Re: [gem5-users] gem5 Mount another disk.

2016-07-19 Thread Oscar Rosell 
Hi,

Not sure why you're getting that. I would try this (in order):

1-  Try doing the umount again.
2-  Reboot the machine (if you can) and try again

Oscar Rosell - Metempsywww.metempsy.com



 On Tue, 19 Jul 2016 17:06:45 +0100 Yifeng Liu yif...@udel.edu 
wrote  

Hello all:

I am new to gem5. I previously mount a ARM disk by using the following command:
sudo mount -o loop,offset=32256 linux-aarch32-ael.img /mnt
sudo umount /mnt
It works fine. however, i want to mount a X86 disk to /mnt 
and I use the following command

sudo mount -o loop,offset=32256 linux-x86.img /mnt

and it gives me a error.

mount: according to mtab /home/yifeng/gem5-Dir/FS-files/X86/disks/linux-x86.img 
is already mounted on /mnt as loop

Any helps would be appreciate. 


___
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

Re: [gem5-users] fail to run a c application in SE mode

2016-07-08 Thread Oscar Rosell 
Hi,

I'd think that the error is just what Gem5 says:

fatal: Object file is a dynamic executable however only static executables are 
supported!
   Please recompile your executable as a static binary and try again.

You can check that the executable is static by using file command:

file /home/anoir/FFT\ work/FFT_opt_v2/fast_fourier_transform

And it will tell you if your binary is "statically linked" or "dynamically 
linked". If it's dynamically linked then you should link your program using 
-static flag. The command line you provided for compilation

gcc -c -static -Wall  fast_fourier_transform.c -lm -o fast_fourier_transform.o

Is only for compilation phase. There should be a linking command that you 
didn't provide and -static should be on that command line. This should compile 
and statically link your program:

gcc -static -Wall fast_fourier_transform.c -lm -o fast_fourier_transform

Thanks,

Oscar
 On Thu, 07 Jul 2016 19:08:42 +0200 anoir nechianoirne...@gmail.com 
wrote  

HI


I'm trying to run a c application of radix4 on gem5 simulator using SE mode. 
This is the link to the code:

http://gweep.net/~rocko/FFT/node6.html#SECTION0006


i used this command to compile it :
gcc -c -static -Wall  fast_fourier_transform.c -lm -o fast_fourier_transform.o


then i used this command to run it :
build/ARM/gem5.opt configs/example/se.py -c /home/anoir/FFT\ 
work/FFT_opt_v2/fast_fourier_transform


and i'm getting this error:

gem5 Simulator System.  http://gem5.org
gem5 is copyrighted software; use the --copyright option for details.

gem5 compiled Jun  9 2016 12:22:14
gem5 started Jul  7 2016 18:53:41
gem5 executing on anoir-Lenovo-IdeaPad-Y510P
command line: build/ARM/gem5.opt configs/example/se.py -c '/home/anoir/FFT 
work/FFT_opt_v2/fast_fourier_transform'

Global frequency set at 1 ticks per second
warn: DRAM device capacity (8192 Mbytes) does not match the address range 
assigned (512 Mbytes)
fatal: Object file is a dynamic executable however only static executables are 
supported!
   Please recompile your executable as a static binary and try again.
 @ tick 0
[create:build/ARM/sim/process.cc, line 644]
Memory Usage: 637548 KBytes
Program aborted at cycle 0
Aborted (core dumped)


could someone tell me what's wrong!!

Thank you
-- 
Anouar NECHIIT Engineer : Industrial systems
Higher Institute of Computer Science
Tunis - El Manar University
Phone : (+216) 50 311 536
E-mail : anoirne...@gmail.com


 



 ___
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

Re: [gem5-users] warn: Returning zero for read from miscreg pmccntr

2016-07-11 Thread Oscar Rosell 
Hi,

I'm not sure if what you're trying to do should work or not. Somebody else may 
be able to help you. 


However, given that you are running your application on Gem5 simulator and 
depending of what you exactly want to achieve, you can alternatively instrument 
your code with m5 ops. An example:


#include "util/m5/m5op.h"


int main()
{
for (int i = 0; i  10; ++i)
{
m5_dumpreset_stats(0,0);
int sum = 1;
for (int j = 0; j  1000; ++j)
sum += sum;
}
return 0;
}



Each m5_dumpreset_stats call should generate a new section in the stats file 
(and reset the stat values). From there you can get the number of cycles and 
many more stats.

Oscar - Metempsywww.metempsy.com



 On Sun, 10 Jul 2016 16:56:24 +0200 anoir nechianoirne...@gmail.com 
wrote  

Hi everybody


I'm trying to run a c application on SE mode using ARM (v7)... in order to get 
performance for a specific loop in the code i did an rdtsc() function that help 
me to het the number of cycles it take the loop to run. this is the code :

a  function to initialize the registers:

static inline void init_perfcounters (int32_t do_reset, int32_t enable_divider)
{
  // in general enable all counters (including cycle counter)
  int32_t value = 1;

  // peform reset:  
  if (do_reset)
  {
value |= 2; // reset all counters to zero.
value |= 4; // reset cycle counter to zero.
  } 

  if (enable_divider)
value |= 8; // enable "by 64" divider for CCNT.

  value |= 16;

  // program the performance-counter control-register:
  asm volatile ("MCR p15, 0, %0, c9, c12, 0\t\n" :: "r"(value));  

  // enable all counters:  
  asm volatile ("MCR p15, 0, %0, c9, c12, 1\t\n" :: "r"(0x800f));  

  // clear overflows:
  asm volatile ("MCR p15, 0, %0, c9, c12, 3\t\n" :: "r"(0x800f));
}


another function to read from the counter:

long long unsigned loop_start,loop_end,radix_start,radix_end,loop_cc,radix_cc;
static inline long long unsigned rdtsc32(void)
{
long long unsigned r = 0;
asm volatile("mrc p15, 0, %0, c9, c13, 0" : "=r"(r) );
return r;
}


then i called this functions in main as following:

loop_start = rdtsc32();

 /*loop*/

loop_end = rdtsc32();
loop_cc=loop_start-loop_end;


unfortunately i keep getting this warning while executing:

warn: Returning zero for read from miscreg pmccntr

meaning that i will get zero in the Cycle Count register!


Can someone help please?


Thank you


-- 
Anouar NECHIIT Engineer : Industrial systems
Higher Institute of Computer Science
Tunis - El Manar University
Phone : (+216) 50 311 536
E-mail : anoirne...@gmail.com


 




 ___
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

Re: [gem5-users] Stuck at "initiateAcc not defined!"

2016-09-07 Thread Oscar Rosell 
As Jason suggested I tried with the latest version of the script 
(./configs/learning_gem5/part1/simple.py) and it worked OK.



gem5 compiled Sep  7 2016 10:51:54

gem5 started Sep  7 2016 11:16:01

gem5 executing on orosell-Inspiron-3847, pid 27899

command line: ./build/X86/gem5.opt ./configs/learning_gem5/part1/simple.py



Global frequency set at 1 ticks per second

warn: DRAM device capacity (8192 Mbytes) does not match the address range 
assigned (512 Mbytes)

0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000

warn: ClockedObject: More than one power state change request encountered 
within the same simulation tick

Beginning simulation!

info: Entering event queue @ 0.  Starting simulation...

Hello world!

Exiting @ tick 454507000 because target called exit()



Oscar Rosell - Metempsy





 On Wed, 07 Sep 2016 09:38:57 +0100 Oscar Rosell 
oscar.ros...@metempsy.comwrote  




Hi,



Did you try Jason's solution? Did it work?



Thanks,



Oscar Rosell - Metempsy





 On Wed, 07 Sep 2016 09:23:02 +0100 Uma S umasuji...@gmail.comwrote 
 










___

gem5-users mailing list

gem5-users@gem5.org

http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users


Dear all,

I had posted a question on 2016/08/30 with the subject, 

Stuck at "initiateAcc not defined!"

I am thankful that two persons replied to that. As per 

Oscar Rosell 's reply I am attaching the "simple.py" file for further
look. Kindly let me know how I can solve this.
_S.Uma

___

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

Re: [gem5-users] Stuck at "initiateAcc not defined!"

2016-09-07 Thread Oscar Rosell 
Hi,



Did you try Jason's solution? Did it work?



Thanks,



Oscar Rosell - Metempsy





 On Wed, 07 Sep 2016 09:23:02 +0100 Uma S umasuji...@gmail.comwrote 
 




Dear all,

I had posted a question on 2016/08/30 with the subject, 

Stuck at "initiateAcc not defined!"


I am thankful that two persons replied to that. As per 

Oscar Rosell 's reply I am attaching the "simple.py" file for further
look. Kindly let me know how I can solve this.
_S.Uma

___

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

Re: [gem5-users] Stuck at "initiateAcc not defined!"

2016-08-31 Thread Oscar Rosell 
Could you send the contents of simple.py file?

Thanks,

Oscar Rosell - Metempsy


 On Wed, 31 Aug 2016 07:32:29 +0100 Uma Sumasuji...@gmail.com wrote 
 

Hello all,


I tried to follow the tutorial as per the instructions in the wisc.edu 
instructions page to run gem5.


I downloaded the simple.py file using the link given in the above said page. 

I tried to run gem5 from the root gem5 directory as: build/X86/gem5.opt 
configs/tutorial/simple.py I got the output as follows.
I dont know where to look for the core dump also.
Kindly help to run this code



suma@suma-G31T-M:~/g51/gem5-stable$ build/X86/gem5.opt -d /home/suma/g51out/1 
configs/tutorial/simple.py
gem5 Simulator System. http://gem5.org
gem5 is copyrighted software; use the --copyright option for details.

gem5 compiled Aug 30 2016 18:29:26
gem5 started Aug 30 2016 14:39:49
gem5 executing on suma-G31T-M
command line: build/X86/gem5.opt -d /home/suma/g51out/1 
configs/tutorial/simple.py

Global frequency set at 1 ticks per second
warn: DRAM device capacity (8192 Mbytes) does not match the address range 
assigned (512 Mbytes)
0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000
Beginning simulation!
info: Entering event queue @ 0. Starting simulation...
panic: initiateAcc not defined!
 @ tick 74000
[initiateAcc:build/X86/cpu/static_inst.hh, line 272]
Memory Usage: 586308 KBytes
Program aborted at cycle 74000
Aborted (core dumped)

 ___
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

Re: [gem5-users] Stuck at "initiateAcc not defined!"

2016-09-09 Thread Oscar Rosell 
The problem is not that scons is not installed. The problem is that it cannot 
find the SConstruct file. You have to be at the root of the repo to build the 
model. Just go there and do "scons build/X86/gem5.opt -j3"



Thanks,



Oscar Rosell - Metempsy





 On Fri, 09 Sep 2016 13:20:29 +0100 Uma S umasuji...@gmail.comwrote 
 




Hello all,


As per Mr.Jason Lowe-Power's suggesion,

I am trying to run the configs/learning_gem5/simple.py from current gem5.

so I have cloned and  gem5(current) in another directory.


but the following command returns error inside gem5(current) directory. I t was 
working alright in the gem5-stable directory.



scons CPU_MODELS="AtomicSimpleCPU,MinorCPU,O3CPU,TimingSimpleCPU" 
build/X86/gem5.opt -j3




output is 



suma@suma-G31T-M:~/g52$ scons 
CPU_MODELS="AtomicSimpleCPU,MinorCPU,O3CPU,TimingSimpleCPU" build/X86/gem5.opt 
-j3



scons: *** No SConstruct file found.

File "/usr/lib/scons/SCons/Script/Main.py", line 905, in _main




so I gave 

suma@suma-G31T-M:~/g52$ whereis scons



answer is



scons: /usr/bin/scons /usr/lib/scons /usr/bin/X11/scons 
/usr/share/man/man1/scons.1.gz



I dont know why I am not able to build the gem5 current. Please help me


___

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

Re: [gem5-users] Stuck at "initiateAcc not defined!"

2016-09-09 Thread Oscar Rosell 
Hi,



The file is in latest gem5 repository:



http://repo.gem5.org/gem5/file/8bc53d5565ba/configs/learning_gem5/part1/simple.py



There's a gem5 repository and an stable repository.



http://www.gem5.org/Repository



The file we're talking about is in gem5 repo. Not sure what's your current-gem5 
repo.



Thanks,



Oscar Rosell - Metempsy





 On Fri, 09 Sep 2016 12:41:27 +0100 Uma S umasuji...@gmail.comwrote 
 




Hi,

 I had posted a question on 2016/08/30 with the subject, Stuck at "initiateAcc 
not defined!"

As per Mr.Jason Lowe-Power's advice, I now tried with current-gem5 instead of 
stable version.
But I could not find (./configs/learning_gem5/part1/simple.py) link in the 
internet?
What am I missing here?
Actually, I had downloaded the "simple.[y" file from 

http://pages.cs.wisc.edu/~david/courses/cs752/Fall2015/gem5-tutorial/_downloads/simple.py

link provided in the tutorial.

I did hg clone to latest gem5 version and did update and merge steps. I used 
the 
above said simple.py file as found in the attachment . But still getting that 
error.

Kindly help.
The screen said,.

suma@suma-G31T-M:~/g51/gem5-stable$ build/X86/gem5.opt -d /home/suma/g51out/2 
configs/tutorial/simple.py

gem5 Simulator System. http://gem5.org

gem5 is copyrighted software; use the --copyright option for details.



gem5 compiled Aug 30 2016 18:29:26

gem5 started Sep 9 2016 16:48:27

gem5 executing on suma-G31T-M

command line: build/X86/gem5.opt -d /home/suma/g51out/2 
configs/tutorial/simple.py



Global frequency set at 1 ticks per second

warn: DRAM device capacity (8192 Mbytes) does not match the address range 
assigned (512 Mbytes)

0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000

Beginning simulation!

info: Entering event queue @ 0. Starting simulation...

panic: initiateAcc not defined!

 @ tick 74000

[initiateAcc:build/X86/cpu/static_inst.hh, line 272]

Memory Usage: 586312 KBytes

Program aborted at cycle 74000

Aborted (core dumped)


___

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

Re: [gem5-users] gem5 checkpoint - Checkpoint not found

2016-09-19 Thread Oscar Rosell 
Hi,



I think (but not sure) that "-r 135051211305000" is wrong. If you have the 
checkpoint in the directory where you are running the command and there's only 
1 checkpoint there try "-r 1". There's a parameter to change the directory 
where the checkpoint is located (--checkpoint-dir).



Thanks,


Oscar Rosell - Metempsy





 On Mon, 19 Sep 2016 09:56:19 +0100 Jasmin Jahic 
jasmin.ja...@gmail.comwrote  




Hello,



I am trying to work with the checkpoints. I have followed instructions at 
http://www.m5sim.org/Checkpoints. 



I make a checkpoint using m5 checkpoint, from telnet console. The checkpoint is 
stored in m5out. 



Then I try to start the gem5 using the checkpoint:

build/X86/gem5.opt configs/example/fs.py -r 135051211305000 --mem-size=512MB 
--kernel=x86_64-vmlinux-2.6.22.9 --disk-image=linux-.img



However, I receive an error:

fatal: Checkpoint 135051211305000 not found



Do you perhaps have any idea what could be the problem? Maybe I should set some 
system path?



