Kiyeon,
I am not sure. I could not look into it due to time constraints. But
it seemed to happen when I recovered from a checkpoint. Also it did
not happen with the atomic core. I ran into this with the detailed
and arm_detailed. If you figure it out let me know.
Thanks,
Jason
Quoting Kiyeon Lee <[email protected]>:
Hi, Jason.
Thanks for your kind help.
I did not try running a multicore simulation yet, but now I can run a
single-core simulation using the EMM version.
Do you have any guess why you are seeing the issue in multicore simulation
using the EMM version?
Thanks.
Kiyeon
On Sun, Apr 15, 2012 at 2:42 PM, Jason Clemons <[email protected]> wrote:
Hey,
Sorry for jumping in midway but I have used a Vexpress machine type for
Gem5 but only the EMM version. I have placed a copy of the kernel (3.2) I
used here:
http://www.eecs.umich.edu/**mevbench/downloads.html<http://www.eecs.umich.edu/mevbench/downloads.html>
The only issue I had was in multicore simulations it would not restore
from checkpoints without getting an error about non cacheable loads.
Single core I had no problems.
Hope this helps.
Thanks,
Jason
Quoting Andrew Cebulski <[email protected]>:
I haven't used VExpress_ELT before, however, I'm testing it out now with
the pre-compiled kernel provided by gem5. Have you tried that in order to
rule out its being a kernel issue? I've rarely seen any chat on the
mailing
lists regarding the VExpress machine types, so I expect they haven't been
debugged/tested much other than by the developers.
-Andrew
On Thu, Apr 12, 2012 at 1:17 PM, Kiyeon Lee <[email protected]> wrote:
Hello, everyone.
I am trying to run a full-system simulation using the ARM "VExpress_ELT"
platform modeled in gem5. I am using the latest source codes from the
gem5 repository.
I compiled linux kernel 2.6.38 to target the ARM Versatile Express
platform.
I used the corresponding config file and the linux kernel source code
patch from the arm website (www.arm.com) to compile the linux kernel.
I am using the "fs.py" script file as it is, except that I changed the
machine type to VExpress_ELT and changed the main memory size to 1GB. I
am also using the linux kernel that I compiled instead of using the
pre-compiled kernel provided by the gem5 website.
However, it seems like the simulation is not proceeding after gem5 loads
the kernel.
I am wondering whether the problem is caused by my linux kernel or by
gem5.
My question is should I somehow change the config file and recompile the
linux kernel or should I debug gem5?
Are the two ARM Versatile Express platforms (VExpress_ELT and
VExpress_EMM) currently stable to use for full-system simulation?
Any hints/comments will be greatly appreciated.
Thanks in advance.
Kiyeon
______________________________**_________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/**mailman/listinfo/gem5-users<http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users>
______________________________**_________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/**mailman/listinfo/gem5-users<http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users>
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
