KVM: x86: Don't WARN if userspace mucks with RCX during string I/O exit

Replace a WARN with a comment to call out that userspace can modify RCX
during an exit to userspace to handle string I/O.  KVM doesn't actually
support changing the rep count during an exit, i.e. the scenario can be
ignored, but the WARN needs to go as it's trivial to trigger from
userspace.

Cc: stable@vger.kernel.org
Fixes: 3b27de271839 ("KVM: x86: split the two parts of emulator_pio_in")
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20211025201311.1881846-2-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index c473498..0cf1082 100644 (file)
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -7122,7 +7122,13 @@ static int emulator_pio_in(struct kvm_vcpu *vcpu, int size,
                           unsigned short port, void *val, unsigned int count)
 {
        if (vcpu->arch.pio.count) {
-               /* Complete previous iteration.  */
+               /*
+                * Complete a previous iteration that required userspace I/O.
+                * Note, @count isn't guaranteed to match pio.count as userspace
+                * can modify ECX before rerunning the vCPU.  Ignore any such
+                * shenanigans as KVM doesn't support modifying the rep count,
+                * and the emulator ensures @count doesn't overflow the buffer.
+                */
        } else {
                int r = __emulator_pio_in(vcpu, size, port, count);
                if (!r)
@@ -7131,7 +7137,6 @@ static int emulator_pio_in(struct kvm_vcpu *vcpu, int size,
                /* Results already available, fall through.  */
        }
 
-       WARN_ON(count != vcpu->arch.pio.count);
        complete_emulator_pio_in(vcpu, val);
        return 1;
 }