mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Avi Kivity <avi@redhat.com>
To: kvm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Subject: [PATCH 25/38] KVM: VMX: Pass cr0.mp through to the guest when the fpu is active
Date: Tue, 16 Feb 2010 12:35:23 +0200	[thread overview]
Message-ID: <1266316536-28936-26-git-send-email-avi@redhat.com> (raw)
In-Reply-To: <1266316536-28936-1-git-send-email-avi@redhat.com>

When cr0.mp is clear, the guest doesn't expect a #NM in response to
a WAIT instruction.  Because we always keep cr0.mp set, it will get
a #NM, and potentially be confused.

Fix by keeping cr0.mp set only when the fpu is inactive, and passing
it through when inactive.

Reported-by: Lorenzo Martignoni <martignlo@gmail.com>
Analyzed-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
---
 arch/x86/kvm/vmx.c |   15 +++++++++------
 1 files changed, 9 insertions(+), 6 deletions(-)

diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index a680d93..7a56879 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -66,7 +66,7 @@ module_param(emulate_invalid_guest_state, bool, S_IRUGO);
 #define KVM_GUEST_CR0_MASK						\
 	(KVM_GUEST_CR0_MASK_UNRESTRICTED_GUEST | X86_CR0_PG | X86_CR0_PE)
 #define KVM_VM_CR0_ALWAYS_ON_UNRESTRICTED_GUEST				\
-	(X86_CR0_WP | X86_CR0_NE | X86_CR0_MP)
+	(X86_CR0_WP | X86_CR0_NE)
 #define KVM_VM_CR0_ALWAYS_ON						\
 	(KVM_VM_CR0_ALWAYS_ON_UNRESTRICTED_GUEST | X86_CR0_PG | X86_CR0_PE)
 #define KVM_CR4_GUEST_OWNED_BITS				      \
@@ -791,12 +791,15 @@ static void vmx_vcpu_put(struct kvm_vcpu *vcpu)
 
 static void vmx_fpu_activate(struct kvm_vcpu *vcpu)
 {
+	ulong cr0;
+
 	if (vcpu->fpu_active)
 		return;
 	vcpu->fpu_active = 1;
-	vmcs_clear_bits(GUEST_CR0, X86_CR0_TS);
-	if (kvm_read_cr0_bits(vcpu, X86_CR0_TS))
-		vmcs_set_bits(GUEST_CR0, X86_CR0_TS);
+	cr0 = vmcs_readl(GUEST_CR0);
+	cr0 &= ~(X86_CR0_TS | X86_CR0_MP);
+	cr0 |= kvm_read_cr0_bits(vcpu, X86_CR0_TS | X86_CR0_MP);
+	vmcs_writel(GUEST_CR0, cr0);
 	update_exception_bitmap(vcpu);
 	vcpu->arch.cr0_guest_owned_bits = X86_CR0_TS;
 	vmcs_writel(CR0_GUEST_HOST_MASK, ~vcpu->arch.cr0_guest_owned_bits);
@@ -807,7 +810,7 @@ static void vmx_decache_cr0_guest_bits(struct kvm_vcpu *vcpu);
 static void vmx_fpu_deactivate(struct kvm_vcpu *vcpu)
 {
 	vmx_decache_cr0_guest_bits(vcpu);
-	vmcs_set_bits(GUEST_CR0, X86_CR0_TS);
+	vmcs_set_bits(GUEST_CR0, X86_CR0_TS | X86_CR0_MP);
 	update_exception_bitmap(vcpu);
 	vcpu->arch.cr0_guest_owned_bits = 0;
 	vmcs_writel(CR0_GUEST_HOST_MASK, ~vcpu->arch.cr0_guest_owned_bits);
@@ -1757,7 +1760,7 @@ static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
 		ept_update_paging_mode_cr0(&hw_cr0, cr0, vcpu);
 
 	if (!vcpu->fpu_active)
-		hw_cr0 |= X86_CR0_TS;
+		hw_cr0 |= X86_CR0_TS | X86_CR0_MP;
 
