Hi all,
I am using the latest version of gem5-stable. I want to restore a checkpoint
with ruby and detailed cores. To take and restore the checkpoint I’ve used
MESI_Two_Level protocol. I get the following error when a checkpoint is
restored.
fatal: Fatal Error:: No Route exists from this Router
Hello all,I do per-core DVFS with detailed (ALPHA) core and ruby infull system
mode. I want to have the same clock frequency for a core and itsattached L1
cache without changing clock of L2 cache. The commandline option --ruby-clock
does not work for me because it sets the same clock forL1 and L
Hi all,
I run simulations on full system mode with detailed CPUs and ruby
(MESI_CMP_directory protocol). In my simulations, L1 cache is private and
L2cache is shared among the cores. As ruby simulates memory subsystem
(includingL1 and L2 caches), I wonder how setting the clock for ruby affects
Runjie,
Thank you again for the further direction.
Regards,
Shervin
--- On Tue, 10/30/12, Runjie Zhang wrote:
From: Runjie Zhang
Subject: Re: [gem5-users] dumping periodic stats with checkpoints
To: "shervin hajiamini"
Cc: gem5-users@gem5.org
Date: Tuesday, October 30, 2012, 4:5
Hi all,
For getting the power data for a core, I have fed McPAT tool
with stats.txt file. The output I get from McPat shows that the power
consumption (which is known as Runtime Dynamic) per core is very large whereas
I want the power consumption (per core) to be less than 1 Watt.
Does anybod
. You pass a
relative time to the function, not an absolute time.
Thanks,
Ali
On Oct 27, 2012, at 11:29 PM, shervin hajiamini wrote:
> Hi Ali, Andreas,
>
>
>
> I have tested dumping periodic stats for only ROI by putting the following
> statements in simulation.py:
>
-
Hi,
Do the cores execute the same code?
--
Fernando A. Endo, PhD student and researcher
CEA Lab
and
Universit? de Grenoble, UJF
France
2012/11/4 shervin hajiamini
> Hi Ali,
>
>
> Thank you very much for further
distribution of the cores, distances, etc.
If your results aren't wrong, they are very interesting!
Regards,
--
Fernando A. Endo, PhD student and researcher
Universit? de Grenoble, UJF
France
2012/11/5 shervin hajiamini
> Hi Fernando,
>
> Yes, all cores execute same code (Canneal be
Hi all,
I am dumping stats (in the ROI of canneal benchmark)
periodically. I have observed that the total number of instructions , sim_insts,
across all cores and for all periods is much larger than the total number of
instructions inside ROI and without period dumping, however the total number
Hi all,
Using periodicStateDump function in gem5, I have dumped the
stats for a benchmark periodically. I have used McPAT to measure the power
consumption
of all the cores PER PERIOD, and I am observing that the power value for each
period is very high (six times greater the power value for th
Hi,
I dump the stats of a benchmark every period and I measure
the power consumption in each period using McPAT. I observe that when the
period size decreases, the power consumption increases and when the period size
increases, the power consumption decreases. For example for period size of one
. You can check this by measuring each of
the 1M cycle intervals in the 10M cycles. You should see power go up and
down, but the average should be 0.6W. You can also compare McPAT's
output to see where the differences are.
Hope that helps.
-Erik
On 11/04/13 16:39, shervin hajiamini w
Hi,
Using statsDump(ticks) function in gem5, I dump statistics
of a benchmark periodically and I am wondering whether there is a lower bound
on the interval(period) size. I mean whether I can dump the statistics with an
arbitrary size for the interval or there is a threshold for the interval si
y it's most likely a bug and we should fix it.
Go ahead and give it a spin. Just make sure you've got plenty disk space.
Andreas
From: shervin hajiamini
mailto:hajamini_2...@yahoo.com>>
Reply-To: gem5 users mailing list
mailto:gem5-users@gem5.org>>
Date: Sunday, 14
Hi all,
I want to dump stats.txt every fixed interval of
instructions. There are already some posted messages in the mailing list about
using “testsys.cpu[i].max_insts_any_thread = options.maxinsts” in Simulation.py.
I have tested those suggestions but they don’t seem to work.
Can anybody ple
er, it should be
> some easier ways to dump stats based on the number of instructions,
> maybe someone else knows!
>
> Cheers
> Negar
>
> On 19/05/2013 00:43, shervin hajiamini wrote:
>>
>> Hi all,
>>
>>
>> I want to dump stats.txt every fixe
Hi all,
I want to dump the statistics periodically based on the fixed
number of instructions. I already followed the following post:
http://www.mail-archive.com/gem5-users@gem5.org/msg07544.html
and this is the code I use in Simulation.py for periodic
dumping (this code works in gem5-master vers
Hi all,
I want to run gem5 (Full System mode) with (out of order) cores having
'different'
frequencies. For example for a 2-core simulation,
frequency_core0=1.25GHz and frequency_core1=2.5GHz.
