@fangfei Lu
can u tell me any site for downloading benchmarks and also
how to run on the simulator. Actually I have simulated ARM architecture but
now my next step is to run a benchmark. Kindly help me out.
Thanks in advance...
On Thu, Sep 27, 2012 at 9:00 AM, wrote:
> Send gem5
I've never seen this problem before, and the BBench source for eBay does
not show anything called ewbay in the source, so I don't know how it's
getting redirected to that URL. Did you modify the BBench source in any
way? Can you send me your exact command line? And, can you tell me the
exact URL th
Hi,
I wanted to experiment with the number of slots each resource has been
assigned.
However, when I look into the code of pipeline_stage.cc, I see that
slots are being freed before instructions from the same stage are being
processed. Doesn't this contradict the usage of the slots?
In a fir
Hi all,
I am having a problem with a setup that used to work. The important flag
is --garnet-network, without which everything works just fine.
Versions from before Sep 05 in the dev repo were working, unfortunately
I don't have the exact commit when it stopped working; if nobody has an
idea what
I started a BBench run this morning with a fresh check out of the latest
dev repo, the Android kernel and an unmodified Gingerbread disk image with
BBench, both from gem5.org/bbench-gem5. I do not see this problem, eBay
renders fine and the simulation continues. I will need more info on your
setup/
Hi all,
Running a benchmark on gem5, I have obtained the number of
busy and idle cycles at each tick. Using an offline algorithm, I have applied
windowing on this data
set and have calculated the number of busy cycles in each window of fixed
size (e.g. 20 busy cycles in widow size of 100 cycl
Hi Arthur:
interesting, I don't recall seeing this behavior in my experiments (it
seems to happen very soon). Can you share a little bit more about your
setup, so I could try to reproduce this issue? Are you running an
unmodified gem5 model?
Thanks,
-Andrea
On Thu, Sep 27, 2012 at 12:11 PM, Arth
Hi Mike,
I made some modifications myself to support SMT on O3 on X86.
After applying this patch (http://reviews.gem5.org/r/1281/#review2999) I
had no issued running SMT in SE mode -- at least in my tests.
-Andrea
On Thu, Sep 27, 2012 at 10:02 AM, Yuval H. Nacson <
yhnac...@tx.technion.ac.il> wrot
Hi Andrea,
At first I noticed this issue on a modified model. However, as I said
the issue comes from the way branching is detected in IEW by calling the
PCState.branching() function (at least I understand it that way), which
is unmodified in my model. Furthermore, I just tried on the latest
Hi Max,
The number of slots each resource is being given is in resource_pool.cc in
the constructor. EAch resource is defined and assigned a user-defined
number of slots.
Slots are only freed in a resource after an instruction makes a request to
that resources, the resource accepts that request, an
Hi,
I am fairly new to gem5 as well but I managed to get this done by
reading through this thread
(http://www.digipedia.pl/usenet/thread/11778/1757/). You'll have to
change the se.py script to run different workloads in different cpu. I
am not sure what you mean by mesh technology though.
Hi Arthur,
let me download the latest version of gem5 and try it.
Do you run 401.bzip2 without arguments?
In my runs I usually invoke it w/ some command line arguments such as:
401.bzip2:
- test: ./bzip2 input.program 5 > input.program.out 2>>
input.program.err; ./bzip2 dryer.jpg 2 > dryer
Hi Arthur,
I should have read your email w/ more attention, I understand that the
problem shows up even w/o arguments.
Would it be possible for you to share the benchmark where you observe the
misbehavior, with my version of bzip I don't seem to incur in similar
instructions -- not as soon as you
On Fri, 28 Sep 2012, Marco Elver wrote:
Hi all,
I am having a problem with a setup that used to work. The important flag
is --garnet-network, without which everything works just fine.
Versions from before Sep 05 in the dev repo were working, unfortunately
I don't have the exact commit when it
Hi All,
Actually the main issue comes from the fact that the mispredicted()
function detect a misprediction by comparing the PredPC and the NextPC
in terms of PCstate objects. Thus the comparison involve both the PC and
the uPC and it is able to detect the branch misprediction inside a
macroo
Hi Jun,
1. Recompiled kernel x86_64-vmlinux-2.6.28.4.smp with SMP enabled and
> CONFIG_NR_CPUS=512.
>
Ok. If my memory serves me correctly, you will need to also change the
CONFIG_NR_CPUS range in the kconfigs for kernel 2.6.28.4. I don't recall
the details on where to find these files, but the
On Fri, 28 Sep 2012, yasar k wrote:
helo all..
am trying to impliment memory level parallelism (mlp) aware cache
replacment policy..need some guidence..how to change cache rep[lacment
policy ..anythng related plz share
Two things --
1. This mail should have been sent to gem5-users list.
2. Yo
Hi zpwu, Tianyi
I am also new on GEm5, My work is also related with Multicore, One of my lab
fellow already worked with GEM5 and and build a Triplet Architecture which is
related with interUnit Now I have to modify and interact between the Objects on
multicore. Core are fashioned in TBHIN Form.
18 matches
Mail list logo
