Hello,
I figured out the answer to my question.
Thanks for your time.
Thanks,
Pavan
On Mon, Oct 1, 2012 at 11:27 AM, Pavan Poluri wrote:
> Yes HEAD_TAIL_ is equivalent to HEAD_ without BODY and TAIL flits. So a
> single control packet is like a path guider to a single data packet?
gt; On Oct 1, 2012, at 1:58 PM, Pavan Poluri wrote:
>
> Hello,
>
> Thanks for your reply. If the 1-flit packets of type HEAD_TAIL_ are
> control messages, I am unable to figure out why do we need to perform
> Routing Computation, Virtual Channel allocation etc computations on these
at 9:53 PM, Tushar Krishna wrote:
> Yes 1-flit packets (of type HEAD_TAIL_) are control messages from the
> protocol.
>
> - Tushar
>
> On Sep 25, 2012, at 8:04 PM, Pavan Poluri wrote:
>
> Hello,
>
> I have a question on the single flit packets generated when syntheti
bytes. My question is,
Are these 1 flit packets of type HEAD_TAIL_ control messages or are they
something different?
Thanks for your time.
Thanks,
Pavan
On Mon, Sep 17, 2012 at 1:56 PM, Pavan Poluri wrote:
> Thanks Tushar.
>
> Thanks,
> Pavan
>
>
> On Mon, Sep 17, 201
Thanks Tushar.
Thanks,
Pavan
On Mon, Sep 17, 2012 at 11:49 AM, Tushar Krishna wrote:
> Yes all that is correct.
>
> data msg size is 72 bytes, not 80. I will correct that on the wiki.
>
> cheers,
> Tushar
>
>
> On Sep 17, 2012, at 2:05 PM, Pavan Poluri wrote:
>
&g
s.
Since the flit size is 16 bytes, an 8 byte control message takes 1 flit and
72 bytes data message takes 5 flits.
Am I correct?
Thanks,
Pavan
On Mon, Sep 17, 2012 at 10:04 AM, Tushar Krishna wrote:
> Answers inline.
>
> On Sep 17, 2012, at 12:42 PM, Pavan Poluri wrote:
>
&g
flits and data message
fitting in 10 flits?
2. Are the sizes of control message (16 bytes) and data message (80 bytes)
fixed? Is it possible to modify their sizes?
Thanks for your time.
Thanks,
Pavan
On Wed, Sep 12, 2012 at 2:54 PM, Pavan Poluri wrote:
> Thanks Tushar.
>
> Thanks
> You could compare the energy numbers for each component from Orion and
> DSENT if you want to see how much they differ.
>
> The cache sizes are in configs/common/Options.py
>
>
> On 09/12/2012 03:26 PM, Pavan Poluri wrote:
>
> Hi Tushar,
>
> The simulation ran fo
ycles did your simulation run for?
> For full system runs in gem5, the network activity is typically very low
> (since network gets flits only on cache misses).
> As a result your dynamic power is very low.
> Network activity can be increased by reducing cache sizes.
>
> cheers,
>
n power.
>>
>> 2. I published a WDDD paper on Orion showing my own brief investigation
>> into why its power/area numbers seemed disconnected with reality. (
>> http://www.ece.wisc.edu/~hayenga/papers/wddd2012_hayenga.pdf)
>>
>> Hope this helps. Maybe the ver
Hello,
I have executed the Blackscholes application of the PARSEC benchmark suite
with 16 threads on the input file set (in_4.txt) with a full system
simulation with 16 cores, 16 L2 caches and 16 directories on a mesh
topology with 4 rows. I have used the MOESI_CMP_directory protocol. The
technolo
ts and pass them to DSENT.
>
> cheers,
> Tushar
>
>
>
> On 08/29/2012 02:32 PM, Pavan Poluri wrote:
>
> Hello,
>
> Thank you for your reply. I have a question on the static and dynamic
> power consumed.
