linux/arch/x86/kernel
Thomas Gleixner fd8a7de177 x86: cpu-hotplug: Prevent softirq wakeup on wrong CPU
After a newly plugged CPU sets the cpu_online bit it enables
interrupts and goes idle. The cpu which brought up the new cpu waits
for the cpu_online bit and when it observes it, it sets the cpu_active
bit for this cpu. The cpu_active bit is the relevant one for the
scheduler to consider the cpu as a viable target.

With forced threaded interrupt handlers which imply forced threaded
softirqs we observed the following race:

cpu 0                         cpu 1

bringup(cpu1);
                              set_cpu_online(smp_processor_id(), true);
		              local_irq_enable();
while (!cpu_online(cpu1));
                              timer_interrupt()
                                -> wake_up(softirq_thread_cpu1);
                                     -> enqueue_on(softirq_thread_cpu1, cpu0);

                                                                        ^^^^

cpu_notify(CPU_ONLINE, cpu1);
  -> sched_cpu_active(cpu1)
     -> set_cpu_active((cpu1, true);

When an interrupt happens before the cpu_active bit is set by the cpu
which brought up the newly onlined cpu, then the scheduler refuses to
enqueue the woken thread which is bound to that newly onlined cpu on
that newly onlined cpu due to the not yet set cpu_active bit and
selects a fallback runqueue. Not really an expected and desirable
behaviour.

