b", I really only created a new branch >> identically to the HEAD -

Re: [gem5-users] Hw_architecture_Gem5

On Apr 24, 2012, at 3:46 AM, Fathallah-XID, Houssem wrote: > Hi all, I am Gem5 beginner, I have run your bbench android example in FS > mode, and I have been able to get the statistics without any problem. > > build/ARM/gem5.opt configs/example/fs.py -b bbench > --kernel=vmlinux.smp.mouse.a

Re: [gem5-users] Adding an Ethernet interface to the ARM platform

On Fri, June 1, 2012 11:43 am, Anirudh Sivaraman wrote: > On Thu, May 31, 2012 at 6:31 PM, Nilay Vaish wrote: >> On Thu, 31 May 2012, Anirudh Sivaraman wrote: >> >>> This is related to my earlier post, but I ll post this as a separate >>> thread to state my problem clearly. I have been trying to d

Re: [gem5-users] Android kernel stalls

> > Sorry about switching between top and bottom posts, gmail defaults to top. > > Originally I thought I was building the 38 kernel; however, because I > did a "git checkout -b", I really only created a new branch > identically to the HEAD - which is the 35 kernel. So, when I removed > that bran

Re: [gem5-users] Building ALPHA with MRUPolicy implemented at L2

On Sun, 3 Jun 2012, Sathyanarayanan S wrote: I want to build ALPHA system with MRUPolicy implemented at L2. How do I do that? What all things did you try out? I think it would make better sense to write to the mailing list if you are facing some problem in running some particular type of si

[gem5-users] Building ALPHA with MRUPolicy implemented at L2

I want to build ALPHA system with MRUPolicy implemented at L2. How do I do that? -- Regards, Sathyanarayanan S ___ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] modifying Ruby memory model

Thanks, Nilay! That helps to get started. On Sun, Jun 3, 2012 at 5:47 PM, Nilay Vaish wrote: > On Sun, 3 Jun 2012, tejasi pimpalkhute wrote: > > Hi there, >> >> I wanted to change the physical memory model for DDR to some other types >> of >> memory like Phase change RAM or so. Could you please

Re: [gem5-users] modifying Ruby memory model

On Sun, 3 Jun 2012, tejasi pimpalkhute wrote: Hi there, I wanted to change the physical memory model for DDR to some other types of memory like Phase change RAM or so. Could you please let me know if it is possible to do so and what all files will I need to touch? (I saw the specifications for

[gem5-users] modifying Ruby memory model

Hi there, I wanted to change the physical memory model for DDR to some other types of memory like Phase change RAM or so. Could you please let me know if it is possible to do so and what all files will I need to touch? (I saw the specifications for DDR are written in dram(.hh/.cc), do I need to wr

[gem5-users] questions about running SPEC CPU benchmark in ARM_FS mode

Hi, I have a question about running SPEC CPU benchmark in ARM_FS mode. Where in the command line can I specify which benchmark binary I want to use. Do I need to write an scrip and use --script? I didn't find some related document for SPEC FS mode. Anyone can give some suggestion?

[gem5-users] Android benchmarks and BBench implementation

I am trying to create and port some benchmarks over to gem5 Android. I've been basing my efforts on BBench, but I have a few questions about its implementation. In init.rc, "m5 readfile" is called to read "start_apk.sh". This is actually the benchmark script (.rcS) that is passed in from Benchmark

Re: [gem5-users] Error "Can't find a path to system files" in System Full mode set up

Have you changed the Path in SysPath.py? On Sun, Jun 3, 2012 at 12:36 PM, wael Amr wrote: > Hello, > > I need to run gem5 under Full System on Ubuntu using virtual machine . > So i typed the following commands for mount the linux-x86.img: > > sudo python util/gem5img.py mount > FullSystemStuff/

Re: [gem5-users] Android kernel stalls

On Sun, Jun 3, 2012 at 3:48 PM, Anirudh Sivaraman wrote: > On Sun, Jun 3, 2012 at 3:20 PM, Ira Ray Jenkins > wrote: >> Well, that's slightly frustrating to have to use a separate toolchain >> than the one distributed with AOSP, but in any case I will attempt it >> this way. Which version do I ne

