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>
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 451a5b492828230c441b136d4d88619d6887c997..ca16091e8fcbee4acf692dc02835c9e0fcd79b46 100644 (file)
--- a/xen/arch/x86/hvm/svm/svm.c
+++ b/xen/arch/x86/hvm/svm/svm.c
@@ -540,7 +540,10 @@ void svm_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 ( vmcb_get_cr0(vmcb) & X86_CR0_TS )
                 svm_fpu_enter(v);
         }
@@ -1029,7 +1032,8 @@ 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 f982fc9faeaf740192af2db3bb86fe1aeaa43f28..65980d686219ef649f9cd40b87288bd29f0b500c 100644 (file)
--- a/xen/arch/x86/hvm/svm/vmcb.c
+++ b/xen/arch/x86/hvm/svm/vmcb.c
@@ -194,8 +194,8 @@ 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 98a601f7fda3aebaa01a7bdecc3865471ca6e4c5..f6adb302c39c40d17724e0d33ca27d7f7c39b549 100644 (file)
--- a/xen/arch/x86/hvm/vmx/vmcs.c
+++ b/xen/arch/x86/hvm/vmx/vmcs.c
@@ -1154,7 +1154,9 @@ static int construct_vmcs(struct vcpu *v)
     __vmwrite(HOST_GS_BASE, 0);
 
     /* 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);
 
@@ -1238,7 +1240,7 @@ 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);
 
     /*

diff --git a/xen/arch/x86/hvm/vmx/vmx.c b/xen/arch/x86/hvm/vmx/vmx.c
index 955643c036ff660c39d9976e13077146fbf857df..1949ce3b734d0a2a27270c47f78ad790d0e963cc 100644 (file)
--- a/xen/arch/x86/hvm/vmx/vmx.c
+++ b/xen/arch/x86/hvm/vmx/vmx.c
@@ -961,7 +961,8 @@ 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);
@@ -1550,7 +1551,10 @@ 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);
         }