So far this has only been observed with forced hard/softirq threading,
but in theory this could happen without forced threaded hard/softirqs
as well. It's probably unobservable as it would take a massive
interrupt storm on the newly onlined cpu which causes the softirq loop
to wake up the softirq thread and an even longer delay of the cpu
which waits for the cpu_online bit.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Peter Zijlstra <peterz@infradead.org>
Cc: stable@kernel.org # 2.6.39
2011-06-08 11:21:19 +02:00
..
acpi x86, ioapic: Consolidate mp_ioapics[] into 'struct ioapic' 2011-05-20 13:40:59 +02:00
apic x86, UV: Add support for SGI UV2 hub chip 2011-05-25 14:20:13 +02:00
cpu Merge branch 'idle-release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-idle-2.6 2011-05-29 11:18:09 -07:00
.gitignore
Makefile x86: Put back -pg to tsc.o and add no GCOV to vread_tsc_64.o 2011-05-27 23:47:16 -04:00
alternative.c Merge branch 'x86-cpu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2011-05-19 17:55:12 -07:00
amd_gart_64.c x86, gart: Rename pci-gart_64.c to amd_gart_64.c 2011-05-10 17:22:06 +02:00
amd_iommu.c x86/amd-iommu: Fix boot crash with hidden PCI devices 2011-06-07 10:06:59 +02:00
amd_iommu_init.c x86/amd-iommu: Fix 3 possible endless loops 2011-06-06 16:10:15 +02:00
amd_nb.c x86, amd-nb: Rename CPU PCI id define for F4 2011-03-31 08:51:38 +02:00
apb_timer.c Merge branch 'consolidate-clksrc-i8253' of master.kernel.org:~rmk/linux-2.6-arm into timers/clocksource 2011-05-14 12:06:36 +02:00
aperture_64.c x86, gart: Don't enforce GART aperture lower-bound by alignment 2011-04-18 10:31:38 -07:00
apm_32.c Merge branch 'idle-release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-idle-2.6 2011-05-29 11:18:09 -07:00
asm-offsets.c x86, asm: Cleanup unnecssary macros in asm-offsets.c 2011-02-25 16:37:32 -08:00
asm-offsets_32.c x86: Partly unify asm-offsets_{32,64}.c 2011-02-10 13:31:37 +01:00
asm-offsets_64.c x86: Partly unify asm-offsets_{32,64}.c 2011-02-10 13:31:37 +01:00
audit_64.c
bootflag.c include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h 2010-03-30 22:02:32 +09:00
check.c x86: Don't check for BIOS corruption in first 64K when there's no need to 2011-03-09 16:36:41 +01:00
cpuid.c BKL: remove extraneous #include <smp_lock.h> 2010-11-17 08:59:32 -08:00
crash.c x86, UV: Make kdump avoid stack dumps 2010-07-21 11:33:27 -07:00
crash_dump_32.c crash_dump: export is_kdump_kernel to modules, consolidate elfcorehdr_addr, setup_elfcorehdr and saved_max_pfn 2011-03-23 19:47:19 -07:00
crash_dump_64.c crash_dump: export is_kdump_kernel to modules, consolidate elfcorehdr_addr, setup_elfcorehdr and saved_max_pfn 2011-03-23 19:47:19 -07:00
devicetree.c x86, ioapic: Consolidate gsi routing info into 'struct ioapic' 2011-05-20 13:41:01 +02:00
doublefault_32.c
dumpstack.c Merge branch 'driver-core-next' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core-2.6 2011-05-19 18:24:11 -07:00
dumpstack_32.c x86, dumpstack: Correct stack dump info when frame pointer is available 2011-03-18 10:51:42 +01:00
dumpstack_64.c x86, dumpstack: Correct stack dump info when frame pointer is available 2011-03-18 10:51:42 +01:00
e820.c crash_dump: export is_kdump_kernel to modules, consolidate elfcorehdr_addr, setup_elfcorehdr and saved_max_pfn 2011-03-23 19:47:19 -07:00
early-quirks.c x86, quirk: Fix SB600 revision check 2011-03-16 14:03:32 +01:00
early_printk.c x86, earlyprintk: Move mrst early console to platform/ and fix a typo 2010-12-06 20:52:04 +01:00
entry_32.S Merge branch 'x86-trampoline-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2011-03-16 10:10:02 -07:00
entry_64.S x86: Fix common misspellings 2011-03-18 10:39:30 +01:00
ftrace.c x86/ftrace: Fix compiler warning in ftrace.c 2011-05-25 19:56:26 -04:00
head.c x86: Use memblock to replace early_res 2010-08-27 11:12:29 -07:00
head32.c Merge branch 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2011-05-19 18:08:06 -07:00
head64.c x86: Cleanup highmap after brk is concluded 2011-03-19 11:58:19 -07:00
head_32.S Merge branch 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2011-03-15 20:01:36 -07:00
head_64.S x86, trampoline: Common infrastructure for low memory trampolines 2011-02-17 21:02:43 -08:00
hpet.c x86: hpet: Cleanup the clockevents init and register code 2011-05-19 14:24:16 +02:00
hw_breakpoint.c x86: Use this_cpu_ops to optimize code 2010-12-30 12:20:28 +01:00
i386_ksyms_32.c
i387.c x86: Fix common misspellings 2011-03-18 10:39:30 +01:00
i8237.c x86: Use syscore_ops instead of sysdev classes and sysdevs 2011-03-23 22:15:54 +01:00
i8253.c x86: Convert PIT to clockevents_config_and_register() 2011-05-19 14:24:16 +02:00
i8259.c x86: Use syscore_ops instead of sysdev classes and sysdevs 2011-03-23 22:15:54 +01:00
init_task.c
io_delay.c
ioport.c x86: Use bitmap library functions 2011-02-17 14:59:22 +01:00
irq.c x86: Don't unmask disabled irqs when migrating them 2011-05-19 14:51:08 +02:00
irq_32.c x86: Fix common misspellings 2011-03-18 10:39:30 +01:00
irq_64.c
irq_work.c irq_work: Add generic hardirq context callbacks 2010-10-18 19:58:50 +02:00
irqinit.c Merge branch 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2011-03-15 20:01:36 -07:00
jump_label.c x86, cpu: Clean up and unify the NOP selection infrastructure 2011-04-18 16:40:21 -07:00
kdebugfs.c llseek: automatically add .llseek fop 2010-10-15 15:53:27 +02:00
kgdb.c Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jwessel/linux-2.6-kgdb 2011-03-25 21:04:56 -07:00
