https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-8/-/blob/ccf40dfacd314ab0fea16cfc6f4eded1a08e2710/arch/x86/kernel/cpu/mce/core.c#L1331 preempt_enable() before nmi_exit(), is this a OK code?

if ((m.cs & 3) == 3) {
        /* If this triggers there is no way to recover. Die hard. */
        BUG_ON(!on_thread_stack() || !user_mode(regs));
        local_irq_enable();
        preempt_enable();

        current->task_struct_rh->mce_ripv = !!(m.mcgstatus & MCG_STATUS_RIPV);
        current->task_struct_rh->mce_whole_page = whole_page(&m);

        if (kill_it || do_memory_failure(&m))
            force_sig(SIGBUS, current);
        preempt_disable();
        local_irq_disable();
    } else {
        if (!fixup_exception(regs, X86_TRAP_MC))
            mce_panic("Failed kernel mode recovery", &m, NULL);
    }

out_ist:
    nmi_exit();
}

In my understanding, the kernel can't preempt if preempt_count > 0, in this case, it is inside NMI, so preempt_count > 0. It is sending SIGBUS:

force_sig(SIGBUS, current);

seems unnecessary to do local_irq_enable and preempt_enable?

I pulled up the console on a virtual machine I manage today and was greeted with some kernel messages: [5912557.130943] Uhhuh. NMI received for unknown reason 20 on CPU 0. [5912557.131115] Do you have a strange power saving mode enabled? [5912557.131287] Dazed and confused, but trying to continue [6064281.393568] Uhhuh. NMI received for unknown reason 30 on CPU 1. [6064281.393888] Do you have a strange power saving mode enabled? [6064281.394235] Dazed and confused, but trying to continue That's just a few of them, both 20 and 30 occur on CPU 0 and 1. - VM is Debian Jessie, BIOS boot ("QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.3-20161025_171302-gandalf 04/01/2014"; kernel 3.16.0-4-amd64) - Hypervisor is libvirt/KVM running on Debian testing (currently Debian's 4.7.0-1-amd64; qemu 1:2.7+dfsg-3). - Hardware is an Opteron 6344 on a [Supermicro H8SGL-F](http://www.supermicro.com/Aplus/motherboard/Opteron6000/SR56x0/H8SGL-F.cfm) with ECC RAM with scrub enabled. I don't see any NMI or EDAC error/warning messages on the host. Any idea what is causing these NMI messages on the guest? Are they anything to worry about? (May be related to https://unix.stackexchange.com/questions/216925/nmi-received-for-unknown-reason-20-do-you-have-a-strange-power-saving-mode-ena but that appears to be bare metal).

Following instructions , I want to disable the NMI watchdog on boot. sudo sh -c "echo '0' > /proc/sys/kernel/nmi_watchdog" However: Couldn't write '0' to 'kernel/nmi_watchdog': Unknown error 524 How to proceed?

I'm trying to create a custom interrupt handler for the NMI hardware button which exists on my motherboard. To test this functionality I've created this simple module: #include #include #include #include static int nmi_custom_handler(unsigned int val, struct pt_regs* regs) { pr_info("My custom NMI: 0x%x\n", val); return NMI_HANDLED; } static int __init nmi_handler_init(void) { pr_info("nmi_handler_init\n"); register_nmi_handler(NMI_UNKNOWN, nmi_custom_handler, 0, "my_custom_nmi"); return 0; } static void __exit nmi_handler_exit(void) { pr_info("nmi_handler_exit\n"); unregister_nmi_handler(NMI_UNKNOWN, "my_custom_nmi"); } module_init(nmi_handler_init); module_exit(nmi_handler_exit); MODULE_AUTHOR("Konstantin Aladyshev "); MODULE_LICENSE("GPL"); If I load this module and press NMI button one time, there will be "My custom NMI" message for every CPU core in my system. The same can be seen in the "/proc/interrupt" interface. NMI interrupt count increases from 0 to 1 for every CPU. But for some reason this works only once. Next button presses don't get logged by my module or the /proc interface. Why? What should I change to be able to use NMI hardware interrupt again?

I have Qemu VMs running FreeBSD, Windows and Linux, and I can send them a hardware NMI via the Qemu monitor.

qm monitor 100
Entering Qemu Monitor for VM 100 - type 'help' for help
qm> help nmi
nmi  -- inject an NMI

When inkecting the NMI to a Windows VM, I get the message is saving a crash dump, and this reboot the VM. On Linux I get the message

[26731.911302] Uhhuh. NMI received for unknown reason 31 on CPU 0.
[26731.911303] Do you have a strange power saving mode enabled?
[26731.911304] Dazed and confused, but trying to continue

How to get the kernel to print a stack trace on the console instead of only this message ? I would need this to debug VMs hanging because of very slow IO.

I'd like to understand what the following lines mean [ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x01] dfl dfl lint[0x0]) [ 0.000000] ACPI: NMI not connected to LINT 1! [ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x02] dfl dfl lint[0x0]) [ 0.000000] ACPI: NMI not connected to LINT 1! [ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x03] dfl dfl lint[0x0]) [ 0.000000] ACPI: NMI not connected to LINT 1! [ 0.000000] ACPI: LAPIC_NMI (acpi_id[0x04] dfl res lint[0x6f]) [ 0.000000] ACPI: NMI not connected to LINT 1! And why the value of the second to last line changes across the kernel version? e.g.: with kernel 4.9.3 it's [0x6f] with kernel 4.7.8 it's [0x1f] and so on

Why we need to keep the nmi_watchdog enabled and what could happen if I disable it permanently ? As some applications recommends to disable NMI watchdog to work properly, what's the advantage of disabling it ? And what does the results of this command, grep -i nmi /proc/interrupts mean ? NMI: 24 18 21 18 Non-maskable interrupts