Re: [gem5-users] Android kernel stalls

On Sun, Jun 3, 2012 at 3:20 PM, Ira Ray Jenkins wrote: > Well, that's slightly frustrating to have to use a separate toolchain > than the one distributed with AOSP, but in any case I will attempt it > this way. Which version do I need to download?Codebench Arm Lite > Edition EABI or Codebench Arm

Re: [gem5-users] Android kernel stalls

What happens when you build the 35 kernel with the config on gem5.org? That output was for the 38 kernel. Also, that output isn't very useful because it prints normal output then stops. What does the simulator's terminal output say? -Tony On Sun, Jun 3, 2012 at 3:44 PM, Ira Ray Jenkins wrote: >

Re: [gem5-users] Android kernel stalls

I have built with the toolchain provided by AOSP. It produces a vmlinux successfully; however, when I run gem5 with that kernel it stalls. I've posted the command line and terminal output in the original post of this thread. I'm currently compiling with the codebench toolchain, maybe that will make

Re: [gem5-users] Android kernel stalls

You don't have to change your toolchain. It just makes it more likely that you'll be able to find support because the codesourcery toolchain is more commonly used. That being said, have you tried building with the compiler/libs provided with Android? If so, what problems are you having? -Tony On

Re: [gem5-users] Android kernel stalls

Well, that's slightly frustrating to have to use a separate toolchain than the one distributed with AOSP, but in any case I will attempt it this way. Which version do I need to download?Codebench Arm Lite Edition EABI or Codebench Arm Lite Edition GNU/Linux? I would eventually like to be able to b

[gem5-users] demand miss hit in MSHR

Hi is there any statistic which show how many demand misses in cache already hit in MSHR? -- // Naderan *Mahmood; ___ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] ALPHA_FS problem -Could not load kernel file-

Hi; You must modify SysPath.py which exists in $GEM5/config/common as following: You can find following fields in SysPath.py: except KeyError: path = [ '/dist/m5/system' ,'.' ] '' must be set by path to disk image. For example if disk image, which downloaded from Gem5 website, ex

[gem5-users] Error "Can't find a path to system files" in System Full mode set up

Hello, I need to run gem5 under Full System on Ubuntu using virtual machine . So i typed the following commands for mount the linux-x86.img: sudo python util/gem5img.py mount FullSystemStuff/x86-system/disks/linux-x86.img /mnt sudo umount /mnt Then i typed :command line: ./build/x86/gem5.opt co

Re: [gem5-users] ~/python2.6/pyconfig.h:1013:5: error: "__BIG_ENDIAN__" is not defined scons: *** [build/ALPHA/sim/main.do] Error 1

Hello, Have you installed python 2.7 ? If no,try to install it. Thanks Best Regards Wael Amr On Sun, Jun 3, 2012 at 6:01 PM, vasu wrote: > I am compiling gem5 on mac OS X 10.6 > $sudo scons build/ALPHA/m5.debug > > and i am getting this output with error > > **console output

[gem5-users] ~/python2.6/pyconfig.h:1013:5: error: "__BIG_ENDIAN__" is not defined scons: *** [build/ALPHA/sim/main.do] Error 1

I am compiling gem5 on mac OS X 10.6 $sudo scons build/ALPHA/m5.debug and i am getting this output with error **console output scons: Reading SConscript files ... Checking for leading underscore in global variables...(cached) yes Checking for C header file Python.h...

