Dear gem5 community, I have successfully simulated the HPCG benchmark using ARM Ubuntu in gem5 in 2 cores. However, when I try to simulate it using more cores (i.e. 8 cores) it takes more simulated time, and I get the following kernel panic after approximately 20 hours of simulation: [ 8.580827] Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000000 [ 8.580827] [ 8.580837] CPU: 0 PID: 1 Comm: login Not tainted 4.18.0+ #1 [ 8.580843] Hardware name: V2P-CA15 (DT) [ 8.580849] Call trace: [ 8.580856] dump_backtrace+0x0/0x1c0 [ 8.580863] show_stack+0x14/0x20 [ 8.580871] dump_stack+0x8c/0xac [ 8.580877] panic+0x130/0x288 [ 8.580884] complete_and_exit+0x0/0x20 [ 8.580891] do_group_exit+0x38/0xa0 [ 8.580898] __wake_up_parent+0x0/0x28 [ 8.580904] el0_svc_naked+0x30/0x34 [ 8.580914] SMP: stopping secondary CPUs [ 8.580921] Kernel Offset: disabled [ 8.580926] CPU features: 0x21c06292 [ 8.580932] Memory Limit: 4096 MB [ 8.580944] ---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000000 [ 8.580944] ]--- I use the following cmd: $GEM5/build/ARM/gem5.fast --listener-mode=on -r -d $GEM5/node0 $GEM5/configs/example/arm/starter_fs.py --kernel=vmlinux.arm64 --num-cores=8 --disk-image=ubuntu-18.04-arm64.img and I use the Latest Linux Kernel Image / Bootloader and Latest Linux Disk Images from here (https://www.gem5.org/documentation/general_docs/fullsystem/guest_binaries). I really appreciate any help! Best regards, Nikos

Hi, Based on my experience, more cores does not means more work, unless you specify more work. To elaborate: running same 1 thread application on 2 cores should be relatively close to 8 cores. More time will be needed (I think close to linearly) when you have more actual work, such as 8 threads on 8 cores. I have not used HPCG, so if thats a single thread benchmark, then what you seeing should not be happening. Also, based on the timestamps, this happens way pass boot up (usually within 1 sec for mine). Then the behavior is very dependent on what you are running(kernel, os, application itself). It does not seems like user application issue. I would guess maybe: any modified kernel function? Such that the scheduler have weird interaction trying to kill your init process. Or, very unlucky, you did not modify kernel in anyway, but you ran the application in a way (with priority tuning, runtime policies changes, etc) trigger this bug. Best, Heng > On Apr 4, 2023, at 04:50, Νικόλαος Ταμπουρατζής via gem5-users <gem5-users@gem5.org> wrote: > > > Dear gem5 community, > > I have successfully simulated the HPCG benchmark using ARM Ubuntu in gem5 in 2 cores. However, when I try to simulate it using more cores (i.e. 8 cores) it takes more simulated time, and I get the following kernel panic after approximately 20 hours of simulation: > > [ 8.580827] Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000000 > [ 8.580827] > [ 8.580837] CPU: 0 PID: 1 Comm: login Not tainted 4.18.0+ #1 > [ 8.580843] Hardware name: V2P-CA15 (DT) > [ 8.580849] Call trace: > [ 8.580856] dump_backtrace+0x0/0x1c0 > [ 8.580863] show_stack+0x14/0x20 > [ 8.580871] dump_stack+0x8c/0xac > [ 8.580877] panic+0x130/0x288 > [ 8.580884] complete_and_exit+0x0/0x20 > [ 8.580891] do_group_exit+0x38/0xa0 > [ 8.580898] __wake_up_parent+0x0/0x28 > [ 8.580904] el0_svc_naked+0x30/0x34 > [ 8.580914] SMP: stopping secondary CPUs > [ 8.580921] Kernel Offset: disabled > [ 8.580926] CPU features: 0x21c06292 > [ 8.580932] Memory Limit: 4096 MB > [ 8.580944] ---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000000 > [ 8.580944] ]--- > > I use the following cmd: $GEM5/build/ARM/gem5.fast --listener-mode=on -r -d $GEM5/node0 $GEM5/configs/example/arm/starter_fs.py --kernel=vmlinux.arm64 --num-cores=8 --disk-image=ubuntu-18.04-arm64.img > > and I use the Latest Linux Kernel Image / Bootloader and Latest Linux Disk Images from here (https://www.gem5.org/documentation/general_docs/fullsystem/guest_binaries). > > I really appreciate any help! > > Best regards, > Nikos > > _______________________________________________ > gem5-users mailing list -- gem5-users@gem5.org > To unsubscribe send an email to gem5-users-leave@gem5.org