Hi Joel, Sorry for my previous reply, the succesfully execution I had was in atomic mode, using the detailed cpu the problem persist. I downloaded the last gem5 version (gem5-782b7284de21) and run the simulation and the problem is still there for detailed cpu.
I simulated using the gem5 without any change, I copy the gem5 output: gem5 Simulator System. http://gem5.orggem5 is copyrighted software; use the --copyright option for details. gem5 compiled May 30 2013 14:39:58gem5 started May 30 2013 15:04:58gem5 executing on mulhacencommand line: ./build/X86/gem5.opt ./configs/example/fs.py --kernel x86_64-vmlinux-2.6.28.4-smp --script=./configs/boot/blackscholes_2c_simdev.rcS -n 2 --caches --l1d_size=32kB --l1d_assoc=8 --l1i_size=32kB --l1i_assoc=8 --l2_size=256kB --l2_assoc=8 --l2cache --cpu-type=detailed -r 1warn: add_child('terminal'): child 'terminal' already has parentGlobal frequency set at 1000000000000 ticks per secondinfo: kernel located at: /scratch/rrodriguez/gem5/gem5-stable/system/binaries/x86_64-vmlinux-2.6.28.4-smp 0: rtc: Real-time clock set to Sun Jan 1 00:00:00 2012Listening for com_1 connection on port 3456warn: Reading current count from inactive timer.0: system.remote_gdb.listener: listening for remote gdb #0 on port 70000: system.remote_gdb.listener: listening for remote gdb #1 on port 7001Switch at curTick count:10000info: Entering event queue @ 4878054131500. Starting simulation...Switched CPUS @ tick 4878054141500switching cpusinfo: Entering event queue @ 4878054141500. Starting simulation...hack: be nice to actually delete the event here**** REAL SIMULATION ****info: Entering event queue @ 4878054142000. Starting simulation...warn: Don't know what interrupt to clear for console.hack: Assuming logical destinations are 1 << id.warn: Tried to clear PCI interrupt 14 m5out/system.pc.com_1.terminalLoading new script...PARSE From: rodrigo.r...@hotmail.com To: gem5-users@gem5.org Date: Wed, 29 May 2013 18:40:49 +0000 Subject: Re: [gem5-users] Parsec problem. Joel, thanks for your help. I am using the version gem5-0a4b702628bd, I updated the file following your the report and all work well now. Thanks again Rodrigo Date: Wed, 29 May 2013 13:19:54 -0500 From: jthestn...@gmail.com To: gem5-users@gem5.org Subject: Re: [gem5-users] Parsec problem. Hi Rodrigo, Unfortunately, there isn't much here to help diagnose the problem you're running into. Your configuration and use of the kernel and disk image look correct for x86. The console output freeze symptom that you're experiencing looks similar to a previous problem we've run into with the x86 interrupts device. A temporary fix to that problem was committed in changeset 9624:43bd6562745e (here: http://repo.gem5.org/gem5?cmd=changeset;node=43bd6562745e). Are you using a revision of gem5 after 9624? If not, I would recommend updating to a more recent changeset and trying this again. If you are already using a revision after 9624, we'll probably need to do more debugging. In this case, please let me know, so we can move this thread over to the gem5-dev email list and I'll get you some info about how to analyze what is going wrong. Thanks, Joel On Tue, May 28, 2013 at 5:24 PM, Rodrigo Reynolds Ramírez <rodrigo.r...@hotmail.com> wrote: Hi Joel, 1. I am attaching you the scripts I am running, mibuilchkpt.sh: It is used for creating the checkpoint it finalized succefully. mibuildbench.sh It is used for running the benchmark (blackscholes)system.pc.com_1.teminal.checkpoint: The terminal output for the checkpoint creation system.pc.com_1.teminal.blackscholes: The terminal output for the benchmark execution blackscholes_2c_simdev.rcS: The benchmark boot file, created automatically I erased the switch cpu because did not work with Alpha neither X86hack_back_ckpt.rcS: Checkpoint creation file, I did not change this file. gem5Output.txt: The gem5 Output for benchmark executionconfig.ini: The simulation structure 2. After the freezing I left the computer simulating 24h, but never ends, gem5Output.txt is the output that gem5 prints 3. I runned again the simulation without attaching the terminal but the result is the same, it is shown in the files attached. I run the same simulation using the alpha image, I just change in mibuild* X86 by ALPHA and the kernel, and it worked correctly. Thanks Best Regards Rodrigo Date: Tue, 28 May 2013 12:58:38 -0500 From: jthestn...@gmail.com To: gem5-users@gem5.org Subject: Re: [gem5-users] Parsec problem. Hi Rodrigo, This doesn't appear to be easily reproducible, and I haven't run into this issue previously. I have a couple questions: 1) The /sbin/m5 error that you are getting appears to be incorrect. Can you reply and attach the script that you are executing in the simulated system, so we can see what you're trying to execute? 