KVM: x86: SVM: Call x86_spec_ctrl_set_guest/host() with interrupts disabled

Mikhail reported the following lockdep splat:

WARNING: possible irq lock inversion dependency detected
CPU 0/KVM/10284 just changed the state of lock:
  000000000d538a88 (&st->lock){+...}, at:
  speculative_store_bypass_update+0x10b/0x170

but this lock was taken by another, HARDIRQ-safe lock
in the past:

(&(&sighand->siglock)->rlock){-.-.}

   and interrupts could create inverse lock ordering between them.

Possible interrupt unsafe locking scenario:

    CPU0                    CPU1
    ----                    ----
   lock(&st->lock);
                           local_irq_disable();
                           lock(&(&sighand->siglock)->rlock);
                           lock(&st->lock);
    <Interrupt>
     lock(&(&sighand->siglock)->rlock);
     *** DEADLOCK ***

The code path which connects those locks is:

   speculative_store_bypass_update()
   ssb_prctl_set()
   do_seccomp()
   do_syscall_64()

In svm_vcpu_run() speculative_store_bypass_update() is called with
interupts enabled via x86_virt_spec_ctrl_set_guest/host().

This is actually a false positive, because GIF=0 so interrupts are
disabled even if IF=1; however, we can easily move the invocations of
x86_virt_spec_ctrl_set_guest/host() into the interrupt disabled region to
cure it, and it's a good idea to keep the GIF=0/IF=1 area as small
and self-contained as possible.

Fixes: 1f50ddb4f418 ("x86/speculation: Handle HT correctly on AMD")
Reported-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Borislav Petkov <bp@suse.de>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: kvm@vger.kernel.org
Cc: x86@kernel.org
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 73e27a98456fbd9cccf3bc1a2c7982cedb1302f0..6276140044d0848b58e6778c31cf4d247520ac44 100644 (file)
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -5586,8 +5586,6 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu)
 
        clgi();
 
-       local_irq_enable();
-
        /*
         * If this vCPU has touched SPEC_CTRL, restore the guest's value if
         * it's non-zero. Since vmentry is serialising on affected CPUs, there
@@ -5596,6 +5594,8 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu)
         */
        x86_spec_ctrl_set_guest(svm->spec_ctrl, svm->virt_spec_ctrl);
 
+       local_irq_enable();
+
        asm volatile (
                "push %%" _ASM_BP "; \n\t"
                "mov %c[rbx](%[svm]), %%" _ASM_BX " \n\t"
@@ -5718,12 +5718,12 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu)
        if (unlikely(!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL)))
                svm->spec_ctrl = native_read_msr(MSR_IA32_SPEC_CTRL);
 
-       x86_spec_ctrl_restore_host(svm->spec_ctrl, svm->virt_spec_ctrl);
-
        reload_tss(vcpu);
 
        local_irq_disable();
 
+       x86_spec_ctrl_restore_host(svm->spec_ctrl, svm->virt_spec_ctrl);
+
        vcpu->arch.cr2 = svm->vmcb->save.cr2;
        vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax;
        vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp;