Dear Ferran,
Thank you so much for all your insights.
I managed to re-compile the benchmark after chroot to the image.
I also solved the external libraries issues and managed to run it on both Gem5
original and the other version of Gem5.
Thanks so much!
Regards,
Elena
-Original
Hi Ferran,
I have never done chroot before, so this was my first attempt. What I did was:
a) Install schroot and debootstrap
b) Creating a chroot at my mounted image
c) Edit the schroot.conf
d) Created basic installation of Ubuntu 14.04 in the chroot
But I encountered errors while trying to
Hi Pierre-Yves,
I did tried compiling it with the -static flag for the first few benchmarks
that I have run, but I got a segmentation fault. So, I tried compiling it
without the -static flag, and it works.
But for this benchmark (QSort and Dijkstra) from MiBench, I tried to compile
without
Hi there,
I run a couple of benchmarks from MiBench Automotive package under FS Mode
successfully.
I wanted to test other benchmarks from the same MiBench, so I compiled those C
programs as usual using GCC (I have GCC 4.8.4 in my Ubuntu). No errors in
compiling and even running it from the
Hi Ferran,
I did as per your suggestions.
Initially, I was still facing with errors /bin/bash: /lib32/libc.so.6: version
'GLIBC_2.11' not found as I am using the image that comes with GemFI (Gem5 with
Fault Injection).
I redo the steps with the image downloaded from official gem5 web and I
By the way,
The original benchmark which compiled successfully, when run with Gem5 in FS
mode, the program was aborted. The error message received was:
Build/XC86/arch/xc86/tlb.cc.351: Fault XC86ISA::TLB::translate(RequestPtr,
ThreadContext*, BaseTLB::Translation*, BaseTLB::Mode, bool&, bool):
Hi Wei Shu,
The issue with Glibc was also something I experienced before. The benchmark was
able to run fine outside Gem5, but in Gem5 it produces that error. Someone
thought me to recompile the benchmark in the image kernel loaded by Gem5. That
step involves chroot to the image kernel and
Hi Wei Shu,
I had similar problems when I tried running benchmarks from MiBench in FS mode
using X86 architecture. The problem I think was because all the benchmarks were
compiled statically. Once I recompiled those benchmarks dynamically, it solved
the segmentation fault problem.
Perhaps
Hi Ciro,
It is the same issue, but there was no answer so I decided to ask a subset
of the question again in a different way.
I am interested in the CPU stats - system.cpu.committedInsts,
system.cpu.dcache.overall_misses, system.cpu.icache.overall_misses
These stats create a different profile
Yes, that was the intention - to access the host file from a Linux guest.
I will study the Workload Automation more in depth as well as look into m5
readfile.
Another way which I was thinking if it is possible to just get the required
stats printed on the terminal, and send those information
Hi all,
I am trying to set up a multicore environment where I have 1 core monitoring
the stats file of another core. In order to do that, the monitoring core needs
to access the stats file during simulation in FS mode. I have tried to open and
read the stats file in the usual way (using a file
Hello there,
Problem solved. I did not set Cpu-Clock and Caches when I checkpoint the
system, hence that’s why it throw the checkpoint in array.
Now I can successfully load the new script file and got the output needed.
Thanks.
Regards,
Elena
From: gem5-users
Thanks for the steps outlined. I have no issue with Step 1, because that’s
exactly what I did. I just have problem with Step 2 because although it can
restore from the checkpoint, it can’t read the new script file.
Best regards,
Elena
From: TungHoang_GMail
mailto:hoangthanht...@gmail.com>>
Hi,
I don’t think the script has any problem because as you can see from the result
below, without any checkpointing, the program was able to execute till the end
after booting. So, I am pretty sure the “bash” is being supported and the
script works fine. I just couldn’t get it to load after
Hi all,
I would like to simulate 2 processors - the idea is to have a main processor,
which is an x86 gem5 running a benchmark and another processor monitoring the
main processor.
My question is, is it possible to have two copies of gem5 running on the same
Ubuntu system or do I need to use
Hi all,
I would like to simulate 2 processors - the idea is to have a main processor,
which is an x86 gem5 running a benchmark and another processor monitoring the
main processor.
My question is, is it possible to have two copies of gem5 running on the same
Ubuntu system or do I need to use
Hi Jason,
I still don’t quite understand you. Please correct me if I am wrong - I assumed
the host you meant the Ubuntu VM that I have installed. If I have one instance
of gem5 running, how to I get another instance of gem5 running? I am trying to
simulate a processor watching another
Hi,
Looks like I am able to run 2 copies of gem5 after all in the same host.
Now I just need to configure my ‘2nd’ copy of gem5 for monitoring.
Thank you.
Thanks and Regards,
Elena Woo Lai Leng
Faculty of Electronic and Computer Science (ECS)
University of Southampton
Email:
18 matches
Mail list logo