linux/arch/x86/kvm
Ladi Prosek 6ed071f051 KVM: x86: fix emulation of RSM and IRET instructions
On AMD, the effect of set_nmi_mask called by emulate_iret_real and em_rsm
on hflags is reverted later on in x86_emulate_instruction where hflags are
overwritten with ctxt->emul_flags (the kvm_set_hflags call). This manifests
as a hang when rebooting Windows VMs with QEMU, OVMF, and >1 vcpu.

Instead of trying to merge ctxt->emul_flags into vcpu->arch.hflags after
an instruction is emulated, this commit deletes emul_flags altogether and
makes the emulator access vcpu->arch.hflags using two new accessors. This
way all changes, on the emulator side as well as in functions called from
the emulator and accessing vcpu state with emul_to_vcpu, are preserved.

More details on the bug and its manifestation with Windows and OVMF:

  It's a KVM bug in the interaction between SMI/SMM and NMI, specific to AMD.
  I believe that the SMM part explains why we started seeing this only with
  OVMF.

  KVM masks and unmasks NMI when entering and leaving SMM. When KVM emulates
  the RSM instruction in em_rsm, the set_nmi_mask call doesn't stick because
  later on in x86_emulate_instruction we overwrite arch.hflags with
  ctxt->emul_flags, effectively reverting the effect of the set_nmi_mask call.
  The AMD-specific hflag of interest here is HF_NMI_MASK.

  When rebooting the system, Windows sends an NMI IPI to all but the current
  cpu to shut them down. Only after all of them are parked in HLT will the
  initiating cpu finish the restart. If NMI is masked, other cpus never get
  the memo and the initiating cpu spins forever, waiting for
  hal!HalpInterruptProcessorsStarted to drop. That's the symptom we observe.

Fixes: a584539b24 ("KVM: x86: pass the whole hflags field to emulator and back")
Signed-off-by: Ladi Prosek <lprosek@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2017-04-27 16:54:09 +02:00
..
cpuid.c KVM: x86: virtualize cpuid faulting 2017-04-21 12:50:06 +02:00
cpuid.h KVM: x86: virtualize cpuid faulting 2017-04-21 12:50:06 +02:00
debugfs.c kvm: x86: export TSC information to user-space 2016-09-16 16:57:48 +02:00
emulate.c KVM: x86: fix emulation of RSM and IRET instructions 2017-04-27 16:54:09 +02:00
hyperv.c sched/headers: Prepare to move cputime functionality from <linux/sched.h> into <linux/sched/cputime.h> 2017-03-02 08:42:39 +01:00
hyperv.h KVM: x86: Hyper-V tsc page setup 2016-09-20 09:26:20 +02:00
i8254.c KVM: x86: Handle the kthread worker using the new API 2016-12-08 15:31:11 +01:00
i8254.h KVM: x86: Handle the kthread worker using the new API 2016-12-08 15:31:11 +01:00
i8259.c KVM: x86: simplify pic_ioport_read() 2017-04-12 20:17:15 +02:00
ioapic.c KVM: x86: rename kvm_vcpu_request_scan_ioapic() 2017-04-12 20:17:14 +02:00
ioapic.h KVM: x86: convert kvm_(set|get)_ioapic() into void 2017-04-12 20:17:14 +02:00
irq_comm.c KVM: x86: use irqchip_kernel() to check for pic+ioapic 2017-04-12 20:17:15 +02:00
irq.c KVM: x86: get rid of pic_irqchip() 2017-04-12 20:17:13 +02:00
irq.h KVM: x86: make kvm_pic_reset() static 2017-04-12 20:17:14 +02:00
Kconfig KVM: x86: drop legacy device assignment 2017-04-07 16:49:00 +02:00
kvm_cache_regs.h KVM, pkeys: add pkeys support for permission_fault 2016-03-22 16:23:37 +01:00
lapic.c KVM: Support vCPU-based gfn->hva cache 2017-02-16 18:42:46 +01:00
lapic.h KVM: x86: preparatory changes for APICv cleanups 2017-02-15 14:54:34 +01:00
Makefile KVM: x86: drop legacy device assignment 2017-04-07 16:49:00 +02:00
mmu_audit.c kvm: rename pfn_t to kvm_pfn_t 2016-01-15 17:56:32 -08:00
mmu.c kvm: nVMX: support EPT accessed/dirty bits 2017-04-07 16:49:00 +02:00
mmu.h kvm: nVMX: support EPT accessed/dirty bits 2017-04-07 16:49:00 +02:00
mmutrace.h tracing: Rename ftrace_event.h to trace_events.h 2015-05-13 14:05:12 -04:00
mtrr.c KVM: MTRR: fix kvm_mtrr_check_gfn_range_consistency page fault 2016-07-05 16:14:43 +02:00
page_track.c KVM: x86: cleanup the page tracking SRCU instance 2017-03-28 14:08:02 +02:00
paging_tmpl.h kvm: nVMX: support EPT accessed/dirty bits 2017-04-07 16:49:00 +02:00
pmu_amd.c perf/x86/amd: Make HW_CACHE_REFERENCES and HW_CACHE_MISSES measure L2 2016-09-16 16:19:49 +02:00
pmu_intel.c KVM: x86: Fix typos 2016-06-14 11:16:28 +02:00
pmu.c KVM: x86: never specify a sample period for virtualized in_tx_cp counters 2017-03-01 14:19:46 +01:00
pmu.h KVM: x86/vPMU: Define kvm_pmu_ops to support vPMU function dispatch 2015-06-23 14:12:14 +02:00
svm.c kvm: better MWAIT emulation for guests 2017-04-21 12:50:28 +02:00
trace.h KVM: x86: support using the vmx preemption timer for tsc deadline timer 2016-06-16 10:07:48 +02:00
tss.h
vmx.c KVM: add kvm_{test,clear}_request to replace {test,clear}_bit 2017-04-27 14:12:22 +02:00
x86.c KVM: x86: fix emulation of RSM and IRET instructions 2017-04-27 16:54:09 +02:00
x86.h kvm: better MWAIT emulation for guests 2017-04-21 12:50:28 +02:00