Best regards,

Jasmin



___

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

Re: [gem5-users] undefined reference to `vtable for Port

2017-07-15 Thread Oscar Rosell 
Hi,

This looks just like a C++ error. Check solution for: 
https://stackoverflow.com/questions/3065154/undefined-reference-to-vtable 


Regards,

   Oscar

> On 15 Jul 2017, at 17:47, Neu, Markus  wrote:
> 
> Hallo,
> 
> at the moment i try to add a additional port to the cache.cc/.hh 
> . The idea is to send a copy of the packets to a new 
> module.
> I use a example from: learning.gem5.org/book/part2/memoryobject.html 
>  as template for the 
> port implementation. Now i have reached a point were the syntax generates no 
> more problems. But the compiler stops with errors:
> 
> 
> build/X86/mem/cache/lib.o.partial: In function 
> `Cache::DapuSidePort::DapuSidePort(std::__cxx11::basic_string std::char_traits, std::allocator > const&, Cache*)':
> /home/osboxes/gem5_raw/build/X86/mem/cache/cache.hh:204: undefined reference 
> to `vtable for Cache::DapuSidePort'
> build/X86/mem/cache/lib.o.partial: In function 
> `Cache::DapuSidePort::~DapuSidePort()':
> /home/osboxes/gem5_raw/build/X86/mem/cache/cache.hh:190: undefined reference 
> to `vtable for Cache::DapuSidePort'
> /home/osboxes/gem5_raw/build/X86/mem/cache/cache.hh:190: undefined reference 
> to `vtable for Cache::DapuSidePort'
> build/X86/exposed_obj/lib.o.partial: In function 
> `Dapu::CPUSidePort::CPUSidePort(std::__cxx11::basic_string std::char_traits, std::allocator > const&, Dapu*)':
> /home/osboxes/gem5_raw/build/X86/exposed_obj/Dapu.hh:43: undefined reference 
> to `vtable for Dapu::CPUSidePort'
> /home/osboxes/gem5_raw/build/X86/exposed_obj/Dapu.hh:43: undefined reference 
> to `vtable for Dapu::CPUSidePort'
> build/X86/exposed_obj/lib.o.partial: In function 
> `Dapu::CPUSidePort::~CPUSidePort()':
> /home/osboxes/gem5_raw/build/X86/exposed_obj/Dapu.hh:25: undefined reference 
> to `vtable for Dapu::CPUSidePort'
> /home/osboxes/gem5_raw/build/X86/exposed_obj/Dapu.hh:25: undefined reference 
> to `vtable for Dapu::CPUSidePort'
> /home/osboxes/gem5_raw/build/X86/exposed_obj/Dapu.hh:25: undefined reference 
> to `vtable for Dapu::CPUSidePort'
> build/X86/exposed_obj/lib.o.partial:/home/osboxes/gem5_raw/build/X86/exposed_obj/Dapu.hh:25:
>  more undefined references to `vtable for Dapu::CPUSidePort' follow
> collect2: error: ld returned 1 exit status
> scons: *** [build/X86/gem5.opt] Error 1
> 
> 
> The gem5 version which i use is from today but the template seems to be much 
> older. The error massages point to the constructor and class definitions:
> 
> 
> Dapu.hh:25 --> class CPUSidePort : public SlavePort
> 
> Dapu.hh:43 -->CPUSidePort(const std::string& name, Dapu *owner) :
> SlavePort(name, owner), owner(owner), 
> needRetry(false),
> blockedPacket(nullptr)
> { }
> 
> 
> Unfortunately i don't understand what the compiler indicates but would assume 
> a missing library. I would appreciate some hints.
> 
> Regards,
> Markus Neu
> ___
> 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

Re: [gem5-users] reading stats.txt in full system mode

2017-07-19 Thread Oscar Rosell 

Hi,

As you say the stats file displays all the results. If your image 
includes the m5 binary then you can reset the stats by adding to the rcS 
file the following line:


/sbin/m5 resetstats

Just before the call to ls. That should work.

Regards,

Oscar

On 19/07/17 13:46, Saad Sheikh wrote:

Hello,
I am new to gem5. I ran a simple 'ls' script on full system mode using 
the example/fs.py file i.e. "build/ARM/gem5.opt configs/example/fs.py 
--disk-image=/home/saad/arm_image/disks/linux-aarch32-ael.img 
--script=configs/boot/test.rcS" where test.rcS contains the command to 
list the current directory items.
Using the stats.txt file, how to I differentiate between the results 
(i.e. no. of hits) caused specifically by the system call 'ls' only 
and those caused by the complete running system because I am assuming 
the that stats.txt file display all the results caused by the system 
from beginning to end(i.e. system startup, kernel system calls, etc..)

Thank you!


___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users




smime.p7s
Description: S/MIME Cryptographic Signature
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] Performance

2017-06-28 Thread Oscar Rosell 

Hi,

Just so that you have a reference, we usually simulate SimPoints of 100M 
instructions (with 10M extra warmup instructions) and on a single core 
O3+L1+L2 model they take between 30 minutes and 75 minutes.


Regards,

   Oscar


On 28/06/17 10:47, Stine, James wrote:

Hi All,

I am sorry to bother everyone.  I am trying to gauge performance and would love 
some feedback on run-time performance.  My main impetus for this Email is due 
to limited information I could find and just wanted to get some feedback on if 
there was some issues related to this topic.  I apologize in advance if I 
missed something specific about this question.

I did some tests on the queens benchmark as well as some others and my run 
times seem to take a long time.  16X grids within queens.c (e.g., queens 16) 
seem to run about 17 hours using AtomicMemory access with caching.  The 
ASPLOS-13 tutorial seems to have very small numCycles, so not sure that is 
accurate for “-o 16” on queens.c.  Eventually, I would love SPEC, but I am 
quite worried if queens.c takes forever, how can I even manage to get SPEC 
through.I also tried some other benchmarks like Matrix Multiplications, but 
some  of them take just as long.  However, queens does take a while to run, 
which I know is typical due to its intense computation mix.  My x86 cycle 
counts (statically compiled with -O3 and loop unrolling) were:  60,055,907,458 
on multi-core Intel extreme processors - again, I might have not run something 
correctly.

If anyone can possibly share their tips/tricks - especially for eventual 
running of SPEC, it would be great.  Does anyone do anything to maximize 
performance?  Even the smallest of tips would be helpful.  Perhaps, I am 
running gem5 with the wrong settings.  Or, perhaps, the settings are correct 
and this is a normal set of run times.  Anyways, I appreciate any help and also 
appreciate the wonder of gem5.  Take care.

All my best,

James
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users





smime.p7s
Description: S/MIME Cryptographic Signature
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] Performance

2017-06-28 Thread Oscar Rosell 

Hi,

I was not really suggesting using SimPoints (even that is a good option 
depending on your needs). As you say, to generate the SimPoints you 
first have to run the full thing (although you can do it with less 
detailed CPU which greatly speeds up things). I was just stating the 
typical throughput (simulated instructions / real time) we're getting 
from Gem5 with O3 CPU so that James can check if his performance is 
"normal".


I have never run perlbench but I have run many workloads taking several 
days to finish so 3 days doesn't look too bad to me.


Regards,

Oscar


On 28/06/17 15:55, Asif Ali Khan wrote:

Hi Oscar,

Yes, I agree with Simpoints its possible to simulate SPEC benchmarks 
but the problems is, how do you generate BBV file for the simpoint 
tool. I tried to generate this BBV file using Gem5 (for perlbench) and 
took like 3 days. Did I do something wrong?


Asif


On Wednesday, June 28, 2017, 3:01:43 PM GMT+2, Oscar Rosell 
<oscar.ros...@metempsy.com> wrote:



Hi,

Just so that you have a reference, we usually simulate SimPoints of 100M
instructions (with 10M extra warmup instructions) and on a single core
O3+L1+L2 model they take between 30 minutes and 75 minutes.