In 'fs.py' I found the following lines about the CPU clock:
TestCPUClass.clock = options.clock
Driv
Hi all,
The statistics file contain information about 'quiesce cycle' which value shows
the amount of cycles that a cpu is in the quiesce state until it receives an
interrupt. I have run a couple of benchmarks (splash2 and parsec) and I see (in
the statistics file) that the amount of quiesceCyc
Hi all,
I want to run a simulation in the Full System mode and with detailed CPU. How
can I make the CPUs remain busy(active) during the entire simulation? I have
been looking through the source code (cpu.cc file) and I found the following
snippet in 'tick()' function:
if (!activityRec.active()
Hi all,
In gem5 (base.hh file) 'numCycles' is declared as 'stats::Scalar'. I want to
cast 'numCycles' to other C++ variable types like 'integer' or 'double',
however gem5 does not allow the cast from 'Scalar' type to other data types.
Here is the error that gem5 gives:
build/ALPHA/cpu/o3/cpu.c
Thank you very much Amin!
On Tuesday, March 11, 2014 9:37 PM, Amin Farmahini wrote:
You can use numCycles.value(). See statistics.hh file.
Thanks,
Amin
On Tue, Mar 11, 2014 at 6:37 PM, shervin hajiamini
wrote:
Hi all,
>
>
>In gem5 (base.hh file) 'numCycles' is
Hi all,
I run my simulations in the full
system mode with detailed CPU (64 detailed CPUs) and I do change the frequency
of the CPUs dynamically during the simulation. I want to calculate the time
penalty of the execution
time of the ROI with dynamic frequency compared to the execution
time of the
Hi all,
I am new to gem5 and I have a
question as follows: I want to declare a variable (e.g. counter variable) and
keep incrementing it inside tick() function in cpu.cc file. I am wondering what
I can do to make gem5 give me JUST the LAST value of the counter variable in
stats.txt or any other
Hi all,
I am running PARSEC suit (canneal)
on 4 cores (O3). The workload is NOT equally distributed among the cores. As I
have created canneal’s .rcS file with 4 threads, could anybody please let me
know how I can assign JUST one thread per core in PARSEC so that the total
workload is distri
ot;windows-1252"
export GOMP_CPU_AFFINITY="0 1 2 3"
put this if u r using 4 core system with 4 threads. it ll pin the threads
On Tue, Jul 24, 2012 at 7:35 AM, shervin hajiamini
wrote:
> Hi all,
>
>
>
> I am running PARSEC suit (canneal) on 4 cores (O3). The workload
n even set that variable from within the
benchmark source code as well (I am sure google can provide a better and
more detailed explanation)
Malek
On Tue, Jul 24, 2012 at 2:25 PM, shervin hajiamini
wrote:
> Hi Biswabandan,
>
> Thank you very much for your help. If I want to have 32 cores
Hi all,
I want to specify in gem5 (cpu.cc file) when the region of interest
(ROI) of a benchmark (splash or parsec) starts and when it ends. Does anyone
know how I can do
this? I would appreciate any help.
Thanks,
Shervin
___
gem5-users mailing l
Hi guys,
Thank you very much for all your replies. My host machine is X86 (32 bits) and
target machine is Linux Ubuntu.
Regards,
Shervin
--- On Fri, 8/17/12, gem5-users-requ...@gem5.org
wrote:
From: gem5-users-requ...@gem5.org
Subject: gem5-users Digest, Vol 73, Issue 32
To: gem5-users@gem5
Hi all,
I have run Canneal benchmark on
gem5 with 64 cores. For windows of a fixed size (e.g. 100 CPU cycles per
window) I want to
calculate the power consumption for EACH window using McPAT tool. For doing
this,
gem5 has to generate stats.txt for each window which is then inputted to McPAT
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
python through swig that allows you to schedule
a periodic dumping of stats, "periodStatDump(long long period)". You can add
that to the script you use for your simulation.
Andreas
From: shervin hajiamini
mailto:hajamini_2...@yahoo.com>>
Reply-To: gem5 users mailing list
mailto:gem
There is a function exported to python through swig that allows you to schedule
a periodic dumping of stats, "periodStatDump(long long period)". You can add
that to the script you use for your simulation.
Andreas
From: shervin hajiamini
mailto:hajamini_2...@yahoo.com>>
Reply-T
from C++ as well, but why modify the source code if you don't
have to?
Andreas
From: shervin hajiamini
mailto:hajamini_2...@yahoo.com>>
Reply-To: gem5 users mailing list
mailto:gem5-users@gem5.org>>
Date: Sunday, 7 October 2012 19:45
To: "gem5-users@gem5.org<m
> To: gem5 users mailing
> list>
> Subject: Re: [gem5-users] dump stats for every N cycles
> Message-ID:>
> Content-Type: text/plain; charset=WINDOWS-1252
>
> I would suggest calling it from the Python scripts, when creating the system.