kprobes.c kprobes, x86: Disable irqs during optimized callback 2011-05-11 13:21:23 +02:00
kvm.c KVM guest: Fix section mismatch derived from kvm_guest_cpu_online() 2011-03-17 13:08:25 -03:00
kvmclock.c x86: Convert remaining x86 clocksources to clocksource_register_hz/khz 2011-02-21 13:33:33 -08:00
ldt.c include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h 2010-03-30 22:02:32 +09:00
machine_kexec_32.c
machine_kexec_64.c x86, cleanups: Use clear_page/copy_page rather than memset/memcpy 2010-09-22 15:36:49 -07:00
mca_32.c x86: Fix common misspellings 2011-03-18 10:39:30 +01:00
microcode_amd.c x86, microcode, AMD: Fix signedness bug in generic_load_microcode() 2011-02-20 14:01:32 +01:00
microcode_core.c x86, microcode: Unregister syscore_ops after microcode unloaded 2011-03-29 11:12:04 +02:00
microcode_intel.c x86/microcode: Fix double vfree() and remove redundant pointer checks before vfree() 2010-12-27 14:33:30 +01:00
mmconf-fam10h_64.c x86-64: Fix and clean up AMD Fam10 MMCONF enabling 2010-11-18 13:41:35 +01:00
module.c jump label: Introduce static_branch() interface 2011-04-04 12:48:08 -04:00
mpparse.c x86, ioapic: Consolidate mp_ioapics[] into 'struct ioapic' 2011-05-20 13:40:59 +02:00
msr.c BKL: remove extraneous #include <smp_lock.h> 2010-11-17 08:59:32 -08:00
paravirt-spinlocks.c
paravirt.c thp: add pmd paravirt ops 2011-01-13 17:32:39 -08:00
paravirt_patch_32.c
paravirt_patch_64.c
pci-calgary_64.c x86: Fix common misspellings 2011-03-18 10:39:30 +01:00
pci-dma.c x86/PCI: Remove dma32_reserve_bootmem 2011-05-10 15:43:32 -07:00
pci-iommu_table.c arch/x86/kernel/pci-iommu_table.c: Convert sprintf_symbol to %pS 2011-05-10 10:21:35 +02:00
pci-nommu.c include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h 2010-03-30 22:02:32 +09:00
pci-swiotlb.c x86, swiotlb: Make SWIOTLB use IOMMU_INIT_* macros. 2010-08-26 15:13:37 -07:00
pcspeaker.c
probe_roms.c x86: Introduce pci_map_biosrom() 2011-03-15 15:34:15 -07:00
process.c Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2011-06-01 02:07:22 +09:00
process_32.c cpuidle/x86/perf: fix power:cpu_idle double end events and throw cpu_idle events from the cpuidle layer 2011-01-12 18:05:16 -05:00
process_64.c x86: mark associated mm when running a task in 32 bit compatibility mode 2011-03-23 16:36:53 -04:00
ptrace.c x86: Get rid of asmregparm 2011-05-24 14:33:35 +02:00
pvclock.c x86/pvclock: Zero last_value on resume 2010-11-28 09:33:20 +01:00
quirks.c x86: HPET force enable for CX700 / VIA Epia LT 2010-09-15 16:27:04 +02:00
reboot.c x86: Reorder reboot method preferences 2011-04-06 10:36:50 +02:00
reboot_32.S x86, reboot: Fix relocations in reboot_32.S 2011-05-02 14:44:46 -07:00
reboot_fixups_32.c x86: Ce4100: Add reboot_fixup() for CE4100 2010-11-12 00:45:41 +01:00
relocate_kernel_32.S
relocate_kernel_64.S
resource.c x86: avoid high BIOS area when allocating address space 2010-12-17 10:01:30 -08:00
rtc.c rtc: cmos: Add OF bindings 2011-02-23 22:27:55 +01:00
setup.c Merge branch 'linus' into x86/urgent 2011-05-26 13:51:35 +02:00
setup_percpu.c x86: Unify CPU -> NUMA node mapping between 32 and 64bit 2011-01-28 14:54:09 +01:00
signal.c x86: signal: sys_rt_sigreturn() should use set_current_blocked() 2011-04-28 13:01:38 +02:00
smp.c sched: Provide scheduler_ipi() callback in response to smp_send_reschedule() 2011-04-14 08:52:32 +02:00
smpboot.c x86: cpu-hotplug: Prevent softirq wakeup on wrong CPU 2011-06-08 11:21:19 +02:00
stacktrace.c x86: Remove warning and warning_symbol from struct stacktrace_ops 2011-05-12 15:31:28 +02:00
step.c x86: Fix common misspellings 2011-03-18 10:39:30 +01:00
sys_i386_32.c i386: Make kernel_execve() suitable for stack unwinding 2010-09-03 08:16:02 +02:00
sys_x86_64.c improve sys_newuname() for compat architectures 2010-03-12 15:52:32 -08:00
syscall_64.c
syscall_table_32.S ns: Wire up the setns system call 2011-05-28 10:48:39 -07:00
tboot.c mm: convert mm->cpu_vm_cpumask into cpumask_var_t 2011-05-25 08:39:21 -07:00
tce_64.c
test_nx.c x86: Eliminate various 'set but not used' warnings 2011-05-21 19:10:33 +02:00
test_rodata.c
time.c x86-64: Clean up vdso/kernel shared variables 2011-05-24 14:51:28 +02:00
tls.c
tls.h
topology.c x86: Fix common misspellings 2011-03-18 10:39:30 +01:00
trampoline.c x86, trampoline: Common infrastructure for low memory trampolines 2011-02-17 21:02:43 -08:00
trampoline_32.S x86, trampoline: Common infrastructure for low memory trampolines 2011-02-17 21:02:43 -08:00
trampoline_64.S x86-64, trampoline: Remove unused variable 2011-02-18 15:50:36 -08:00
traps.c x86, NMI: Clean-up default_do_nmi() 2011-01-07 15:08:53 +01:00
tsc.c x86-64: Move vread_tsc into a new file with sensible options 2011-05-24 14:51:29 +02:00
tsc_sync.c
verify_cpu.S x86: Fix common misspellings 2011-03-18 10:39:30 +01:00
vm86_32.c thp: split_huge_page_mm/vma 2011-01-13 17:32:41 -08:00
vmlinux.lds.S Merge branch 'x86-vdso-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2011-05-26 12:19:31 -07:00
vread_tsc_64.c x86-64: Move vread_tsc into a new file with sensible options 2011-05-24 14:51:29 +02:00
vsmp_64.c
vsyscall_64.c Merge branch 'x86-vdso-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2011-05-26 12:19:31 -07:00
x86_init.c Merge branch 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2011-05-19 18:08:06 -07:00
x8664_ksyms_64.c x86-64, mem: Convert memmove() to assembly file and fix return value bug 2011-01-25 16:58:39 -08:00
xsave.c x86: Fix common misspellings 2011-03-18 10:39:30 +01:00