> I have a simulated a 8 core 8 router mesh NoC with synthe
Aug 28, 2012, at 12:30 PM, Pavan Poluri wrote:
>
> > Hello,
> >
> > I have done a full system simulation with 8 cores, 8 routers, 8 L2
> caches on a mesh topology with garnet's fixed pipeline implementation. The
> simulation statistics can be seen in ruby.stats file t
Hello,
I have done a full system simulation with 8 cores, 8 routers, 8 L2 caches
on a mesh topology with garnet's fixed pipeline implementation. The
simulation statistics can be seen in ruby.stats file that contains router
power statistics like router dynamic power, router static power, router
clo
int here.
BTW am curious why you need to simulate a network-only protocol
(with no real application) for 10^11 cycles? From my experience the
network behavior stabilizes by 1M or 10M cycles.
- Tushar
On 04/30/2012 02:48 PM, Pavan Poluri wrote:
Hello,
I would like to run the network test
> converted into 1 flit, while data packets in vnet 2 get converted into 5
> flits.
> So if you look at ruby.stats, the number of total packets injected will be
> = flits_injected_vnet0 + flits_injected_vnet1 +(flits_injected_vnet2)/5
>
> - Tushar
>
> On Apr 29, 2012, at
injected
will be = flits_injected_vnet0 + flits_injected_vnet1 +
(flits_injected_vnet2)/5
- Tushar
On Apr 29, 2012, at 4:43 PM, Pavan Poluri wrote:
Hi Tushar,
Thanks for the quick reply. So if I run the experiment with
different random seeds over a large number of cycles the average
number
the flits injected
> stat from it by 5 to get packets injected.
>
> Let me know if you have more questions.
>
> - Tushar
>
>
> On Apr 29, 2012, at 3:25 PM, Pavan Poluri wrote:
>
> > Hello,
> >
> > I am trying to simulate the interconnection network usi
Hello,
I am trying to simulate the interconnection network using the Network_test
protocol. My question is on the file networktest.cc that is responsible for
generating and sending the packets. I understand that if I use --fixed-pkts
and --maxpackets options together I enforce each cpu node to onl
version of gem5 with Garnet?
Thanks for your time.
Thanks,
Pavan
On Mar 19, 2012, at 11:37 AM, Malogro wrote:
Did you remember to use the BigTsunami osfpal? It's supplied
together with the rest of the binaries.
Alexandre Yasuo Yamamoto
On Wed, Mar 7, 2012 at 3:41 PM, Pavan Poluri
Hello,
I have downloaded the most recent gem5 source code from the repository and
am trying to execute the PARSEC benchmark. I have built the gem5 binary
(gem5.opt) for full system simulation and also enabled RUBY (RUBY=true) and
PROTOCOL=MOESI_CMP_directory. But when I try to execute the PARSEC
b
Hello,
I am trying to run PARSEC benchmark on gem5 with Garnet model of fixed
pipeline. I was able to execute it for number of cpus = 4 successfully.
However, when I tried with number of cpus = 8 or 16, the program aborted
after running for like approximately 26 hrs. I am trying to figure out what
Hello,
I have a question regarding the synthetic traffic generated in garnet. Is
it possible to generate the exact same synthetic traffic in garnet in gem5
over and over again?
Thanks for your time!!!
Thanks,
Pavan
___
gem5-users mailing list
gem5-user
Hi,
My simulation finally ended after running for nearly 30 hours and I got
some stats. Thanks a lot Tushar and Alexandre for helping me in solving out
the issues I faced.
Thanks,
Pavan
On Sat, Feb 4, 2012 at 10:23 AM, Pavan Poluri wrote:
> Hi Alexandre,
>
> I issued a hg clone co
at 9:44 AM, Malogro wrote:
> I find it curious because I never ran into this problem. Are you using the
> latest repository version? Have you tried issuing the hg clone command
> again and recompiling just to see what happens?
>
> Alexandre Yasuo Yamamoto
>
>
>
> On
ystem Files" for ALPHA. Uncompress them in
> your /tmp/path. You should see the console binary and your problem should
> go away.