 	vmcs_writel(CR0_READ_SHADOW, cr0);
 	vmcs_writel(GUEST_CR0, hw_cr0);
-- 
1.6.5.3


  parent reply	other threads:[~2010-02-16 10:41 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-02-16 10:34 [PATCH 00/38] KVM updates for the 2.6.34 merge window (batch 3/4) Avi Kivity
2010-02-16 10:34 ` [PATCH 01/38] KVM: export <asm/hyperv.h> Avi Kivity
2010-02-16 10:35 ` [PATCH 02/38] KVM: VMX: Remove emulation failure report Avi Kivity
2010-02-16 10:35 ` [PATCH 03/38] KVM: fix Hyper-V hypercall warnings and wrong mask value Avi Kivity
2010-02-16 10:35 ` [PATCH 04/38] KVM: fix cleanup_srcu_struct on vm destruction Avi Kivity
2010-02-16 10:35 ` [PATCH 05/38] KVM: x86: Use macros for x86_emulate_ops to avoid future mistakes Avi Kivity
2010-02-16 10:35 ` [PATCH 06/38] KVM: VMX: Fix exceptions of mov to dr Avi Kivity
2010-02-16 10:35 ` [PATCH 07/38] KVM: VMX: Fix emulation of DR4 and DR5 Avi Kivity
2010-02-16 10:35 ` [PATCH 08/38] KVM: VMX: Clean up DR6 emulation Avi Kivity
2010-02-16 10:35 ` [PATCH 09/38] KVM: SVM: Clean up and enhance mov dr emulation Avi Kivity
2010-02-16 10:35 ` [PATCH 10/38] KVM: SVM: Trap all debug register accesses Avi Kivity
2010-02-16 10:35 ` [PATCH 11/38] KVM: Fix kvm_coalesced_mmio_ring duplicate allocation Avi Kivity
2010-02-16 10:35 ` [PATCH 12/38] KVM: x86: fix checking of cr0 validity Avi Kivity
2010-02-16 10:35 ` [PATCH 13/38] KVM: Allow kvm_load_guest_fpu() even when !vcpu->fpu_active Avi Kivity
2010-02-16 10:35 ` [PATCH 14/38] KVM: Drop kvm_{load,put}_guest_fpu() exports Avi Kivity
2010-02-16 10:35 ` [PATCH 15/38] KVM: Activate fpu on clts Avi Kivity
2010-02-16 10:35 ` [PATCH 16/38] KVM: Add a helper for checking if the guest is in protected mode Avi Kivity
2010-02-16 10:35 ` [PATCH 17/38] KVM: Move cr0/cr4/efer related helpers to x86.h Avi Kivity
2010-02-16 10:35 ` [PATCH 18/38] KVM: Rename vcpu->shadow_efer to efer Avi Kivity
2010-02-16 10:35 ` [PATCH 19/38] KVM: Optimize kvm_read_cr[04]_bits() Avi Kivity
2010-02-16 10:35 ` [PATCH 20/38] KVM: trace guest fpu loads and unloads Avi Kivity
2010-02-16 10:35 ` [PATCH 21/38] KVM: MMU: Remove some useless code from alloc_mmu_pages() Avi Kivity
2010-02-16 10:35 ` [PATCH 22/38] KVM: PPC E500: Add register l1csr0 emulation Avi Kivity
2010-02-16 10:35 ` [PATCH 23/38] KVM: PPC: Add PVR/PIR init for E500 Avi Kivity
2010-02-16 10:35 ` [PATCH 24/38] KVM: PPC E500: fix tlbcfg emulation Avi Kivity
2010-02-16 10:35 ` Avi Kivity [this message]
2010-02-16 10:35 ` [PATCH 26/38] KVM: mark segments accessed on HW task switch Avi Kivity
2010-02-16 10:35 ` [PATCH 27/38] KVM: Fix msr trace Avi Kivity
2010-02-16 10:35 ` [PATCH 28/38] KVM: Trace failed msr reads and writes Avi Kivity
2010-02-16 10:35 ` [PATCH 29/38] KVM: VMX: Remove redundant check in vm_need_virtualize_apic_accesses() Avi Kivity
2010-02-16 10:35 ` [PATCH 30/38] KVM: enable PCI multiple-segments for pass-through device Avi Kivity
2010-02-16 10:35 ` [PATCH 31/38] KVM: fix load_guest_segment_descriptor() to return X86EMUL_* Avi Kivity
2010-02-16 10:35 ` [PATCH 32/38] KVM: fix kvm_fix_hypercall() " Avi Kivity
2010-02-16 10:35 ` [PATCH 33/38] KVM: VMX: Wire up .fpu_activate() callback Avi Kivity
2010-02-16 10:35 ` [PATCH 34/38] KVM: ia64: Fix string literal continuation lines Avi Kivity
2010-02-16 10:35 ` [PATCH 35/38] KVM: VMX: Remove redundant test in vmx_set_efer() Avi Kivity
2010-02-16 10:35 ` [PATCH 36/38] KVM: Introduce kvm_host_page_size Avi Kivity
2010-02-16 10:35 ` [PATCH 37/38] KVM: VMX: emulate accessed bit for EPT Avi Kivity
2010-02-16 10:35 ` [PATCH 38/38] KVM: Remove redundant reading of rax on OUT instructions Avi Kivity

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1266316536-28936-26-git-send-email-avi@redhat.com \
    --to=avi@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

Powered by JetHome