----- Original Message ----- > 于 2014/4/27 7:43, Dave Anderson 写道: > > > > > > ----- Original Message ----- > >> On 2014/4/26 3:20, Dave Anderson wrote: > >>> > >>> > >>> ----- Original Message ----- > >>>> For ARM32 platfrom, The system will "offline" all CPUs except the > >>>> crashing one, by clear the cpu_online_mask. So we need to find > >>>> another way to get online-CPUs number for crash utility. > >>>> > >>>> This patch uses cpu_active_mask to get that value. > >>>> > >>>> Signed-off-by: Liu Hua <sdu.liu@xxxxxxxxxx> > >>>> --- > >>>> arm.c | 2 +- > >>>> 1 file changed, 1 insertion(+), 1 deletion(-) > >>>> > >>>> diff --git a/arm.c b/arm.c > >>>> index 3c38cd5..c94d7bf 100644 > >>>> --- a/arm.c > >>>> +++ b/arm.c > >>>> @@ -1518,7 +1518,7 @@ arm_display_machine_stats(void) > >>>> static int > >>>> arm_get_smp_cpus(void) > >>>> { > >>>> - return get_cpus_online(); > >>>> + return get_cpus_active(); > >>>> } > >>>> > >>>> /* > >>> > >>> When did this start happening? I ask because no other ARM users have > >>> reported this until now. I've only got one sample SMP vmcore, and > >>> clearly the non-crashing cpu was not offlined: > >>> > >> > >> > >> Hi Dave, > >> > >> commit 36a2e5010fe87 "kexec: offline non panic CPUs on Kdump panic " > >> introduces this change. <v3.10.27-1-g36a2e50> > >> > >> The following messges come from the ARM boards on my hand. The crash > >> utility has added my patchs. > >> > >> (1) ARM A15 16 CPUs > >> > >> crash> sys > >> > >> KERNEL: vmlinux > >> DUMPFILE: vmcore_my [PARTIAL DUMP] > >> CPUS: 16 > >> DATE: Fri Feb 28 10:07:55 2014 > >> UPTIME: 00:03:57 > >> LOAD AVERAGE: 0.03, 0.01, 0.01 > >> TASKS: 140 > >> NODENAME: arma15el > >> RELEASE: 3.10.37+ > >> VERSION: #2 SMP Thu Apr 24 21:13:28 CST 2014 > >> MACHINE: armv7l (unknown Mhz) > >> MEMORY: 3.1 GB > >> PANIC: "Internal error: Oops: a07 [#1] SMP ARM" (check log for > >> details > >> > >> crash> help -k > >> .... > >> cpu_possible_map: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 > >> cpu_present_map: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 > >> cpu_online_map: 0 > >> ... > >> > >> > >> (2) ARM A9 4 CPUs > >> > >> crash> sys > >> KERNEL: vmlinux > >> DUMPFILE: vmcore > >> CPUS: 2 > >> DATE: Mon Apr 14 11:09:10 2014 > >> UPTIME: 00:06:17 > >> LOAD AVERAGE: 0.16, 0.05, 0.02 > >> TASKS: 45 > >> NODENAME: arma9el > >> RELEASE: 3.10.37+ > >> VERSION: #6 SMP Tue Apr 22 13:23:30 CST 2014 > >> MACHINE: armv7l (unknown Mhz) > >> MEMORY: 1.5 GB > >> PANIC: "Internal error: Oops: 817 [#1] SMP ARM" (check log for > >> details) > >> > >> crash> help -k > >> .... > >> cpu_possible_map: 0 1 > >> cpu_present_map: 0 1 > >> cpu_online_map: 1 > >> > >> > >> crash> p per_cpu__runqueues > >> > >> PER-CPU DATA TYPE: > >> struct rq runqueues; > >> PER-CPU ADDRESSES: > >> [0]: c069a680 > >> [1]: c06a2680 > >> > >> With my patches, "sys" command is ok. And I can get per_cpu variables > >> correctly. > >> But something should be done for "help -k". May be I should add fully > >> support > >> for cpu_active_mask. If necessory, I will resend this patch series. > > > > It's not necessary, but it would be nice to have, and would be > > easy to add. > > > > Also, would it be possible that you can let me have a copy of your > > 16-cpu vmlinux/vmcore pair? That would be a great sample for > > my testing. > > > > Thanks, > > Dave > > Hi Dave, > > Sure, there seems no extra problems caused by the commit above. We can add it > when necessory. Will you be re-posting the original patch? (with or without the help -k support) > > And, I am very sorry to tell you that I have no authority for sending binary files out of my > company, becasue of it's security rules. I have no ways to send that huge file out. Maybe > I can do some test for you on my platform. OK I understand. Thanks, Dave -- Crash-utility mailing list Crash-utility@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/crash-utility