>
> Regards,
>
> Alexandre Yasuo Yamamoto
>
>
>
> On Wed, Feb 1, 2012 at 11:03 PM, Pavan Poluri wrote:
>
>> Hi,
>>
&g
"Done :D"
>> /sbin/m5 exit
>>
>> Do not forget to download system and benchmark images! Also, it takes a
>> while to run.
>>
>> If you need to do it using SE, I'm afraid I haven't tried to do it and
>> wouldn't be of much help.
&g
want to modify it to your specific needs of cores/etc.
>
> Regards,
>
> Alexandre Yasuo Yamamoto
>
>
>
> On Wed, Feb 1, 2012 at 1:30 PM, Pavan Poluri wrote:
>
>> Hi Tushar,
>>
>> I was wondering if it is possible to invoke GARNET through benchmark
>&
data messages (typically 5-flit wide) are
> typically more than 1 buffer deep.
>
> - Tushar
>
>
>
> On 01/30/2012 01:48 PM, Pavan Poluri wrote:
>
> Hi Tushar,
>
> Thanks for the tip!! Now, I am able to run the code without it being
> aborted. I am also able to see
ypically 1-flit wide) are 1
> buffer deep.
> VCs within vnets that carry data messages (typically 5-flit wide) are
> typically more than 1 buffer deep.
>
> - Tushar
>
>
>
> On 01/30/2012 01:48 PM, Pavan Poluri wrote:
>
> Hi Tushar,
>
> Thanks for the tip!! Now,
m = False
i += 1
# ADD THE FOLLOWING LINE:
system.system_port = system.ruby._sys_port_proxy.port
But remember to remove it when you do a hg pull -u later when the
patch gets pushed in.
cheers,
Tushar
On 1/28/12 2:19 PM, Pavan Poluri wrote:
Hi Tushar,
I did not receive any emails regardin
_port = system.ruby._sys_port_proxy.port
But remember to remove it when you do a hg pull -u later when the
patch gets pushed in.
cheers,
Tushar
On 1/28/12 2:19 PM, Pavan Poluri wrote:
Hi Tushar,
I did not receive any emails regarding the patch for the issue with
GARNET. I was just wondering may
On Jan 27, 2012, at 12:39 PM, Tushar Krishna wrote:
No emails related to code reviews and changes only come on the dev
list.
On 1/27/12 2:18 PM, Pavan Poluri wrote:
Hi Tushar,
I am currently building gem5 binary with MOESI_hammer. I will try
to invoke garnet with that binary as you
Thanks.. I subscribed to the dev list.
Thanks,
Pavan
On Jan 27, 2012, at 12:39 PM, Tushar Krishna wrote:
No emails related to code reviews and changes only come on the dev
list.
On 1/27/12 2:18 PM, Pavan Poluri wrote:
Hi Tushar,
I am currently building gem5 binary with MOESI_hammer. I
seems to have broken recently due to some other
> changes.
>
> cheers,
> Tushar
>
>
>
> On 1/27/12 1:51 PM, Pavan Poluri wrote:
>
> Hi Tushar,
>
> Thank you for your quick reply. Can you please tell me how can I get
> access to this patch you are suggesting? And
ev list to fix this issue...
Keep a look out on that list for the patch to be pushed, and then
try it ..
cheers,
Tushar
On 1/27/12 1:11 AM, Pavan Poluri wrote:
Hello,
I am trying to invoke garnet module in gem5.
As given in the documentation I built gem5 binary using
scons build/ALPHA_S
Keep a look out on that list for the patch to be pushed, and then
try it ..
cheers,
Tushar
On 1/27/12 1:11 AM, Pavan Poluri wrote:
Hello,
I am trying to invoke garnet module in gem5.
As given in the documentation I built gem5 binary using
scons build/ALPHA_SE_Network_Test/gem5.debug
After th
Hello,
I am trying to invoke garnet module in gem5.
As given in the documentation I built gem5 binary using
scons build/ALPHA_SE_Network_Test/gem5.debug
After that, I executed the following command
./build/ALPHA_SE_Network_test/gem5.debug configs/example/
ruby_network_test.py --num-cpus=16 --
38 matches
Mail list logo
