Hi all,
I have a problem with KVM cpu with full system mode. Actually, it was
working for a month, but suddenly it stopped working today after I reboot
the local machine. I regularly reboot every week, and it only happens today.
I don't have any error message, but it freezes during the boot up. Here is
the point that my machine freezes during boot up.
==== m5 terminal: Terminal 0 ====
Linux version 4.8.13 (yin@Workstation) (gcc version 7.5.0 (Ubuntu
7.5.0-3ubuntu1~18.04) ) #1 SMP Wed Apr 22 10:04:56 CDT 2020
Command line: earlyprintk=ttyS0 console=ttyS0 lpj=7999923 root=/dev/hda1
*//Skip few lines of output*
Memory: 1021560K/1048184K available (4207K kernel code, 593K rwdata, 1624K
rodata, 992K init, 636K bss, 26624K reserved, 0K cma-reserved)
Hierarchical RCU implementation.
Build-time adjustment of leaf fanout to 64.
RCU restricting CPUs from NR_CPUS=8192 to nr_cpu_ids=1.
RCU: Adjusting geometry for rcu_fanout_leaf=64, nr_cpu_ids=1
NR_IRQS:524544 nr_irqs:256 16
Console: colour dummy device 80x25
console [ttyS0] enabled
console [ttyS0] enabled
bootconsole [earlyser0] disabled
bootconsole [earlyser0] disabled
*___FREEZE AFTER THIS POINT___*
I also tried *strace* on the host machine to find the reason.
*//Skip the first 40K outputs*
perf_event_open({type=PERF_TYPE_HARDWARE, size=PERF_ATTR_SIZE_VER0,
config=PERF_COUNT_HW_CPU_CYCLES, ...}, 0, -1, -1, 0) = 29
mmap(NULL, 8192, PROT_READ|PROT_WRITE, MAP_SHARED, 29, 0) = 0x7f56320e8000
perf_event_open({type=PERF_TYPE_HARDWARE, size=PERF_ATTR_SIZE_VER0,
config=PERF_COUNT_HW_INSTRUCTIONS, ...}, 0, -1, 29, 0) = 30
mmap(NULL, 8192, PROT_READ|PROT_WRITE, MAP_SHARED, 30, 0) = 0x7f56320e6000
gettid() = 27415
timer_create(CLOCK_MONOTONIC, {sigev_signo=SIGRT_2,
sigev_notify=SIGEV_THREAD_ID, sigev_notify_thread_id=27415}, [1]) = 0
futex(0x560033196f48, FUTEX_WAIT_PRIVATE, 0, NULL) = 0
futex(0x560033196ef8, FUTEX_WAKE_PRIVATE, 1) = 0
*//Skip 6K outputs*
futex(0x56002e66bb68, FUTEX_WAKE_PRIVATE, 1) = 0
getpid() = 27415
tgkill(27415, 27662, SIGRT_2) = 0
futex(0x560033196f48, FUTEX_WAIT_PRIVATE, 0, NULL) = ? ERESTARTSYS (To be
restarted if SA_RESTART is set)
--- SIGIO {si_signo=SIGIO, si_code=SI_KERNEL} ---
rt_sigreturn({mask=[RT_2]}) = 202
I tried the following to solve this problem, but none of them worked.
1. Reinstall the gem5
2. Use other kernel or disk image
I also have another machine with the same code. I tried the following with
another machine to validate the setup is correct, and there is no issue.
1. To validate disk image and kernel image, copy the image and kernel into
another machine, and launch gem5 full system mode.
2. To validate the code, clone the repository and build with the same
gcc/g++/python/scons version.
I believe something goes wrong in my machine environment, but I'm not sure
how to figure it out. I would appreciate any idea/suggestion/experience.
Thanks,
Jeageun
_______________________________________________
gem5-users mailing list -- [email protected]
To unsubscribe send an email to [email protected]
%(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
