[PATCH v4 08/21] kvm: x86: Check and enable permitted dynamic xfeatures at KVM_SET_CPUID2

[Yang Zhong <yang.zhong@xxxxxxxxx>]

From: Jing Liu <jing2.liu@xxxxxxxxx>

Guest xstate permissions should be set by userspace VMM before vcpu
creation. Extend KVM_SET_CPUID2 to verify that every feature reported
in CPUID[0xD] has proper permission set. If permission allows, enable
all xfeatures in guest cpuid with fpstate buffer sized accordingly.

This avoids introducing new KVM exit reason for reporting permission
violation to userspace VMM at run-time and also removes the need of
tricky fpstate buffer expansion in the emulation and restore path of
XCR0 and IA32_XFD MSR.

Signed-off-by: Jing Liu <jing2.liu@xxxxxxxxx>
Signed-off-by: Kevin Tian <kevin.tian@xxxxxxxxx>
Signed-off-by: Yang Zhong <yang.zhong@xxxxxxxxx>
---
 arch/x86/kvm/cpuid.c | 62 +++++++++++++++++++++++++++++++++-----------
 1 file changed, 47 insertions(+), 15 deletions(-)

diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index 4855344091b8..acbc10db550e 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -81,9 +81,12 @@ static inline struct kvm_cpuid_entry2 *cpuid_entry2_find(
 	return NULL;
 }
 
-static int kvm_check_cpuid(struct kvm_cpuid_entry2 *entries, int nent)
+static int kvm_check_cpuid(struct kvm_vcpu *vcpu,
+			   struct kvm_cpuid_entry2 *entries,
+			   int nent)
 {
 	struct kvm_cpuid_entry2 *best;
+	int r = 0;
 
 	/*
 	 * The existing code assumes virtual address is 48-bit or 57-bit in the
@@ -93,11 +96,40 @@ static int kvm_check_cpuid(struct kvm_cpuid_entry2 *entries, int nent)
 	if (best) {
 		int vaddr_bits = (best->eax & 0xff00) >> 8;
 
-		if (vaddr_bits != 48 && vaddr_bits != 57 && vaddr_bits != 0)
-			return -EINVAL;
+		if (vaddr_bits != 48 && vaddr_bits != 57 && vaddr_bits != 0) {
+			r = -EINVAL;
+			goto out;
+		}
 	}
 
-	return 0;
+	/*
+	 * Check guest permissions for dynamically-enabled xfeatures.
+	 * Userspace VMM is expected to acquire permission before vCPU
+	 * creation. If permission allows, enable all xfeatures with
+	 * fpstate buffer sized accordingly. This avoids complexity of
+	 * run-time expansion in the emulation and restore path of XCR0
+	 * and IA32_XFD MSR.
+	 */
+	best = cpuid_entry2_find(entries, nent, 0xd, 0);
+	if (best) {
+		u64 xfeatures;
+
+		xfeatures = best->eax | ((u64)best->edx << 32);
+		if (xfeatures & ~vcpu->arch.guest_fpu.perm) {
+			r = -ENXIO;
+			goto out;
+		}
+
+		if (xfeatures != vcpu->arch.guest_fpu.xfeatures) {
+			r = fpu_update_guest_perm_features(
+						&vcpu->arch.guest_fpu);
+			if (r)
+				goto out;
+		}
+	}
+
+out:
+	return r;
 }
 
 static void kvm_update_kvm_cpuid_base(struct kvm_vcpu *vcpu)
@@ -277,21 +309,21 @@ u64 kvm_vcpu_reserved_gpa_bits_raw(struct kvm_vcpu *vcpu)
 static int kvm_set_cpuid(struct kvm_vcpu *vcpu, struct kvm_cpuid_entry2 *e2,
                         int nent)
 {
-    int r;
+	int r;
 
-    r = kvm_check_cpuid(e2, nent);
-    if (r)
-        return r;
+	r = kvm_check_cpuid(vcpu, e2, nent);
+	if (r)
+		return r;
 
-    kvfree(vcpu->arch.cpuid_entries);
-    vcpu->arch.cpuid_entries = e2;
-    vcpu->arch.cpuid_nent = nent;
+	kvfree(vcpu->arch.cpuid_entries);
+	vcpu->arch.cpuid_entries = e2;
+	vcpu->arch.cpuid_nent = nent;
 
-    kvm_update_kvm_cpuid_base(vcpu);
-    kvm_update_cpuid_runtime(vcpu);
-    kvm_vcpu_after_set_cpuid(vcpu);
+	kvm_update_kvm_cpuid_base(vcpu);
+	kvm_update_cpuid_runtime(vcpu);
+	kvm_vcpu_after_set_cpuid(vcpu);
 
-    return 0;
+	return 0;
 }
 
 /* when an old userspace process fills a new kernel module */