2) The freezing may not be incorrect functionality, but a symptom of another problem. Did you run this simulation until it exited? If not, how long did you let it run? If so, can you please reply and attach the output that gem5 prints? 3) I'd recommend not attaching a terminal/telnet to the simulation. For automated full-system simulation, the terminal output is automatically printed to the file system.pc.com_1.terminal (x86), and you can just tail this file instead of attaching a terminal. Note that by attaching the terminal, you may perturb what the simulated system is doing and cause non-deterministic results (e.g. it is suspicious that the first output you pasted here has "PARSE" and the second has "PARSEC" before it stops printing). Thanks, Joel On Mon, May 27, 2013 at 4:21 PM, Rodrigo Reynolds Ramírez <rodrigo.r...@hotmail.com> wrote: Hello everyone, I tried the image for x86 and the image for alpha. For both of them first at all I created a checkpoint, later I runned the backscholes benchmark, for both actions I used two cpus (-n 2), for running the bechmark I used a cpu detailed (--cpu-type=detailed). The benchmark for Alpha finished succefully but the benchmark for x86 did not finish, the execution freezes as I indicated before. It is strange that the simulation for a cpu-type atomic finish succefully. I copy the output for the terminal and telnet. Switched CPUS @ tick 4878000507500switching cpus**** REAL SIMULATION ****info: Entering event queue @ 4878000507500. Starting simulation... warn: Don't know what interrupt to clear for console.hack: Assuming logical destinations are 1 << id.warn: Tried to clear PCI interrupt 14 Trying 127.0.0.1...Connected to localhost.Escape character is '^]'.==== m5 slave terminal: Terminal 0 ====Loading new script...PARSEC Thanks in advance Rodrigo From: rodrigo.r...@hotmail.com To: gem5-users@gem5.org Date: Fri, 24 May 2013 18:37:48 +0000 Subject: Re: [gem5-users] Parsec problem. Hi Joel, I downloaded and runned a simulation using the new image, the simulation works well using an atomic CPU but it does not work with a detailed CPU with 2 CPUs. I copy the terminal output, for some reason the simulation freezes during the printf. Trying 127.0.0.1...Connected to localhost.Escape character is '^]'.==== m5 slave terminal: Terminal 0 ====Loading new script... usage: /sbin/m5 exit [delay] /sbin/m5 resetstats [delay [period]] /sbin/m5 dumpstats [delay [period]] /sbin/m5 dumpresetstats [delay [period]] /sbin/m5 readfile [filename] /sbin/m5 execfile <filename> /sbin/m5 checkpoint [delay [period]] /sbin/m5 loadsymbol <address> <symbol> /sbin/m5 initparam /sbin/m5 sw99param /sbin/m5 pin <cpu> <program> [args ...] All times in nanoseconds!PARSE Thanks in advance Rodrigo Date: Fri, 24 May 2013 12:02:15 -0500 From: jthestn...@gmail.com To: gem5-users@gem5.org Subject: Re: [gem5-users] Parsec problem. Hi Rodrigo, I see your problem. The disk image that you are trying to use is for the ALPHA instruction set (all binaries on the disk are compiled for the ALPHA ISA), but you're running gem5 using the x86 instruction set. The loop modprobe error indicated that you were trying to mount a disk with 32-bit Linux configs, but you will need a disk image for 64-bit (x86_64 to be specific). We've had plenty of demand for an x86 disk image, so this morning, I cleaned one up and uploaded it to our site. Not all of the benchmarks work (e.g. facesim, ferret and vips are known to fail, and rtview is not on the image). That said, I've thoroughly tested blackscholes, bodytrack, canneal, dedup, fluidanimate, freqmine, streamcluster, swaptions and x264, and I can confirm that they work correctly. If you spend any time on the rest of the benchmarks to get them to work, we would really appreciate if you could contribute back binaries that can be added to the disk image. I've updated the PARSEC on gem5 site (http://www.cs.utexas.edu/~parsec_m5/), and you can download the new image here: http://www.cs.utexas.edu/~parsec_m5/x86root-parsec.img.bz2 Hope this helps. Please let me know if you have any trouble, Joel On Fri, May 24, 2013 at 6:58 AM, Rodrigo Reynolds Ramírez <rodrigo.r...@hotmail.com> wrote: Hello Joel, I am trying to boot the kernel x86_64-vmlinux-2.6.28.4-smp, I downloaded it from http://www.cs.utexas.edu/~parsec_m5 I have also tried with x86_64-vmlinux-2.6.22.9.smp but neither worked. I include the config.ini file. Thanks in advance Rodrigo Date: Thu, 23 May 2013 19:02:25 -0500 From: jthestn...