I have figured out how to it. I added a boolean named roi in the system.hh
header file, right after the totalNumInsts variable. This is now available
in src/mem/cache/base.cc, so I can print the cache data only in the region
of interest of the benchmarks.
I'm also interested in changing the
When I enable the use_virtual_addresses ans start feeding the prefetchers
with VA address, I see no translation happening, and I ended up with
situation where all new addresses are dropped because the reference to tlb
is null. Any help is appreciated.
use_virtual_addresses = Param.Bool(True,
Alright thanks!
De: Ciro Santilli
Enviado: sexta-feira, 25 de outubro de 2019 13:58
Para: gem5-users@gem5.org ; iagosilvestr...@hotmail.com
Assunto: Re: [gem5-users] Do I need to re-create checkpoints if I change the
CPU model parameters
Considering that you
I sucessfully booted the simulation with HPI cpu model with one core but
whenever I try to run it with 4 I get an error.
panic: panic condition (pkt->needsWritable() != pkt->isInvalidate()) &&
!pkt->req->isCacheMaintenance() occurred: global got snoop WriteReq
[80945000:80945003] UC where
Considering that you can even checkpoint from completely different CPUs such at
Atomic and restore to HPI, I would expect that such HPI changes will also
magically just work.
From: gem5-users on behalf of Iago .
Sent: Friday, October 25, 2019 5:22 PM
To:
I have been trying to use HPI cpu model in my control algorithm performance
study and searching for more info on it I found that the default HPI parameters
set the l1 cache latency to 1 which seems to be very low, If I already created
checkpoints with this CPU model can i just alter the
Thanks for the report Javed,
I reproduce on other configs as well e.g. se.py.
This seems to have been caused by 9d3b9e96c56386ee6539657c21cba95e118e576a.
I'm not very familiar with the ports, and could not fix it immediately, I'll
have a look later on if Gabe doesn't get a chance to fix it
Hello
I cloned a cleaned version of GEM5 from the repository and got the ARM Isa
gem5.opt built.
However, when I run the simple hello world test for the HPI() arm model, I am
getting the following error:
[j00533938@lhrplinux1 gem5]$ ./build/ARM/gem5.opt
configs/example/arm/starter_se.py