x86/HVM: don't cause #NM to be raised in Xen

The changes for XSA-267 did not touch management of CR0.TS for HVM
guests. In fully eager mode this bit should never be set when
respective vCPU-s are active, or else hvmemul_get_fpu() might leave it
wrongly set, leading to #NM in hypervisor context.

{svm,vmx}_enter() and {svm,vmx}_fpu_dirty_intercept() become unreachable
this way. Explicit {svm,vmx}_fpu_leave() invocations need to be guarded
now.

With no CR0.TS management necessary in fully eager mode, there's also no
need anymore to intercept #NM.

Reported-by: Charles Arnold <carnold@suse.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Release-acked-by: Juergen Gross <jgross@suse.com>
master commit: 488efc29e4e996bb3805c982200f65061390cdce
master date: 2018-06-28 09:07:06 +0200

diff --git a/xen/arch/x86/hvm/svm/svm.c b/xen/arch/x86/hvm/svm/svm.c
index 673a38c5746f4dc9ee446bded4124e9b2b483707..165500e3f2c2b683b046b48e46c7032074688674 100644 (file)
--- a/xen/arch/x86/hvm/svm/svm.c
+++ b/xen/arch/x86/hvm/svm/svm.c
@@ -574,7 +574,10 @@ void svm_update_guest_cr(struct vcpu *v, unsigned int cr, unsigned int flags)
         if ( !(v->arch.hvm_vcpu.guest_cr[0] & X86_CR0_TS) )
         {
             if ( v != current )
-                hw_cr0_mask |= X86_CR0_TS;
+            {
+                if ( !v->arch.fully_eager_fpu )
+                    hw_cr0_mask |= X86_CR0_TS;
+            }
             else if ( vmcb_get_cr0(vmcb) & X86_CR0_TS )
                 svm_fpu_enter(v);
         }
@@ -1083,7 +1086,8 @@ static void svm_ctxt_switch_from(struct vcpu *v)
     if ( unlikely((read_efer() & EFER_SVME) == 0) )
         return;
 
-    svm_fpu_leave(v);
+    if ( !v->arch.fully_eager_fpu )
+        svm_fpu_leave(v);
 
     svm_save_dr(v);
     svm_lwp_save(v);

diff --git a/xen/arch/x86/hvm/svm/vmcb.c b/xen/arch/x86/hvm/svm/vmcb.c
index f2d0f4c0c22f3349cda1475fcd795a03b1dad957..04518fd344359d19790d0f1a682492d826793ec2 100644 (file)
--- a/xen/arch/x86/hvm/svm/vmcb.c
+++ b/xen/arch/x86/hvm/svm/vmcb.c
@@ -180,8 +180,8 @@ static int construct_vmcb(struct vcpu *v)
     paging_update_paging_modes(v);
 
     vmcb->_exception_intercepts =
-        HVM_TRAP_MASK
-        | (1U << TRAP_no_device);
+        HVM_TRAP_MASK |
+        (v->arch.fully_eager_fpu ? 0 : (1U << TRAP_no_device));
 
     if ( paging_mode_hap(v->domain) )
     {

diff --git a/xen/arch/x86/hvm/vmx/vmcs.c b/xen/arch/x86/hvm/vmx/vmcs.c
index b783ff0d4541e3261c45fc1246d916459bc0a303..258fc08f728f8f377c474c0c5f6938db8d1b90d6 100644 (file)
--- a/xen/arch/x86/hvm/vmx/vmcs.c
+++ b/xen/arch/x86/hvm/vmx/vmcs.c
@@ -1144,7 +1144,9 @@ static int construct_vmcs(struct vcpu *v)
     __vmwrite(HOST_TR_SELECTOR, TSS_ENTRY << 3);
 
     /* Host control registers. */
-    v->arch.hvm_vmx.host_cr0 = read_cr0() | X86_CR0_TS;
+    v->arch.hvm_vmx.host_cr0 = read_cr0() & ~X86_CR0_TS;
+    if ( !v->arch.fully_eager_fpu )
+        v->arch.hvm_vmx.host_cr0 |= X86_CR0_TS;
     __vmwrite(HOST_CR0, v->arch.hvm_vmx.host_cr0);
     __vmwrite(HOST_CR4, mmu_cr4_features);
 
@@ -1223,7 +1225,7 @@ static int construct_vmcs(struct vcpu *v)
 
     v->arch.hvm_vmx.exception_bitmap = HVM_TRAP_MASK
               | (paging_mode_hap(d) ? 0 : (1U << TRAP_page_fault))
-              | (1U << TRAP_no_device);
+              | (v->arch.fully_eager_fpu ? 0 : (1U << TRAP_no_device));
     vmx_update_exception_bitmap(v);
 
     v->arch.hvm_vcpu.guest_cr[0] = X86_CR0_PE | X86_CR0_ET;

diff --git a/xen/arch/x86/hvm/vmx/vmx.c b/xen/arch/x86/hvm/vmx/vmx.c
index 970751494c52cbed4f40a7d83c879b7878c1ec7f..610c8d6eb9993700cb68e819931927e2b6e40034 100644 (file)
--- a/xen/arch/x86/hvm/vmx/vmx.c
+++ b/xen/arch/x86/hvm/vmx/vmx.c
@@ -936,7 +936,8 @@ static void vmx_ctxt_switch_from(struct vcpu *v)
         vmx_vmcs_reload(v);
     }
 
-    vmx_fpu_leave(v);
+    if ( !v->arch.fully_eager_fpu )
+        vmx_fpu_leave(v);
     vmx_save_guest_msrs(v);
     vmx_restore_host_msrs();
     vmx_save_dr(v);
@@ -1493,7 +1494,10 @@ static void vmx_update_guest_cr(struct vcpu *v, unsigned int cr,
         if ( !(v->arch.hvm_vcpu.guest_cr[0] & X86_CR0_TS) )
         {
             if ( v != current )
-                hw_cr0_mask |= X86_CR0_TS;
+            {
+                if ( !v->arch.fully_eager_fpu )
+                    hw_cr0_mask |= X86_CR0_TS;
+            }
             else if ( v->arch.hvm_vcpu.hw_cr[0] & X86_CR0_TS )
                 vmx_fpu_enter(v);
         }