kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] KVM: X86: Raise #GP when clearing CR0_PG in 64 bit mode
@ 2021-12-07  9:52 Lai Jiangshan
  2021-12-09 17:55 ` Paolo Bonzini
  0 siblings, 1 reply; 6+ messages in thread
From: Lai Jiangshan @ 2021-12-07  9:52 UTC (permalink / raw)
  To: linux-kernel, kvm, Paolo Bonzini
  Cc: Lai Jiangshan, Sean Christopherson, Vitaly Kuznetsov, Wanpeng Li,
	Jim Mattson, Joerg Roedel, Thomas Gleixner, Ingo Molnar,
	Borislav Petkov, Dave Hansen, x86, H. Peter Anvin

From: Lai Jiangshan <laijs@linux.alibaba.com>

In the SDM:
If the logical processor is in 64-bit mode or if CR4.PCIDE = 1, an
attempt to clear CR0.PG causes a general-protection exception (#GP).
Software should transition to compatibility mode and clear CR4.PCIDE
before attempting to disable paging.

Signed-off-by: Lai Jiangshan <laijs@linux.alibaba.com>
---
 arch/x86/kvm/x86.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 00f5b2b82909..78c40ac3b197 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -906,7 +906,8 @@ int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
 	    !load_pdptrs(vcpu, kvm_read_cr3(vcpu)))
 		return 1;
 
-	if (!(cr0 & X86_CR0_PG) && kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE))
+	if (!(cr0 & X86_CR0_PG) &&
+	    (is_64_bit_mode(vcpu) || kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE)))
 		return 1;
 
 	static_call(kvm_x86_set_cr0)(vcpu, cr0);
-- 
2.19.1.6.gb485710b


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH] KVM: X86: Raise #GP when clearing CR0_PG in 64 bit mode
  2021-12-07  9:52 [PATCH] KVM: X86: Raise #GP when clearing CR0_PG in 64 bit mode Lai Jiangshan
@ 2021-12-09 17:55 ` Paolo Bonzini
  2025-06-07  0:58   ` H. Peter Anvin
  0 siblings, 1 reply; 6+ messages in thread
From: Paolo Bonzini @ 2021-12-09 17:55 UTC (permalink / raw)
  To: Lai Jiangshan, linux-kernel, kvm
  Cc: Lai Jiangshan, Sean Christopherson, Vitaly Kuznetsov, Wanpeng Li,
	Jim Mattson, Joerg Roedel, Thomas Gleixner, Ingo Molnar,
	Borislav Petkov, Dave Hansen, x86, H. Peter Anvin

On 12/7/21 10:52, Lai Jiangshan wrote:
> From: Lai Jiangshan <laijs@linux.alibaba.com>
> 
> In the SDM:
> If the logical processor is in 64-bit mode or if CR4.PCIDE = 1, an
> attempt to clear CR0.PG causes a general-protection exception (#GP).
> Software should transition to compatibility mode and clear CR4.PCIDE
> before attempting to disable paging.
> 
> Signed-off-by: Lai Jiangshan <laijs@linux.alibaba.com>
> ---
>   arch/x86/kvm/x86.c | 3 ++-
>   1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 00f5b2b82909..78c40ac3b197 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -906,7 +906,8 @@ int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
>   	    !load_pdptrs(vcpu, kvm_read_cr3(vcpu)))
>   		return 1;
>   
> -	if (!(cr0 & X86_CR0_PG) && kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE))
> +	if (!(cr0 & X86_CR0_PG) &&
> +	    (is_64_bit_mode(vcpu) || kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE)))
>   		return 1;
>   
>   	static_call(kvm_x86_set_cr0)(vcpu, cr0);
> 

Queued, thanks.