Regards,

Oscar


On 28/06/17 10:47, Stine, James wrote:
> Hi All,
>
> I am sorry to bother everyone.  I am trying to gauge performance and 
would love some feedback on run-time performance.  My main impetus for 
this Email is due to limited information I could find and just wanted 
to get some feedback on if there was some issues related to this 
topic.  I apologize in advance if I missed something specific about 
this question.

>
> I did some tests on the queens benchmark as well as some others and 
my run times seem to take a long time.  16X grids within queens.c 
(e.g., queens 16) seem to run about 17 hours using AtomicMemory access 
with caching.  The ASPLOS-13 tutorial seems to have very small 
numCycles, so not sure that is accurate for “-o 16” on queens.c.  
Eventually, I would love SPEC, but I am quite worried if queens.c 
takes forever, how can I even manage to get SPEC through.I also 
tried some other benchmarks like Matrix Multiplications, but some  of 
them take just as long.  However, queens does take a while to run, 
which I know is typical due to its intense computation mix.  My x86 
cycle counts (statically compiled with -O3 and loop unrolling) were:  
60,055,907,458 on multi-core Intel extreme processors - again, I might 
have not run something correctly.

>
> If anyone can possibly share their tips/tricks - especially for 
eventual running of SPEC, it would be great.  Does anyone do anything 
to maximize performance?  Even the smallest of tips would be helpful.  
Perhaps, I am running gem5 with the wrong settings.  Or, perhaps, the 
settings are correct and this is a normal set of run times.  Anyways, 
I appreciate any help and also appreciate the wonder of gem5.  Take care.

>
> All my best,
>
> James
> ___
> gem5-users mailing list
> gem5-users@gem5.org <mailto:gem5-users@gem5.org>
> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

___
gem5-users mailing list
gem5-users@gem5.org <mailto: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




smime.p7s
Description: S/MIME Cryptographic Signature
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] "Invalid magic string" error in GEM5

2017-06-02 Thread Oscar Rosell 
Hi,

I think that error is in the application you want to simulate, not in Gem5:


https://github.com/codekansas/tinier-nn/search?utf8=%E2%9C%93q=Invalid+magic+stringtype=


Regards,

Oscar Rosell - Metempsy


 On Fri, 02 Jun 2017 16:48:52 +0200 Raul Garcia 
raul.gar...@manchester.ac.uk wrote  

  Hello Gem5 Users,
 
 
 I am want to run a simulation with the following command line:
 
 command line: ./build/ARM/gem5.opt --debug-flags=Exec 
--stats-file=stats_cnn.txt --debug-file=trace_cnn.txt ./configs/example/se.py 
-I 1000 --cpu-type=arm_detailed --caches --l1d_size=32kB --l1i_size=32kB 
--l2cache --l2_size=256kB --sys-clock=650MHz --cpu-clock=650MHz -c 
../benchmarks/tinier-nn/eval/run_arm '--options= 
../benchmarks/tinier-nn/models/model.def'
 
 
 
 
 Global frequency set at 1 ticks per second
 warn: DRAM device capacity (8192 Mbytes) does not match the address range 
assigned (512 Mbytes)
 0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000
  REAL SIMULATION 
 info: Entering event queue @ 0.  Starting simulation...
 info: Increasing stack size by one page.
 
 Invalid magic string.Exiting @ tick 42761014 because target called exit()
 
 I couldn't find any reference to the Invalid magic string error. Any ideas of 
what is causing it?
 
 
 Raul.
 
 ___
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

Re: [gem5-users] BBV file not generated(or found)

2017-06-12 Thread Oscar Rosell 

Hi,

I don't see any evident problem. Have you tried to generate any BBV 
file? For example:


./build/ARM/gem5.opt configs/example/se.py -c 
./tests/test-progs/hello/bin/arm/linux/hello  --simpoint-profile 
--simpoint-interval 1 --fastmem


should work.

Regards,

   Oscar

On 09/06/17 19:07, Muzamil Rafique wrote:

Hi All,

I am trying to generate BBV file for SimPoint creation using the 
following command:


build/X86/gem5.opt configs/example/spec06_config.py --benchmark bzip2 
--benchmark_stdout=m5out/bzip2.out --benchmark_stderr=m5out/bzip2.err 
-I 1 --simpoint-profile --simpoint-interval 1 --fastmem


But at the end of simulation, I am not getting the BBV 
file(simpoint.bb.gz), neither in gem5 run directory nor in output 
directory.


Can anyone help me out with this issue? Am i doing something wrong?

Muzamil


___
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

Re: [gem5-users] Re：sim_seconds is a huge different between x86-se mode and x86- fs mode ,under the same program condition

2017-09-05 Thread Oscar Rosell 

Hi,

> Then, I change my rcS file followed you said ， I get a new stats,txt 
,the sim_seconds  is the actual simulation time for this program ,Am I 
right?


After you first changed your rcS file, the sim_seconds show the time 
from m5 resetstats until m5 exit. So yeah, should pretty approximately 
correspond with the time for arraywritenew.


> Next I changed rcS file again,it has two Simulation Statistics ,Is 
sim_seconds(  0.004455   ) the actual simulation time for this 
program? why different the sim_seconds   ( 0.004447 )


That's the simulation time for the program, right. Why is it different? 
Note first that the difference is pretty small. The difference is 
probably just due to the differences in the rcS file. Note that the 
calls to m5 are actual instructions that the simulator will execute so 
that will make some difference here and there. I wouldn't worry much 
about it. Just as a sanity check you can make arraywritenew to perform 
more/less work and see if the difference in sim_seconds between the two 
rcS is constant (which I'd expect) or linear.


Regards,

Oscar

On 02/09/17 08:25, 李莉 wrote:


Hi Oscar.

Thanks your rapid reply

First of all ,you are right ,the fs mode commend line edit 
error, because of my careless


Second , as you except ,the number of instructions executed is different.

Then, I change my rcS file followed you said ， I get a new stats,txt 
,the sim_seconds  is the actual simulation time for this program ,Am I 
right?


this  is my rcs file

#!/bin/sh
m5 resetstats
/arraywritenew
m5 exit

common line：build/X86/gem5.opt configs/example/fs.py 
--cpu-type=TimingSimpleCPU --caches --script=configs/boot/arraywrite.rcS


this is a part of my stats.txt

sim_seconds 0.004447   # Number of seconds simulated
sim_ticks 4447491000   # Number of ticks simulated
final_tick 5311293823000   # Number of ticks from 
beginning of simulation (restored from checkpoints and never reset)
sim_freq 1   # Frequency of simulated 
ticks


system.cpu.dtb.rdAccesses 305360   # TLB accesses 
on read requests
system.cpu.dtb.wrAccesses 227953   # TLB accesses 
on write requests
system.cpu.dtb.rdMisses 556   # TLB misses on read 
requests
system.cpu.dtb.wrMisses 175   # TLB misses on 
write requests


- End Simulation Statistics   --

Next I changed rcS file again,it has two Simulation Statistics ,Is 
sim_seconds(  0.004455   ) the actual simulation time for this 
program? why different the sim_seconds   ( 0.004447 )


this  is   my rcs file

m5 resetstats
/arraywritenew
m5 dumpstats
m5 exit

common line：build/X86/gem5.opt configs/example/fs.py 
--cpu-type=TimingSimpleCPU --caches --script=configs/boot/arraywrite.rcS


this is a part of  my stats.txt

-- Begin Simulation Statistics --
sim_seconds 0.004455   # Number of seconds simulated
sim_ticks 445488   # Number of ticks simulated
final_tick 531130128   # Number of ticks from 
beginning of simulation (restored from checkpoints and never reset)
sim_freq 1   # Frequency of simulated 
ticks


system.cpu.dtb.rdAccesses 306413   # TLB accesses 
on read requests
system.cpu.dtb.wrAccesses 228544   # TLB accesses 
on write requests
system.cpu.dtb.rdMisses 556   # TLB misses on read 
requests
system.cpu.dtb.wrMisses 175   # TLB misses on 
write requests


