Comment 71 for bug 254668

About the clonezilla kernel: If I hadn't been too tired yesterday it would have been clear to me why there are no pauses. This command line has pretty much any no<whatever> option set.

Especially nosmp, which also disables lapic. In that case the HPET is directly responsible for triggering timer interrupts. So there is no need for the code to go through broadcast and those messages are expected to not show up.

To document my thoughts on the noisy debug run:

[ 2.924895] CPU#1: enter broadcast
[ 2.924958] clockevents_set_mode(lapic, 1)
[ 2.925018] CPU#0: enter broadcast
[ 2.925077] clockevents_set_mode(lapic, 1)
<both CPUs idle>
[ 2.925136] CPU#1: exit broadcast
[ 2.925195] clockevents_set_mode(lapic, 3)
<CPU#0 idle CPU#1 active>
[ 2.925287] CPU#0: exit broadcast
[ 2.925353] clockevents_set_mode(lapic, 3)
<both CPUs active>
[ 2.925415] CPU#1: enter broadcast
[ 2.925422] ACPI: CPU0 (power states: C1[C1] C2[C2] C3[C3])
[ 2.925543] clockevents_set_mode(lapic, 1)
<CPU#0 active CPU#1 idle>
[ 2.925603] CPU#1: exit broadcast
[ 2.925661] clockevents_set_mode(lapic, 3)
<both CPUs active>
[ 2.925767] processor ACPI0007:00: registered as cooling_device0
[ 2.925771] CPU#1: enter broadcast
[ 2.925772] clockevents_set_mode(lapic, 1)
<CPU#0 active CPU#1 idle>
[ 2.925953] ACPI: Processor [CPU0] (supports 8 throttling states)
[ 2.926144] CPU#1: exit broadcast
[ 2.926210] clockevents_set_mode(lapic, 3)
<both CPUs active>
[ 2.926412] ACPI: CPU1 (power states: C1[C1] C2[C2] C3[C3])
[ 2.926703] processor ACPI0007:01: registered as cooling_device1
[ 2.926771] ACPI: Processor [CPU1] (supports 8 throttling states)
[ 2.932389] ACPI Exception (thermal-0377): AE_OK, No or invalid critical threshold [20080609]
<snip>
[ 3.733585] ohci_hcd: 2006 August 04 USB 1.1 'Open' Host Controller (OHCI) Driver
[ 3.734069] ACPI: PCI Interrupt Link [LUS0] enabled at IRQ 18
[ 3.734135] IOAPIC[0]: Set PCI routing entry (2-18 -> 0xb9 -> IRQ 18 Mode:1 Active:1)
[ 3.734215] ohci_hcd 0000:00:02.0: PCI INT A -> Link[LUS0] -> GSI 18 (level, low) -> IRQ 18
[ 3.734297] ohci_hcd 0000:00:02.0: setting latency timer to 64
[ 3.734361] ohci_hcd 0000:00:02.0: OHCI Host Controller
[ 3.734471] ohci_hcd 0000:00:02.0: new USB bus registered, assigned bus number 1
[ 3.734570] ohci_hcd 0000:00:02.0: irq 18, io mem 0xf6486000
[ 3.771357] libata version 3.00 loaded.
[ 3.790187] usb usb1: configuration #1 chosen from 1 choice
[ 3.790284] hub 1-0:1.0: USB hub found
[ 3.790355] hub 1-0:1.0: 7 ports detected
<Pause, both CPUs not idle>
[ 17.460971] Clocksource tsc unstable (delta = 4398045749678 ns)
[ 17.462224] ACPI: PCI Interrupt Link [Z018] enabled at IRQ 18
[ 17.462294] ohci_hcd 0000:00:04.0: PCI INT A -> Link[Z018] -> GSI 18 (level, low) -> IRQ 18
[ 17.462379] ohci_hcd 0000:00:04.0: setting latency timer to 64

So I would get it from that log that neither CPU had broadcast active at the time of the pause. The question is whether the CPUs can still be in a sort of idle state...