@gmail.com To: gem5-users@gem5.org Subject: Re: [gem5-users] Parsec problem. @Rodrigo: Which kernel are you trying to boot? Can you please attach the config.ini file that gem5 produced when you ran this? Thanks, Joel On Thu, May 23, 2013 at 5:06 PM, Rodrigo Reynolds Ramírez <rodrigo.r...@hotmail.com> wrote: Thanks Joel, that explains why Archive Manager opened the file, I downloaded the file again,, I extracted it using bunzip and I get the same error. RPC: Registered udp transport module. RPC: Registered tcp transport module.input: PS/2 Generic Mouse as /class/input/input1EXT2-fs warning: checktime reached, running e2fsck is recommendedVFS: Mounted root (ext2 filesystem). Freeing unused kernel memory: 332k freedserial8250: too much work for irq4request_module: runaway loop modprobe binfmt-464crequest_module: runaway loop modprobe binfmt-464c request_module: runaway loop modprobe binfmt-464crequest_module: runaway loop modprobe binfmt-464crequest_module: runaway loop modprobe binfmt-464c I have been looking for these error in Internet and some people says it could means the filesystem is damaged. I have tried to build my own image but it did not work. Is there another image for parsec available for the users? Rodrigo Date: Thu, 23 May 2013 16:07:30 -0500 From: jthestn...@gmail.com To: gem5-users@gem5.org Subject: Re: [gem5-users] Parsec problem. Hi Rodrigo, The disk image isn't a tar archive, but instead it is bzipped to save space and download time. You simply need to unzip it: % bunzip2 linux-parsec-2-1-m5-with-test-inputs.img.bz2 Hope this helps, Joel On Thu, May 23, 2013 at 12:39 PM, Rodrigo Reynolds Ramírez <rodrigo.r...@hotmail.com> wrote: The problem is that the disk image is damage, when I extracted it using tar I receive this error: tar -xvjf linux-parsec-2-1-m5-with-test-inputs.img.bz2 tar: This does not look like a tar archive tar: Skipping to next header tar: Exiting with failure status due to previous errors But I could extracted it with Archive Manager. I tried to boot with another image, http://www.m5sim.org/dist/current/x86/x86-system.tar.bz2%5D but I cannot make the parsec work with it. Does somebody has a working image with parsec? Thanks in advance Rodrigo From: rodrigo.r...@hotmail.com To: gem5-users@gem5.org Date: Thu, 23 May 2013 17:08:08 +0000 Subject: [gem5-users] Parsec problem. Hello everyone, I am trying to run the parsec suite, I download the file provided here: http://www.cs.utexas.edu/~parsec_m5/ but I got a error when I run the system, I got these message: VFS: Mounted root (ext2 filesystem). Freeing unused kernel memory: 332k freed request_module: runaway loop modprobe binfmt-464c request_module: runaway loop modprobe binfmt-464c request_module: runaway loop modprobe binfmt-464c request_module: runaway loop modprobe binfmt-464c request_module: runaway loop modprobe binfmt-464c Connection closed by foreign host. My running command is: ./build/X86/gem5.opt ./configs/example/fs.py \ --kernel x86_64-vmlinux-2.6.28.4-smp \ --script=./configs/boot/blackscholes_2c_simdev.rcS \ -n 1 --caches --l2cache \ --l1d_size=32kB --l1i_size=32kB --l2_size=256kB \ --l1d_assoc=8 --l1i_assoc=8 --l2_assoc=8 --cpu-type=atomic I have tried changing the number of cpus and the cpu-type, I always have the same error. Also I tried with the hack_back_ckpt.rcS script. Thanks in advance Rodrigo _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users -- Joel Hestness PhD Student, Computer Architecture Dept. of Computer Science, University of Wisconsin - Madison http://pages.cs.wisc.edu/~hestness/ _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users -- Joel Hestness PhD Student, Computer Architecture Dept. of Computer Science, University of Wisconsin - Madison http://pages.cs.wisc.edu/~hestness/ _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users -- Joel Hestness PhD Student, Computer Architecture Dept. of Computer Science, University of Wisconsin - Madison http://pages.cs.wisc.edu/~hestness/ _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users -- Joel Hestness PhD Student, Computer Architecture Dept. of Computer Science, University of Wisconsin - Madison http://pages.cs.wisc.edu/~hestness/ _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users -- Joel Hestness PhD Student, Computer Architecture Dept. of Computer Science, University of Wisconsin - Madison http://pages.cs.wisc.edu/~hestness/ _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users _______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
_______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