- End Simulation Statistics   --

-- Begin Simulation Statistics --
sim_seconds 0.006399   # Number of seconds simulated
sim_ticks 6399292000   # Number of ticks simulated
final_tick 5313245692000   # Number of ticks from 
beginning of simulation (restored from checkpoints and never reset)
sim_freq 1   # Frequency of simulated 
tickssystem.cpu.dtb.rdAccesses 430950   # TLB 
accesses on read requests
system.cpu.dtb.wrAccesses 323985   # TLB accesses 
on write requests
system.cpu.dtb.rdMisses 825   # TLB misses on read 
requests
system.cpu.dtb.wrMisses 249   # TLB misses on 
write requests


- End Simulation Statistics   --




___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] sim_seconds is a huge different between x86-se mode and x86- fs mode , under the same program condition

2017-09-01 Thread Oscar Rosell 

Hi,

First of all, those command lines look wrong. Both seem to use 
configs/example/se.py which is for "system emulation" mode.


Are those the only stats that differ? I would expect that the number of 
instructions executed would also differ. When running on full system 
mode an operating system has to be brought up. That's also simulated and 
will add up on the stats unless you reset them. Are you resetting the 
stats? In the rcS file you can add a call to m5 to clear the stats so it 
would look like:


/sbin/m5 resetstats

command_lines_to_execute_what_you_want_to_evaluate

/sbin/m5 exit

Regards,

Oscar


On 01/09/17 05:13, 李莉 wrote:


hello all,

  I emulate the same program in x86-se and x86-fs modes ,I have two 
puzzles  when I get the stats.txt


puzzle 1: the sim_seconds is huge different

puzzle 2：the TLB accesses is huge different

*here is my program*

#include 
void main(){
int a [10240];
int i=0;
int sum;
for(i=0;i<=10239;i++)
   {
 a[i]=i;

}
printf("%d\n",a[i-1]);

}

*The following is the simulation results*

*se model*  ：
commend line：build/X86/gem5.opt configs/example/se.py 
--cpu-type=TimingSimpleCPU --caches - c mountfile/arrraywritenew


stats.txt:

-- Begin Simulation Statistics --
sim_seconds 0.000515   # Number of seconds simulated
sim_ticks 515234000   # Number of ticks simulated

system.cpu.dtb.rdAccesses 63141   # TLB accesses 
on read requests
system.cpu.dtb.wrAccesses 32281   # TLB accesses 
on write requests
system.cpu.dtb.rdMisses 72   # TLB misses on read 
requests
system.cpu.dtb.wrMisses 24   # TLB misses on write 
requests
system.cpu.itb.rdAccesses 0   # TLB accesses on 
read requests
system.cpu.itb.wrAccesses 225198   # TLB accesses 
on write requests
system.cpu.itb.rdMisses 0   # TLB misses on read 
requests
system.cpu.itb.wrMisses 74   # TLB misses on write 
requests


*FS model*：
common line：build/X86/gem5.opt configs/example/se.py 
--cpu-type=TimingSimpleCPU --caches --script=configs/boot/arraywrite.rcS



stats.txt
- Begin Simulation Statistics --
sim_seconds 5.308574   # Number of seconds simulated
sim_ticks 5308574319000   # Number of ticks simulated
final_tick   5308574319000

 system.cpu.dtb.rdAccesses 16111843   # TLB 
accesses on read requests
system.cpu.dtb.wrAccesses 10662308   # TLB 
accesses on write requests
system.cpu.dtb.rdMisses 7951   # TLB misses on 
read requests

system.cpu.dtb.wrMisses 1181
  system.cpu.itb.rdAccesses 0   # TLB accesses on 
read requests
system.cpu.itb.wrAccesses 163647017   # TLB 
accesses on write requests
system.cpu.itb.rdMisses 0   # TLB misses on read 
requests

system.cpu.itb.wrMisses  3946

Thanks in advance

regards

lily



___
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

Re: [gem5-users] chroot failing on full system files

2018-05-09 Thread Oscar Rosell 

Hi,

Sorry, I'm not sure which exact image you're using at this point. First 
check that the binary exists inside the image. If you have mounted it 
you should be able to check it (you will need sudo, I think) just by 
doing "ls /mnt_point/bin/" and checking if bash is there. If it's 
present, do a file command on it "file /mnt_point/bin/bash" and send the 
output.


Regards,

    Oscar

On 09/05/18 16:24, Raman Arora wrote:

Hi Gabe,

Could you please suggest any measures i could take in order to debug 
the problem?
Any document or suggestion that could help me solve this issue would 
be helpful.


Raman

On Wed, 9 May 2018 at 01:13, Gabe Black > wrote:


The problem is that your system doesn't think it can run
/bin/bash. In the case of ARM, it's because the binary in the
image is for ARM, but the host computer is x86. In that case, you
need qemu so that your x86 computer can run the ARM /bin/bash from
the image. In your case, I'm assuming your host is x86, so it
should be able to run a /bin/bash from an x86 disk image with no
problem.

If, however, it can't run /bin/bash for some other reason, like if
the binary doesn't exist, your user doesn't have permission to run
it for some reason, etc., then it could fail with a similar error
message.

Gabe

On Tue, May 8, 2018 at 8:13 PM, Raman Arora
> wrote:

Hi All,

I am trying to run the full system files for X86 architecture
following the presentation


https://www.youtube.com/watch?v=Oh3NK12fnbg=LLpWtB08-0ZzqUaF98EZQqAA=4=300s

and Document


https://docs.google.com/document/d/1B7nZSqMLwkwoVNEj_58tMPTk4bKWvoEMbokOAjqeC-k/preview

But I am facing the following error,

*chroot: failed to run command ‘/bin/bash’: No such file or
directory*
**
*I have installed qemu,qemu-user,qemu-user-static,qemu-system
and the dependencies. I am using image file
*Linux-parsec-2-1-m5.img.* *

I have looked and the closest I have got to is a fix for ARM
architecture

https://www.mail-archive.com/gem5-users@gem5.org/msg15112.html

and and unresolved fix for X86

https://groups.google.com/forum/#!topic/openpiton/MdPPkRKWl9A


Could you please let me know how to fix the same for X86 as well?

Thanks and Regards,

Raman

___
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


___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] chroot failing on full system files

2018-05-09 Thread Oscar Rosell 

Hi,

Yeah, it seems that's the ALPHA image. Sorry, but I don't know where the 
X86 PARSEC image is, if it exists. Maybe somebody else has a link to it. 
In any case, I suppose you can use any base X86 image, chroot to it and 
download and install PARSEC there (you may probably need to resize the 
image though, if you use a preexisting image).