[gem5-users] ALPHA_FS problem -Could not load kernel file-

Hi All I have this error. How to solve this BR Abbas abbas@fake:~/gem5-stable/configs/common> build/ALPHA_FS/gem5.opt -d /tmp/output configs/example/fs.py -bash: build/ALPHA_FS/gem5.opt: No such file or directory abbas@fake:~/gem5-stable/configs/common> cd .. abbas@fake:~/gem5-stable/configs>

[gem5-users] GEM5 System software architecture

Hi All I need complete software architecture of the GEM5. but  I can't find  any system detail architecture. Anybody has information about it ? Best Regards Abbas ___ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinf

Re: [gem5-users] Ruby + OoO + with check-pointing support

On Sun, 3 Jun 2012, Volos Stavros wrote: But when running a ruby simulation, the cpu used is TimingSimpleCPU right? So, actually I am creating a checkpoint with that mode. Even in that case, I cannot restore the checkpoint with O3. I am definite that you created a checkpoint using an Atomic

Re: [gem5-users] Ruby + OoO + with check-pointing support

But when running a ruby simulation, the cpu used is TimingSimpleCPU right? So, actually I am creating a checkpoint with that mode. Even in that case, I cannot restore the checkpoint with O3. So it seems that the only way to restore a Timing/O3 cpu is by having a checkpoint that was created us

Re: [gem5-users] TLB statistics in ARM SE mode

If you want to account for these (bearing in mind that even a miss won't delay execution) you can add the stat increments to translateSe.cc in arch/arm/tlb.cc Ali Sent from my ARM powered mobile device On Jun 3, 2012, at 5:18 AM, Nathanaël Prémillieu wrote: > Then why the stats of the TLBs a

Re: [gem5-users] Ruby + OoO + with check-pointing support

On Sun, 3 Jun 2012, Volos Stavros wrote: Hi Nilay, I am aware of this feature. I do restore the checkpoint using --restore-with-cpu. When setting that parameter to timing, it works. However, the simulation will not be detailed. I want to run an OoO+ruby simulation as soon as the checkpoint

Re: [gem5-users] Ruby + OoO + with check-pointing support

Hi Nilay, I am aware of this feature. I do restore the checkpoint using --restore-with-cpu. When setting that parameter to timing, it works. However, the simulation will not be detailed. I want to run an OoO+ruby simulation as soon as the checkpoint is loaded. I tried several combinations: a)

Re: [gem5-users] Android kernel stalls

On Sat, Jun 2, 2012 at 8:40 PM, Ira Ray Jenkins wrote: > On Sat, Jun 2, 2012 at 7:54 PM, Anirudh Sivaraman > wrote: >> This thread may be useful : >> http://m5sim.org/cgi-bin/mailman/private/gem5-users/2012-June/019665.html >> since I had a similar problem today morning. I used 2.6.35 and after

Re: [gem5-users] Ruby + OoO + with check-pointing support

On Sun, 3 Jun 2012, Volos Stavros wrote: Hi Nilay, Thanks for your reply. When trying with gem5_fast I get the following error or commenting out the call to setupFuncEvents, I get the following error: fatal: Can't unserialize 'system.cpu0:locked' @ cycle 2573654901000[paramIn:build/ALPHA_MO

Re: [gem5-users] Ruby + OoO + with check-pointing support

Hi Nilay, Thanks for your reply. When trying with gem5_fast I get the following error or commenting out the call to setupFuncEvents, I get the following error: fatal: Can't unserialize 'system.cpu0:locked' @ cycle 2573654901000[paramIn:build/ALPHA_MOESI_hammer/sim/serialize.cc, line 228] I re

Re: [gem5-users] TLB statistics in ARM SE mode

Then why the stats of the TLBs are all zero in my runs ? Nathanaël Le 02/06/2012 19:58, Gabe Black a écrit : There *is* a TLB in SE mode, and address translation. The primary difference is that in SE mode on misses it's filled from a magical page table structure which the simulator maintains in

Re: [gem5-users] bzip2 can't run properly in x86 O3 mode

I faced that before. Thank to Ali, the problem is now fixed You should modify two files: 1) src/sim/process.cc you should find something like this: if (stack_base - stack_min > 8 * 1024 * 1024) fatal("Over max stack size for one thread\n"); 2) src/arch/x86/process.cc you should find two o

Re: [gem5-users] bzip2 can't run properly in x86 O3 mode

Hi, I am sorry to be too late to rely. I tracediffed the output of 8842 and 8841 and attached the output. I revised one place of the output format in 8841 to make the output more similar, but it seems the format changed a lot, and the diff may not be helpful. I also put a breakpoint, printed out th