Fix reboot on non-preemptible kernels
cover Fix reboot on non-preemptible kernels
Reboot with guests running, on Intel hosts, with a non-preemptible host kernel is broken. This patchset fixes the issue.
Avi Kivity (2):
KVM: Don't spin on virt instruction faults during reboot
KVM: VMX: Return 0 from a failed VMREAD
arch/x86/include/asm/kvm_host.h | 8 ++++++-- arch/x86/kvm/vmx.c | 4 ++-- virt/kvm/kvm_main.c | 13 ++++--------- 3 files changed, 12 insertions(+), 13 deletions(-)
PATCH 1/2 KVM: Don't spin on virt instruction faults during reboot
Since vmx blocks INIT signals, we disable virtualization extensions during reboot. This leads to virtualization instructions faulting; we trap these faults and spin while the reboot continues.
Unfortunately spinning on a non-preemptible kernel may block a task that reboot depends on; this causes the reboot to hang.
Fix by skipping over the instruction and hoping for the best.
--- arch/x86/include/asm/kvm_host.h | 8 ++++++-- virt/kvm/kvm_main.c | 13 ++++--------- 2 files changed, 10 insertions(+), 11 deletions(-)
arch/x86/include/asm/kvm_host.h +6 -2
@@ -790,14 +790,18 @@ enum {* reboot turns off virtualization while processes are running. * Trap the fault and ignore the instruction if that happens. */ -asmlinkage void kvm_handle_fault_on_reboot(void); +asmlinkage void kvm_spurious_fault(void); +extern bool kvm_rebooting;#define __kvm_handle_fault_on_reboot (insn) \ "666: " insn "\n\t" \ + "668: \n\t" \ ".pushsection .fixup, \"ax\" \n" \ "667: \n\t" \ + "cmpb $0, kvm_rebooting \n\t" \ + "jne 668b \n\t" \ __ASM_SIZE(push) " $666b \n\t" \ - "jmp kvm_handle_fault_on_reboot \n\t" \ + "call kvm_spurious_fault \n\t" \ ".popsection \n\t" \ ".pushsection __ex_table, \"a\" \n\t" \ _ASM_PTR " 666b, 667b \n\t" \
virt/kvm/kvm_main.c +4 -9
@@ -90,7 +90,8 @@ static void hardware_disable_all(void);static void kvm_io_bus_destroy (struct kvm_io_bus *bus); -static bool kvm_rebooting; +bool kvm_rebooting; +EXPORT_SYMBOL_GPL (kvm_rebooting);static bool largepages_enabled = true;
@@ -2179,18 +2180,12 @@ static int kvm_cpu_hotplug(struct notifier_block *notifier, unsigned long val, } -asmlinkage void kvm_handle_fault_on_reboot(void) +asmlinkage void kvm_spurious_fault(void) { -if (kvm_rebooting) { -/* spin while reset goes on */ -local_irq_enable (); -while (true) -cpu_relax (); - }/* Fault while not rebooting. We want the trace. */ BUG (); } -EXPORT_SYMBOL_GPL (kvm_handle_fault_on_reboot); +EXPORT_SYMBOL_GPL (kvm_spurious_fault);static int kvm_reboot (struct notifier_block *notifier,unsigned long val,void *v)
Signature · 1 line
1.7.1
PATCH 2/2 KVM: VMX: Return 0 from a failed VMREAD
In reply to Avi KivityIf we execute VMREAD during reboot we'll just skip over it. Instead of returning garbage, return 0, which has a much smaller chance of confusing the code. Otherwise we risk a flood of debug printk()s which block the reboot process if a serial console or netconsole is enabled.
--- arch/x86/kvm/vmx.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)
arch/x86/kvm/vmx.c +2 -2
@@ -565,10 +565,10 @@ static inline void ept_sync_individual_addr(u64 eptp, gpa_t gpa)static unsigned long vmcs_readl (unsigned long field) { -unsigned long value; +unsigned long value =0 ; asm volatile (__ex(ASM_VMX_VMREAD_RDX_RAX) - : "=a"(value) : "d"(field) : "cc"); + : "+a"(value) : "d"(field) : "cc"); return value; }
Signature · 1 line
1.7.1