Regards,

    Oscar


On 09/05/18 17:11, Raman Arora wrote:

Hi Oscar,

Thank you for your prompt reply,I executed the commands that you 
asked. Here is the output.


ram@ram-VirtualBox:/mnt$ ls /bin/
*bash * cp fusermount login   
networkctl    ping6 stty    umount
bunzip2   cpio   getfacl loginctl nisdomainname 
plymouth su  uname
busybox   dash   grep    lowntfs-3g ntfs-3g   
ps sync    uncompress
bzcat date   gunzip  ls ntfs-3g.probe pwd 
systemctl   unicode_start
bzcmp dd gzexe   lsblk ntfs-3g.secaudit  rbash 
systemd vdir
bzdiff    df gzip    lsmod ntfs-3g.usermap   
readlink systemd-ask-password    wdctl
bzegrep   dir    hciconfig   mkdir ntfscat   red 
systemd-escape  which
bzexe dmesg  hostname    mknod ntfscluster   rm 
systemd-hwdb    whiptail
bzfgrep   dnsdomainname  ip  mktemp ntfscmp   
rmdir systemd-inhibit ypdomainname
bzgrep    domainname journalctl  more ntfsfallocate rnano 
systemd-machine-id-setup    zcat
bzip2 dumpkeys   kbd_mode    mount ntfsfix   
run-parts systemd-notify  zcmp
bzip2recover  echo   kill    mountpoint ntfsinfo  
sed systemd-tmpfiles    zdiff
bzless    ed kmod    mt ntfsls    setfacl 
systemd-tty-ask-password-agent  zegrep
bzmore    efibootmgr less    mt-gnu ntfsmove  
setfont tailf   zfgrep
cat   egrep  lessecho    mv ntfstruncate  setupcon 
tar zforce
chacl false  lessfile    nano ntfswipe  sh 
tempfile    zgrep
chgrp fgconsole  lesskey nc open  
sh.distrib touch   zless
chmod fgrep  lesspipe    nc.openbsd openvt    
sleep true    zmore
chown findmnt    ln  netcat pidof ss 
udevadm znew
chvt  fuser  loadkeys    netstat ping  
static-sh   ulockmgr_server

ram@ram-VirtualBox:/mnt$ file bin/bash
bin/bash: ELF 64-bit LSB executable, Alpha (unofficial), version 1 
(SYSV), dynamically linked, interpreter /lib/ld-linux.so.2, for 
GNU/Linux 2.4.3, stripped.


The bash executable is available it seems, as i have highlighted.

From the output, it seems that i have downloaded the wrong 
pre-compiled disk image for X86. Could you please let me know of any 
link I can download the correct one?


Thank You for helping me out. :)

Raman

On Wed, May 9, 2018 at 10:44 AM, Oscar Rosell 
<oscar.ros...@metempsy.com <mailto:oscar.ros...@metempsy.com>> wrote:


Hi,

Sorry, I'm not sure which exact image you're using at this point.
First check that the binary exists inside the image. If you have
mounted it you should be able to check it (you will need sudo, I
think) just by doing "ls /mnt_point/bin/" and checking if bash is
there. If it's present, do a file command on it "file
/mnt_point/bin/bash" and send the output.

Regards,

    Oscar

On 09/05/18 16:24, Raman Arora wrote:

Hi Gabe,

Could you please suggest any measures i could take in order to
debug the problem?
Any document or suggestion that could help me solve this issue
would be helpful.

Raman

On Wed, 9 May 2018 at 01:13, Gabe Black <gabebl...@google.com
<mailto:gabebl...@google.com>> wrote:

The problem is that your system doesn't think it can run
/bin/bash. In the case of ARM, it's because the binary in the
image is for ARM, but the host computer is x86. In that case,
you need qemu so that your x86 computer can run the ARM
/bin/bash from the image. In your case, I'm assuming your
host is x86, so it should be able to run a /bin/bash from an
x86 disk image with no problem.

If, however, it can't run /bin/bash for some other reason,
like if the binary doesn't exist, your user doesn't have
permission to run it for some reason, etc., then it could
fail with a similar error message.

Gabe

On Tue, May 8, 2018 at 8:13 PM, Raman Arora
<mail.arorara...@gmail.com

Re: [gem5-users] chroot failing on full system files

2018-05-10 Thread Oscar Rosell 

Hi,

Not sure what X86 image you're using. The apt-get command is present 
only in Linux distributions from the Debian family (Ubuntu and Debian 
are the most popular). If it doesn't exist it most probably means that 
the image you are using is not one of those (Fedora/CentOS use yum).


As an alternative, note that there is people that already tried to get a 
Gem5/PARSEC/X86 image. A Google search got me to this (kinda old) GitHub:


https://github.com/anthonygego/gem5-parsec3

So you can try to follow the instructions above or contact the author 
(which I am not). IMPORTANT WARNING: I haven't checked what the code in 
the GitHub above does. Running random code from the Internet may end up 
erasing your disk, compromising your system or even worse!


Regards,

    Oscar

On 09/05/18 21:48, Raman Arora wrote:

Hi Oscar,

Thanks a ton for your help, I tried downloading an x86 image and the 
chroot worked on it :D.


But after i did apt-get update i got a message bash: apt-get command 
not found. Any solutions to this?


I already executed the command echo "nameserver 8.8.8.8" > 
/etc/resolv.conf once i successfully did a chroot on my mount point .


Thanks and Regards,

Raman

On Wed, May 9, 2018 at 11:25 AM, Oscar Rosell 
<oscar.ros...@metempsy.com <mailto:oscar.ros...@metempsy.com>> wrote:


Hi,

Yeah, it seems that's the ALPHA image. Sorry, but I don't know
where the X86 PARSEC image is, if it exists. Maybe somebody else
has a link to it. In any case, I suppose you can use any base X86
image, chroot to it and download and install PARSEC there (you may
probably need to resize the image though, if you use a preexisting
image).

Regards,

    Oscar


On 09/05/18 17:11, Raman Arora wrote:

Hi Oscar,

Thank you for your prompt reply,I executed the commands that you
asked. Here is the output.

ram@ram-VirtualBox:/mnt$ ls /bin/
*bash * cp fusermount  login   networkctl ping6  
stty umount
bunzip2   cpio   getfacl loginctl   
nisdomainname plymouth su   uname
busybox   dash   grep lowntfs-3g  ntfs-3g  
ps sync uncompress
bzcat date   gunzip ls  ntfs-3g.probe
pwd systemctl unicode_start
bzcmp dd gzexe lsblk   ntfs-3g.secaudit 
rbash systemd   vdir
bzdiff    df gzip lsmod   ntfs-3g.usermap  
readlink systemd-ask-password   wdctl
bzegrep   dir    hciconfig mkdir  
ntfscat   red systemd-escape   which
bzexe dmesg  hostname mknod  
ntfscluster   rm systemd-hwdb whiptail
bzfgrep   dnsdomainname  ip mktemp  ntfscmp  
rmdir systemd-inhibit ypdomainname
bzgrep    domainname journalctl more   
ntfsfallocate rnano systemd-machine-id-setup   zcat
bzip2 dumpkeys   kbd_mode mount  
ntfsfix   run-parts systemd-notify   zcmp
bzip2recover  echo   kill mountpoint  ntfsinfo 
sed systemd-tmpfiles   zdiff
bzless    ed kmod mt  ntfsls   
setfacl systemd-tty-ask-password-agent  zegrep
bzmore    efibootmgr less mt-gnu  ntfsmove 
setfont tailf   zfgrep
cat   egrep  lessecho mv 
ntfstruncate  setupcon tar   zforce
chacl false  lessfile nano   
ntfswipe  sh tempfile   zgrep
chgrp fgconsole  lesskey nc 
open  sh.distrib touch   zless
chmod fgrep  lesspipe nc.openbsd 
openvt    sleep true   zmore
chown findmnt    ln netcat  pidof ss
udevadm   znew
chvt  fuser  loadkeys netstat
ping  static-sh ulockmgr_server
ram@ram-VirtualBox:/mnt$ file bin/bash
bin/bash: ELF 64-bit LSB executable, Alpha (unofficial), version
1 (SYSV), dynamically linked, interpreter /lib/ld-linux.so.2, for
GNU/Linux 2.4.3, stripped.

The bash executable is available it seems, as i have highlighted.

From the output, it seems that i have downloaded the wrong
pre-compiled disk image for X86. Could you please let me know of
any link I can download the correct one?

