Hi Ali and GEM5 Users I have tried again and could boot both ALPHA and X86, as follows:
> Time to shell prompt at ALPHA: 0m38.826s (running on [Intel i7 Extreme]@[3.33GHz+16GB RAM]) $ cd /home/user/gem5; export M5_PATH=/home/user/gem5; time ./build/ALPHA/gem5.opt --outdir=/tmp/output ./configs/example/fs.py --cpu-type=atomic --clock=2000000000 --mem-size=256MB --caches --disk-image=./linux-latest.img --kernel=./binaries/vmlinux --num-cpus=1 > Time to shell prompt at X86_64: 2m35.474s (running on [Intel i7 Extreme]@[3.33GHz+16GB RAM]) $ cd /home/user/gem5; export M5_PATH=/home/user/gem5; time ./build/X86/gem5.opt --outdir=/tmp/output ./configs/example/fs.py --cpu-type=atomic --clock=2000000000 --mem-size=256MB --caches --disk-image=./linux-x86.img --kernel=./binaries/x86_64-vmlinux-2.6.22.9 --num-cpus=1 I have tried also to run the following hello.rcS script: --- cut here --- #!/bin/sh # Wait for system to calm down sleep 10 # Reset the stats m5 resetstats # Run hello /hello # Exit the simulation m5 exit --- cut here --- and it worked at ALPHA, as follows: > Time to "hello" and finish simulation: 1m26.648s (running on [Intel i7 Extreme]@[3.33GHz+16GB RAM]) $ cd /home/user/gem5; export M5_PATH=/home/user/gem5; time ./build/ALPHA/gem5.opt --outdir=/tmp/output ./configs/example/fs.py --cpu-type=atomic --clock=2000000000 --mem-size=256MB --caches --disk-image=./linux-latest.img --kernel=./binaries/vmlinux --num-cpus=1 --script=./configs/boot/hello.rcS but it does not work on X86_64, as follows: > Does not work yet! Script doesn't start but keep running without any error message... $ cd /home/user/gem5; export M5_PATH=/home/user/gem5; time ./build/X86/gem5.opt --outdir=/tmp/output ./configs/example/fs.py --cpu-type=atomic --clock=2000000000 --mem-size=256MB --caches --disk-image=./linux-x86.img --kernel=./binaries/x86_64-vmlinux-2.6.22.9 --script=./configs/boot/hello.rcS --num-cpus=1 Well, I think now all architectures (ARM, X86, and ALPHA) boots on FS mode and the remaining problem is to see how to fix running scripts on a X86 machine. After that, I will start to test other parameters I was using before this experiments, to see which is not working well when I was simulating ALPHA or X86, using my instance of GEM5. Also starts to use checkpoints to switch to O3CPU, backing to the focus of the original forum message. I hope it helps and thanks in advance for your attention and time. Regards Romulo On Tue, Jul 24, 2012 at 8:05 PM, Romulo Curty Cerqueira <cu...@ieee.org>wrote: > Hi Ali > > I am a very new user and after your answer I tried the following > simulations: > > 1) I have tried to simulate X86 boot in FS mode using --cpu-type=atomic > all the time, and running only a "hello program", for 1165m59.515s with no > success or any error message. > > 2) I have also tried to simulate ALPHA boot in FS mode using > --cpu-type=atomic all the time, and running only a "hello program", for > 1195m10.024s with no success or any error message. > > I suppose my machine is able to simulate this scenarios in <1h because it > is a [Intel i7 Extreme]@[3.33GHz+16GB RAM] machine, as briefly follows: > > $ uname -a > Linux i7-extreme 3.3.8-1.fc16.x86_64 #1 SMP Mon Jun 4 20:49:02 UTC 2012 > x86_64 x86_64 x86_64 GNU/Linux > > # lshw -short > H/W path Device Class Description > ==================================================== > system () > /0 bus Intel DX58SO Motherboard > /0/0 processor Intel(R) Core(TM) i7 CPU > 975 @ 3.33GHz > /0/0/1 memory 8MiB L3 cache > /0/0/3 memory 256KiB L2 cache > /0/0/4 memory 32KiB L1 cache > /0/2 memory 32KiB L1 cache > /0/5 