[gem5-users] m5threads for Aarch64 SE

2021-12-07 Thread adrian via gem5-users
Hi guys, I’m trying to generate some multithreaded workloads for aarch64 in SE. However, m5threads only supports arm32 bits. Is there any way to generate such workloads? Thank you ___ gem5-users mailing list -- gem5-users@gem5.org To unsubscribe send

[gem5-users] Re: gem5 + GPU support?

2021-06-30 Thread adrian via gem5-users
Hi Jason, Thanks a lot for this useful information. Is it possible to have several CPUs + several GPUs running in parallel? I'm using Arm as the CPU's ISA. I don't know how this GPU ISA deals with other ISAs. Sincerely, Adrián ___ gem5-users mailing

[gem5-users] gem5 + GPU support?

2021-06-30 Thread adrian via gem5-users
Hi, I am interested in simulating a GPU in gem5. Just by googling I discovered a repo from 2017 but I've also seen gem5 v21 has some files regarding to the GPU. Could you please confirm gem5 v21 supports a GPU model? There is not much info about it on internet. How can this GPU model be used?

[gem5-users] Re: Ruby in Arm?

2021-05-09 Thread adrian via gem5-users
Hi Giacomo, Thanks for your response. I've managed to run the simulator with a single core. However, I cannot do the same with more than one. This is the command I use to boot. ./build/ARM/gem5.opt --outdir m5out/booted_machine_4 configs/example/fs.py --kernel=binaries/vmlinux.arm64

[gem5-users] Ruby in Arm?

2021-05-07 Thread adrian via gem5-users
Hi guys, I've always used the classic memory model when running timing simulations using Arm, but I'd like to run some experiments using Ruby. I've seen some people using it and asking questions about it, but my current version of gem5 shows this warning when enabling it: "warn: You are

[gem5-users] Power model for bigLITTLE configuration with a L3 cache.

2021-03-18 Thread adrian via gem5-users
Hi gem5 users, I'm doing some experiments with the power model in the simulator. I'm currently using the default "fs_power.py" script from the "stable" branch. I'm not having any issues when adding the power model to cores and caches when only using a single bigCluster. However, at the moment I