Hi chunfeng,
Are the packets in your routing algorithm moving clockwise or
counterclockwise? This type of routing algorithm can be prone to deadlock.
To prevent deadlock from occurring, consider implementing deadlock-free
algorithms such as XY routing.
Alternatively, you can use deadlock-free
Hi Jiayi,
Thanks for reply. I have run parsec on garnet in the FS mode
successfully using ruby. Thanks a lot :)
best,
Zifeng Zhao
--
该邮件从移动设备发送
--原始邮件--
发件人:"JiayiHuangviagem5-users "___
gem5-users mailing list --
Hello,
I have run the standalone simulation(random, tornado traffic,
etc.) over the Garnet model.
I noticed that there are some papers that had run the PARSEC benchmark to
evaluate
their NoC scheme's performance.
Then I look through some current full system script(e.g.,
Dear Professor Tushar,
Can DSENT estimate the power/area of control module that added
into the Garnet?
Best wishes,
Zifeng Zhao
--Original--
From:
"Krishna,Tusharviagem5-users"https://github.com/mit-carbon/Graphite/tree/master/contrib/dsent/dsent-core
Hi all,
I am using Garnet model for my research. And I want to know how to
evaluate the power/area consumption when
I have changed the Garnet model (i.e., added some additional control logic, not
simply change the number of buffer, vc, vnet. etc.).
Can DSENT model do this?
Best wishes,
hi,
The virtual networks are mainly used for avoiding protocol deadlocks,
i.e., one message type per vnet.
the output 5 virtual channels is because you set vcs-per-vnet=1(i.e., one
virtual channel per vnet)
and the garnet model has 5 virtual networks (I think so)
As for the last problem, it's
hi Soliman,
The parameter"simcycles" in the command line actually refers to "ticks"
In ruby the default value of a cycle is 500 ticks. So simcycles=1000 equal to
1000 ticks or 2 cycles
Zifeng Zhao
--
该邮件从移动设备发送
--原始邮件--
发件人:"gem5-users "