> You can call it from C++ as wel
Hi Sascha,
I added "from m5.internal.stats import periodicStatDump as statDump" to fs.py,
however while building gem5, it gives me the following error:
can not import name periodicStatDump
I can not find the definition of "periodStatDump" function in any of the python
files. I would appreciate
[mailto:gem5-users-boun...@gem5.org] On
Behalf Of shervin hajiamini
Sent: 09 October 2012 16:28
To: gem5-users@gem5.org
Subject: Re: [gem5-users] dump stats for every N cycles
Hi Sascha,
I added "from m5.internal.stats import periodicStatDump as statDump" to fs.py,
however while building gem5
once the simulation is running though.
Sascha
From: gem5-users-boun...@gem5.org [mailto:gem5-users-boun...@gem5.org] On
Behalf Of shervin hajiamini
Sent: 10 October 2012 15:25
To: gem5-users@gem5.org
Subject: Re: [gem5-users] dump stats for every N cycles
Hi Sascha,
Thank you very much for
Hi Sascha,
After I run gem5 with periodicStatDump(period), it seems that the dump data (in
stats.txt) for each period overwrites the previous dumps rather than appending
the dump data to the previous ones. Could you please let me know how I can make
the dump data (at each period) appended to th
Hi Andreas,
Thank you very much for the clarification. Besides using
periodicDumpStat(period), I found (through searching gem5 mailing list) that
the stats can be also dumped periodically by puting the statements below in
.rcS file:
m5 dumpstats[400]
10 100
m5 resetstats
This means tha
Hi Ali,
Thanks for the clarification.
Regards,
Shervin
Hi Shervin,
They're nano-seconds.
Ali
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
Hi Sascha, Andreas,
I have put periodicStatDump(period) in “options.banchmark”
section of fs.py, however after running the simulation, I have not seen dump
data in stats.txt so far.
Please let me
know whether I have put this function in the right place inside fs.py? If not
where shall I exact
Hi Sascha,
Thanks a lot, I try that.
By the way is there any chance that the stats are dumped only for the Region of
Interest (ROI) of a benchmark rather than for the entire benchmark. Any idea?
Regards,
Shervin
Hi Shervin,
I can only assume th
Hi Runjie,
Thank you very much.
Using periodicStatDump(period) which has been recently added to gem5, is it
possible to dump stats for just ROI (of a benchmark) in every N cycles/ticks?
Regards,
Shervin
--- On Wed, 10/17/12, Runjie Zhang wrote:
From: Runjie Zhang
Subject: Re: [gem5-users]
Hi all,
For dumping stats periodically I
am using statDump(period) which is called in fs.py. Does anyone know how I can
use this function call
to dump stats for JUST the region of interest of a benchmark?
Using statDump(period),
how can I create checkpoints for dumping stats periodically for
.S file in
your makefile.
Hope it will help.
2012/10/23, shervin hajiamini :
>
>
> Hi all,
>
>
> For dumping stats periodically I
> am using statDump(period)
which is called in fs.py. Does anyone know how I
> can use this function call
> to dump stats for JUST
can periodically dump the
stats at that point.
Ali
On Oct 24, 2012, at 1:07 AM, shervin hajiamini wrote:
>
> Hello Fernando,
>
> Thanks for the link.
>
> It seems that for getting the periodic stats of a benchmark I need to change
> the source code of the benchmark, h
5.simulate(ticks_till_ROI)
statDump(10)
m5.simulate(m5.MaxTick)
Ali
On Oct 24, 2012, at 1:43 AM, shervin hajiamini wrote:
> Ali,
>
> Based on the stats I get for the entire ROI (non-periodic) of a benchmark, I
> know approximately at which tick ROI starts and at which tick the ROI
). Could you please let me know how I can specify that in
simulation.py?
Regards,
Shervin
--- On Thu, 10/25/12, shervin hajiamini wrote:
From: shervin hajiamini
Subject: Re: dumping periodic stats with checkpoints
To: gem5-users@gem5.org
Date: Thursday, October 25, 2012, 5:17 AM
Hi Ali,
Thank
Hi Ali, Andreas,
I have tested dumping periodic stats for only ROI by putting
the following statements in simulation.py:
M5.simulate(start_tick_ROI)
statDump(40)
exit_event =
m5.simulate(end_tick_ROI)
The stats get dumped (every period) starting from start_tick_ROI, but
it does
second simulate should be end_tick_ROI - start_tick_ROI. You pass a
relative time to the function, not an absolute time.
Thanks,
Ali
On Oct 27, 2012, at 11:29 PM, shervin hajiamini wrote:
> Hi Ali, Andreas,
>
>
>
> I have tested dumping periodic stats for only ROI by puttin
.
Andreas
From: shervin hajiamini
Reply-To: shervin hajiamini , gem5 users mailing list
Date: Thursday, 1 May 2014 17:41
To: "gem5-users@gem5.org"
Subject: [gem5-users] Incorrect simulation time with dynamic frequency
Hi all,
I run my simulations in the full system mode with detail
52 matches
Mail list logo