Thank You for helping me out. :)

Raman

On Wed, May 9, 2018 at 10:44 AM, Oscar Rosell
<oscar.ros...@metempsy.com <mailto:oscar.ros...@metempsy.com>> wrote:

Hi,

Sorry, I'm not sure which exact image you're using at this
point. First check that the binary exists inside the image.
If you have mounted it you should be able to check it (you
will need sudo, I think) just by doing "ls /mnt_point/bin/"
and checking if bash is there. If it's present, do a file
command on it "file /mnt_p

Re: [gem5-users] Getting a C++ program (with tensorflow) to work.

2018-06-11 Thread Oscar Rosell 

Hi,

I'd expect libloader.so file to be a library and that's suspicious... 
Besides that, I have had much more success running complex workloads on 
FS mode than in SE mode (however, I've worked mostly with ARM 
architecture). I'd advise you try that.


Regards,

    Oscar

On 11/06/18 11:09, Matthew D'Alonzo wrote:

Hello, everyone.

I’ve been working on a job where I need to get a C++ program to run 
through the simulator, and so far I’ve had no luck. I compiled the 
program statically, so all the necessary libraries would be included.

The following is the command that I use to run the program:
build/X86/gem5.opt configs/example/se.py -n 4 --cpu-type DerivO3CPU 
--caches --l1d_size=64kB --l1i_size=64kB --l2_size=256kB 
--cacheline_size=64 --l2cache -c 
/Private/tensortwo/tensorflow/bazel-bin/tensorflow/loader/libloader.so


Here is the output of gem5 when I try to run it:

Global frequency set at 1 ticks per second
warn: DRAM device capacity (8192 Mbytes) does not match the address 
range assigned (512 Mbytes)

0: system.remote_gdb: listening for remote gdb on port 7000
0: system.remote_gdb: listening for remote gdb on port 7001
0: system.remote_gdb: listening for remote gdb on port 7002
0: system.remote_gdb: listening for remote gdb on port 7003
 REAL SIMULATION 
info: Entering event queue @ 0.  Starting simulation...
warn: ignoring syscall set_robust_list(...)
warn: ignoring syscall rt_sigaction(...)
(further warnings will be suppressed)
warn: ignoring syscall rt_sigprocmask(...)
(further warnings will be suppressed)
gem5 has encountered a segmentation fault!

--- BEGIN LIBC BACKTRACE ---
build/X86/gem5.opt(_Z15print_backtracev+0x15)[0x132f0a5]
build/X86/gem5.opt[0x1343a3d]
/lib64/libpthread.so.0(+0xf680)[0x7fb24d13e680]
/opt/crc/g/gcc/7.1.0/lib64/libgcc_s.so.1(_Unwind_Resume+0x42)[0x7fb24be545f2]
build/X86/gem5.opt(_ZN6X86ISA7Decoder10decodeInstENS_11ExtMachInstE+0x352c4)[0xc4cb44]
build/X86/gem5.opt(_ZN6X86ISA7Decoder6decodeENS_11ExtMachInstEm+0x234)[0xbdeb14]
build/X86/gem5.opt(_ZN6X86ISA7Decoder6decodeERNS_7PCStateE+0x214)[0xbdee04]
build/X86/gem5.opt(_ZN12DefaultFetchI9O3CPUImplE5fetchERb+0x8d3)[0xb745b3]
build/X86/gem5.opt(_ZN12DefaultFetchI9O3CPUImplE4tickEv+0xbc)[0xb754fc]
build/X86/gem5.opt(_ZN9FullO3CPUI9O3CPUImplE4tickEv+0x110)[0xb55580]
build/X86/gem5.opt(_ZN10EventQueue10serviceOneEv+0xd5)[0x13365a5]
build/X86/gem5.opt(_Z9doSimLoopP10EventQueue+0x6b)[0x135075b]
build/X86/gem5.opt(_Z8simulatem+0xc3e)[0x13516de]
build/X86/gem5.opt[0x13fc997]
build/X86/gem5.opt[0x7ea5a7]
/lib64/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x730a)[0x7fb24d43020a]
/lib64/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x7ed)[0x7fb24d43203d]
/lib64/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x663c)[0x7fb24d42f53c]
/lib64/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x7ed)[0x7fb24d43203d]
/lib64/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x663c)[0x7fb24d42f53c]
/lib64/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x7ed)[0x7fb24d43203d]
/lib64/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x663c)[0x7fb24d42f53c]
/lib64/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x7ed)[0x7fb24d43203d]
/lib64/libpython2.7.so.1.0(PyEval_EvalCode+0x32)[0x7fb24d432142]
/lib64/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x5513)[0x7fb24d42e413]
/lib64/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x7ed)[0x7fb24d43203d]
/lib64/libpython2.7.so.1.0(PyEval_EvalFrameEx+0x663c)[0x7fb24d42f53c]
/lib64/libpython2.7.so.1.0(PyEval_EvalCodeEx+0x7ed)[0x7fb24d43203d]
/lib64/libpython2.7.so.1.0(PyEval_EvalCode+0x32)[0x7fb24d432142]
/lib64/libpython2.7.so.1.0(+0x10057f)[0x7fb24d44b57f]
/lib64/libpython2.7.so.1.0(PyRun_StringFlags+0x65)[0x7fb24d44c3e5]
build/X86/gem5.opt(_Z6m5MainiPPc+0x53)[0x13429e3]
--- END LIBC BACKTRACE ---
Segmentation fault


I’m not sure if this is a lost cause or not, but if there’s any help I 
could get given this information, that would be great.


Thanks!
Matthew


___
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

Re: [gem5-users] execution problem hello.c in architecture armv8 big.LITTLE

2018-06-01 Thread Oscar Rosell 

Hi,

The important line is this one:

/ tmp / script: line 3: cd: / home / root / parsec: No such file or
directory

No way you can get the binary running if you cannot cd to the directory 
where the binary is located. That has no relation with how the binary 
was compiled.


My intuition is that the directory is in fact "/root/parsec". It's just 
a guess based on the absence of home directory for root user in my 
Ubuntu (although I don't know which image you're using). To be sure just 
add an ls to show the directory at each point like this:


PARSEC_DIR = "/ home / root / parsec"
ls
cd /home/
ls
cd root
ls
cd parse
ls

On 01/06/18 08:45, Ciro Santilli wrote:

This is not required, the best thing to do is to find what is the
proper compatible cross compiler for the image:
https://stackoverflow.com/questions/31929092/trying-to-run-a-cross-compiled-executable-on-target-device-fails-with-no-such-f/49993116#49993116

On Fri, Jun 1, 2018 at 6:34 AM, Tung Hoang  wrote:

You may need “-static” for arm cross-compiling.

/T

On Thu, May 31, 2018 at 3:19 PM commerce _com 
wrote:

Hi ciro,

I'm sure the binary of the compilation  is in the image on the path: /
home / root / parsec

but I did not understand how to solve this problem; Please.


com_.

2018-05-31 23:02 GMT+02:00 Ciro Santilli :

Likely incompatible compiler using wrong dynamic loader, do "file
hello", see "interpreter /some/path", and check if "/some/path" is
present on guest.

On Thu, May 31, 2018 at 9:33 PM, commerce _com 
wrote:

Hi all,

i need to run hello.c in an architecture armv8 big.LITTLE
I compile hello_word.c by a crosscompiler here is the command:
arm-linux-gnueabihf-gcc hello.c -o hello

I added the binarie of the compilation to the linaro aarch64 image,
with I
generated this .rcs script as follows:
#! / Bin / bash
PARSEC_DIR = "/ home / root / parsec"
cd $ PARSEC_DIR
pwd
./Hello

I typed the following command:

sudo build / ARM / gem5.opt configs / example / arm / fs_bigLITTLE.py
--kernel = / media / ali / ali / gem5-master / aarch-system-20180409 /
binaries / vmlinux.vexpress_gem5_v1_64 --dtb = / media / ali / ali /
gem5-master / aarch-system-20180409 / binaries /
armv8_gem5_v1_big_little_2_2.dtb --bootscript = / home / ali / desktop
/
rcs_file / parsec_rcs / hello.rcS - caches

the execution of the kernel it works normal but when it passes the
stage of
execution of the hello I found this result:

/ tmp / script: line 3: cd: / home / root / parsec: No such file or
directory
/
/ tmp / script: line 5: ./hello: No such file or directory

despite I added the binary hello to the linaro image.

please if you have an idea to solve the problem.

here is my file system.terminal:

thanks advance.

com_.


