Hi,
Very interesting observation. There are several explanations here. First,
you may want to look at branch mispredictions. The pipeline flushes due to
memory order violations may be shadowed by flushes due to branch
mispredictions. You may want to increase your LSQ size to allow more
inflight
Hi,
This is because of address space info is not saved during serialization.
In src/sim/process.cc,
In void Process::serialize(CheckpointOut ) const, add the following lines
paramOut(cp, "mem_state.brk_point", memState->getBrkPoint());
paramOut(cp, "mem_state.stack_base",
Hi,
You can try running the benchmarks in timing CPU instead of detailed to see
if it works. As far as I know, there are various bugs in Ruby and using
detailed CPU makes the bugs more likely to be exposed.
On Mon, Aug 15, 2016 at 4:37 PM, George Mappouras <
georgios.mappou...@duke.edu> wrote:
Hi,
Are you taking checkpoints? If yes then getting a deadlock is normal
On Mon, Aug 15, 2016 at 2:28 PM, George Mappouras <
georgios.mappou...@duke.edu> wrote:
> Hi Jason,
>
> Thanks for the suggestions. I use MESI_Two_Level and I also compliled gem5
> for that protocol like this:
> *scons
Hi,
The implementation of LL/SC is buggy and wrongly in gem5. Some protocols
are buggy too.
This may be the cause of your problem. Can you guys try to run SPEC to see
if it runs?
On Tue, Mar 15, 2016 at 4:53 PM, Marcelo Melo Linck <
marcelo.li...@acad.pucrs.br> wrote:
> Hi Hao Zheng,
>
> I
Hi all,
I am going to simulate a machine with a distributed shared L2 cache. This
machine has a 4x4 mesh network. I plan to use the ruby memory model and
garnet network model.
I was wondering if anyone could give me any suggestions?
Thanks in advance.
Ruohuang Zheng
Hi all,
I am pretty new to gem5. I was trying to run gem5 with ruby memory mode.
I compiled gem5 with the following command:
scons build/ALPHA_MOESI_hammer/gem5.opt
Then I ran gem5:
./build/ALPHA/gem5.opt \
configs/example/ruby_random_test.py \
--num-cpus=16