memory 64KiB BIOS > /0/14 memory 16GiB System Memory > /0/14/0 memory 4GiB DIMM DDR3 Synchronous 1333 > MHz (0.8 ns) > /0/14/1 memory 4GiB DIMM DDR3 Synchronous 1333 > MHz (0.8 ns) > /0/14/2 memory 4GiB DIMM DDR3 Synchronous 1333 > MHz (0.8 ns) > /0/14/3 memory 4GiB DIMM DDR3 Synchronous 1333 > MHz (0.8 ns) > /0/100 bridge 5520/5500/X58 I/O Hub to ESI Port > /0/100/1 bridge 5520/5500/X58 I/O Hub PCI Express > Root Port 1 > /0/100/3 bridge 5520/5500/X58 I/O Hub PCI Express > Root Port 3 > /0/100/3/0 display GT200b [GeForce GTX 285] > /0/100/7 bridge 5520/5500/X58 I/O Hub PCI Express > Root Port 7 > /0/100/10 generic 5520/5500/X58 Physical and Link > Layer Registers Port 0 > /0/100/10.1 generic 5520/5500/X58 Routing and > Protocol Layer Registers Port 0 > /0/100/14 generic 5520/5500/X58 I/O Hub System > Management Registers > /0/100/14.1 generic 5520/5500/X58 I/O Hub GPIO and > Scratch Pad Registers > /0/100/14.2 generic 5520/5500/X58 I/O Hub Control > Status and RAS Registers > /0/100/14.3 generic 5520/5500/X58 I/O Hub Throttle > Registers > ... > /0/100/1e bridge 82801 PCI Bridge > /0/100/1e/3 bus TSB43AB22A IEEE-1394a-2000 > Controller (PHY/Link) [iOHCI-Lynx] > /0/100/1f bridge 82801JIR (ICH10R) LPC Interface > Controller > /0/100/1f.2 storage 82801 SATA RAID Controller > /0/100/1f.3 bus 82801JI (ICH10 Family) SMBus > Controller > /0/101 bridge Xeon 5500/Core i7 QuickPath > Architecture Generic Non-Core Registers > /0/102 bridge Xeon 5500/Core i7 QuickPath > Architecture System Address Decoder > /0/103 bridge Xeon 5500/Core i7 QPI Link 0 > /0/104 bridge Xeon 5500/Core i7 QPI Physical 0 > /0/105 bridge Xeon 5500/Core i7 Integrated > Memory Controller > /0/106 bridge Xeon 5500/Core i7 Integrated > Memory Controller Target Address Decoder > /0/107 bridge Xeon 5500/Core i7 Integrated > Memory Controller Test Registers > /0/108 bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 0 Control Registers > /0/109 bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 0 Address Registers > /0/10a bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 0 Rank Registers > /0/10b bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 0 Thermal Control Registers > /0/10c bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 1 Control Registers > /0/10d bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 1 Address Registers > /0/10e bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 1 Rank Registers > /0/10f bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 1 Thermal Control Registers > /0/110 bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 2 Control Registers > /0/111 bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 2 Address Registers > /0/112 bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 2 Rank Registers > /0/113 bridge Xeon 5500/Core i7 Integrated > Memory Controller Channel 2 Thermal Control Registers > /0/1 scsi0 storage > /0/1/0.0.0 /dev/sda disk 1TB ST31000528AS > ... > /0/3 scsi1 storage > /0/3/0.0.0 /dev/sdb disk 1TB ST31000528AS > /0/3/0.0.0/1 /dev/sdb1 volume 931GiB Linux raid autodetect > partition > /0/4 scsi2 storage > /0/4/0.0.0 /dev/sdc disk 1TB ST31000528AS > /0/4/0.0.0/1 /dev/sdc1 volume 931GiB Linux raid autodetect > partition > /0/6 scsi3 storage > /0/6/0.0.0 /dev/sdd disk 1TB ST31000528AS > /0/6/0.0.0/1 /dev/sdd1 volume 931GiB Linux raid autodetect > partition > /0/7 scsi4 storage > /0/7/0.0.0 /dev/sde disk 1TB ST31000528AS > /0/7/0.0.0/1 /dev/sde1 volume 931GiB Linux raid autodetect > partition > /0/8 scsi5 storage > /0/8/0.0.0 /dev/cdrom2 disk DVD RW AD-7240S > ... > > I suppose to have a GEM5 instance compiled according to the spec because I > have followed the instructions from the GEM5 official web site, and I can > boot ARM architectures in FS mode and run the "hello program" without any > problem or error message. > > But I'm not the only user having problems to boot architectures in FS > mode, mainly for X86 architecture. > > What should we do? > > Thanks in advance for your attention and time. > > Regards > > Romulo > > > > On Mon, Jul 23, 2012 at 12:44 AM, Romulo Curty Cerqueira > <cu...