___
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

___
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

Re: [gem5-users] Compilation error

2018-06-05 Thread Oscar Rosell 

Hi,

Most (if not all) of the errors are just a consequence of the first one 
so they don't really matter. Most probably, you're missing a ";" 
somewhere before the "namespace py = pybind11;" line.


Regards,

    Oscar


On 04/06/18 22:44, Mao Ye wrote:

Hi all,

I have been using gem5 for a while without any problem. But today I 
met a problem with my new modification and it failed at compilation 
stage. However, I diff the following file with the one under another 
directory of mine, showing no difference at all. So what might be the 
reason causing so many errors? Any suggestion will be appreciated.


The error message is as below :

[VER TAGS]  -> X86/sim/tags.cc
[ CXX] X86/python/_m5/param_DRAMCtrl.cc -> .o
build/X86/python/_m5/param_DRAMCtrl.cc:53:1: error: expected 
unqualified-id before 'namespace'

 namespace py = pybind11;
 ^
build/X86/python/_m5/param_DRAMCtrl.cc:56:13: error: 'py' has not been 
declared

 module_init(py::module _internal)
 ^
build/X86/python/_m5/param_DRAMCtrl.cc:56:24: error: expected ',' or 
'...' before '&' token

 module_init(py::module _internal)
    ^
build/X86/python/_m5/param_DRAMCtrl.cc:137:33: error: expected 
identifier before string constant
 static EmbeddedPyBind embed_obj("DRAMCtrl", module_init, 
"AbstractMemory");

   ^
build/X86/python/_m5/param_DRAMCtrl.cc:137:33: error: expected ',' or 
'...' before string constant
build/X86/python/_m5/param_DRAMCtrl.cc:137:75: error: expected '}' at 
end of input
 static EmbeddedPyBind embed_obj("DRAMCtrl", module_init, 
"AbstractMemory");

^
build/X86/python/_m5/param_DRAMCtrl.cc: In static member function 
'static void FECache::module_init(int)':
build/X86/python/_m5/param_DRAMCtrl.cc:58:5: error: 'py' has not been 
declared

 py::module m = m_internal.def_submodule("param_DRAMCtrl");
 ^
build/X86/python/_m5/param_DRAMCtrl.cc:59:5: error: 'py' has not been 
declared
 py::class_std::unique_ptr>(m, "

 ^
build/X86/python/_m5/param_DRAMCtrl.cc:59:30: error: expected 
primary-expression before ',' token
 py::class_std::unique_ptr>(m, "

  ^
build/X86/python/_m5/param_DRAMCtrl.cc:59:52: error: expected 
primary-expression before ',' token
 py::class_std::unique_ptr>(m, "

  ^
build/X86/python/_m5/param_DRAMCtrl.cc:59:86: error: 'py' was not 
declared in this scope
 py::class_std::unique_ptr>(m, "

^
build/X86/python/_m5/param_DRAMCtrl.cc:59:98: error: template argument 
2 is invalid
 py::class_std::unique_ptr>(m, "DRAM

^
build/X86/python/_m5/param_DRAMCtrl.cc:59:101: error: 'm' was not 
declared in this scope
 :class_std::unique_ptr>(m, "DRAMCtr

^
build/X86/python/_m5/param_DRAMCtrl.cc:60:14: error: 'py' is not a 
class, namespace, or enumeration

 .def(py::init<>())
  ^
build/X86/python/_m5/param_DRAMCtrl.cc:60:23: error: expected 
primary-expression before '>' token

 .def(py::init<>())
   ^
build/X86/python/_m5/param_DRAMCtrl.cc:60:25: error: expected 
primary-expression before ')' token

 .def(py::init<>())
 ^
build/X86/python/_m5/param_DRAMCtrl.cc:132:5: error: 'py' is not a 
class, namespace, or enumeration
 py::class_py::nodelete>>(m, "DRAMCtrl")

 ^
build/X86/python/_m5/param_DRAMCtrl.cc:132:24: error: expected 
primary-expression before ',' token
 py::class_py::nodelete>>(m, "DRAMCtrl")

    ^
build/X86/python/_m5/param_DRAMCtrl.cc:132:40: error: expected 
primary-expression before ',' token
 py::class_py::nodelete>>(m, "DRAMCtrl")

  ^
build/X86/python/_m5/param_DRAMCtrl.cc:132:68: error: the value of 
'py' is not usable in a constant expression
 py::class_py::nodelete>>(m, "DRAMCtrl")

^
build/X86/python/_m5/param_DRAMCtrl.cc:59:86: note: 'py' was not 
declared 'constexpr'
 py::class_std::unique_ptr>(m, "

^
build/X86/python/_m5/param_DRAMCtrl.cc:132:80: error: type/value 
mismatch at argument 2 in template parameter list for 'template_Tp, class _Dp> class std::unique_ptr'
 py::class_py::nodelete>>(m, "DRAMCtrl")

^
build/X86/python/_m5/param_DRAMCtrl.cc:132:80: note: expected a type, 
got 'py'

build/X86/python/_m5/param_DRAMCtrl.cc: At global scope:
build/X86/python/_m5/param_DRAMCtrl.cc:135:1: error: expected 
unqualified-id at end of input

 };
 ^
scons: *** [build/X86/python/_m5/param_DRAMCtrl.o] Error 1
scons: building terminated because of errors.


--
Mao Ye,


___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users




smime.p7s
Description: S/MIME Cryptographic Signature
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] syscall error while simulating

2018-07-17 Thread Oscar Rosell 

Hi,

There's nothing magical about it. Those syscalls are probably just not 
implemented. Depending on the Gem5 version you're using they may have 
been implemented later (didn't check). Running in Full System mode 
instead should work.


Regards,

    Oscar


On 17/07/18 06:22, Puneet Saraf wrote:

Dear All,

I am getting a syscall unimplemented error while running the 
simulation. For some syscall it gives warning and proceed with the 
simulation but for some it stopped there itself. Below are the syscall 
for which my simulation stopped.


fadvise64
timer_create
timer_settime
timer_gettime
timer_getoverrun
timer_delete,
clock_settime
clock_gettime
clock_getres
clock_nanosleep

These error are getting generated from *src/arch/x86/linux/process.cc 
*


 /* 221 */ SyscallDesc("fadvise64", unimplementedFunc),
/* 222 */ SyscallDesc("timer_create", unimplementedFunc),
/* 223 */ SyscallDesc("timer_settime", unimplementedFunc),
/* 224 */ SyscallDesc("timer_gettime", unimplementedFunc),
/* 225 */ SyscallDesc("timer_getoverrun", unimplementedFunc),
/* 226 */ SyscallDesc("timer_delete", unimplementedFunc),
/* 227 */ SyscallDesc("clock_settime", unimplementedFunc),
/* 228 */ SyscallDesc("clock_gettime", clock_gettimeFunc),
/* 229 */ SyscallDesc("clock_getres", clock_getresFunc),
/* 230 */ SyscallDesc("clock_nanosleep", unimplementedFunc),


I don’t know what is causing these errors and how to debug and rectify 
them.

Please help if anybody has any idea about these errors.




Regards
Puneet Saraf
M.S. Scholar
PACE Lab, Dept. Of Computer Science & Engg.
IIT Madras, Chennai-600036


___
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