@ieee.org>wrote: > >> A: ... Normally people boot linux with a atomic simple CPU, take a >> checkpoint and then switch to the O3 cpu to run their application of >> interest. >> ..., so it seems like there is something wrong on your end. >> >> >> Ok. I will do that boot and check my end. >> >> Thank you Ali :-) >> >> >> >> On Mon, Jul 23, 2012 at 12:36 AM, Ali Saidi <sa...@umich.edu> wrote: >> >>> All models should boot in <1h, although there is no particularly good >>> reason to boot the OS with the O3 cpu. Normally people boot linux with a >>> atomic simple CPU, take a checkpoint and then switch to the O3 cpu to run >>> their application of interest. >>> >>> We have regression tests that run weekly using the disk images that take >>> approximately 1h, so it seems like there is something wrong on your end. >>> >>> Ali >>> >>> On Jul 22, 2012, at 10:22 PM, Romulo Curty Cerqueira wrote: >>> >>> Hi Anderson >>> >>> I think that now I understand your question because I have tried to >>> simulate the "hello" test program (that comes with GEM5) using gem5.opt for >>> the first time at FS Alpha and FS x86, and the surprising results are: >>> >>> Architecture Simulation Time(*) >>> ---------------------------------------------- >>> ARM 4 CPUs 66m50.399s >>> ALPHA 1 CPU >749m39.720s (unfinished) >>> X86 1 CPU >675m49.498s (unfinished) >>> ---------------------------------------------- >>> >>> (*) All tests running "alone" at same [Intel i7 Extreme]@[3.33GHz+16GB >>> RAM] machine. >>> >>> It is really a strange FS behavior and I don't know how to explain that. >>> Maybe the disk images (provided by GEM5 download page) are not >>> lightweight enough to provide a good comparison between simulation time for >>> these architectures running at FS mode. But I'm not sure if it could be the >>> main reason. >>> >>> Regards >>> >>> Romulo >>> >>> >>> On Tue, Jul 17, 2012 at 12:58 PM, Anderson Faustino < >>> andersonfaust...@gmail.com> wrote: >>> >>>> Hi, Romulo. >>>> >>>> I didn't explain correct my problem, in my last message. There is no >>>> problem with ALPHA. >>>> >>>> But, X86 doesn't boot with 4 O3CPU, the other cpus work fine, but 03CPU >>>> doesn't. >>>> >>>> Yesterday, I started a boot: >>>> >>>> ---------------- >>>> build/X86/gem5.opt configs/example/fs.py >>>> --kernel=x86_64-vmlinux-2.6.22.9.smp --cpu-type=detailed --l2cache --caches >>>> --script=run.sh -n 4 >>>> ---------------- >>>> >>>> And until now it didn't finish. The process is stopped in: >>>> >>>> Initializing CPU#3 >>>> Calibrating delay loop (skipped)... 3999.96 BogoMIPS preset >>>> CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line) >>>> CPU: L2 Cache: 1024K (64 bytes/line) >>>> Fake M5 x86_64 CPU stepping 01 >>>> Brought up 4 CPUs >>>> migration_cost=16 >>>> >>>> Until now, it took 1173 minutes. >>>> >>>> I really want to understand: why doesn't a configuration with 4 O3CPU >>>> boot the system? But other works fine. >>>> >>>> Thanks. >>>> >>> >>> _______________________________________________ >>> 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