Paolo

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] KVM: X86: Raise #GP when clearing CR0_PG in 64 bit mode
  2021-12-09 17:55 ` Paolo Bonzini
@ 2025-06-07  0:58   ` H. Peter Anvin
  2025-06-09 14:23     ` Sean Christopherson
  0 siblings, 1 reply; 6+ messages in thread
From: H. Peter Anvin @ 2025-06-07  0:58 UTC (permalink / raw)
  To: Paolo Bonzini, Lai Jiangshan, linux-kernel, kvm
  Cc: Lai Jiangshan, Sean Christopherson, Vitaly Kuznetsov, Wanpeng Li,
	Jim Mattson, Joerg Roedel, Thomas Gleixner, Ingo Molnar,
	Borislav Petkov, Dave Hansen, x86

On 2021-12-09 09:55, Paolo Bonzini wrote:
> On 12/7/21 10:52, Lai Jiangshan wrote:
>> From: Lai Jiangshan <laijs@linux.alibaba.com>
>>
>> In the SDM:
>> If the logical processor is in 64-bit mode or if CR4.PCIDE = 1, an
>> attempt to clear CR0.PG causes a general-protection exception (#GP).
>> Software should transition to compatibility mode and clear CR4.PCIDE
>> before attempting to disable paging.
>>
>> Signed-off-by: Lai Jiangshan <laijs@linux.alibaba.com>
>> ---
>>   arch/x86/kvm/x86.c | 3 ++-
>>   1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
>> index 00f5b2b82909..78c40ac3b197 100644
>> --- a/arch/x86/kvm/x86.c
>> +++ b/arch/x86/kvm/x86.c
>> @@ -906,7 +906,8 @@ int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned 
>> long cr0)
>>           !load_pdptrs(vcpu, kvm_read_cr3(vcpu)))
>>           return 1;
>> -    if (!(cr0 & X86_CR0_PG) && kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE))
>> +    if (!(cr0 & X86_CR0_PG) &&
>> +        (is_64_bit_mode(vcpu) || kvm_read_cr4_bits(vcpu, 
>> X86_CR4_PCIDE)))
>>           return 1;
>>       static_call(kvm_x86_set_cr0)(vcpu, cr0);
>>
> 
> Queued, thanks.
> 

Have you actually checked to see what real CPUs do in this case?

	-hpa


^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] KVM: X86: Raise #GP when clearing CR0_PG in 64 bit mode
  2025-06-07  0:58   ` H. Peter Anvin
@ 2025-06-09 14:23     ` Sean Christopherson
  2025-06-09 18:16       ` Jim Mattson
  0 siblings, 1 reply; 6+ messages in thread
From: Sean Christopherson @ 2025-06-09 14:23 UTC (permalink / raw)
  To: H. Peter Anvin
  Cc: Paolo Bonzini, Lai Jiangshan, linux-kernel, kvm, Lai Jiangshan,
	Vitaly Kuznetsov, Wanpeng Li, Jim Mattson, Joerg Roedel,
	Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, x86

On Fri, Jun 06, 2025, H. Peter Anvin wrote:
> On 2021-12-09 09:55, Paolo Bonzini wrote:
> > On 12/7/21 10:52, Lai Jiangshan wrote:
> > > From: Lai Jiangshan <laijs@linux.alibaba.com>
> > > 
> > > In the SDM:
> > > If the logical processor is in 64-bit mode or if CR4.PCIDE = 1, an
> > > attempt to clear CR0.PG causes a general-protection exception (#GP).
> > > Software should transition to compatibility mode and clear CR4.PCIDE
> > > before attempting to disable paging.
> > > 
> > > Signed-off-by: Lai Jiangshan <laijs@linux.alibaba.com>
> > > ---
> > >   arch/x86/kvm/x86.c | 3 ++-
> > >   1 file changed, 2 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> > > index 00f5b2b82909..78c40ac3b197 100644
> > > --- a/arch/x86/kvm/x86.c
> > > +++ b/arch/x86/kvm/x86.c
> > > @@ -906,7 +906,8 @@ int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned
> > > long cr0)
> > >           !load_pdptrs(vcpu, kvm_read_cr3(vcpu)))
> > >           return 1;
> > > -    if (!(cr0 & X86_CR0_PG) && kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE))
> > > +    if (!(cr0 & X86_CR0_PG) &&
> > > +        (is_64_bit_mode(vcpu) || kvm_read_cr4_bits(vcpu,
> > > X86_CR4_PCIDE)))
> > >           return 1;
> > >       static_call(kvm_x86_set_cr0)(vcpu, cr0);
> > > 
> > 
> > Queued, thanks.
> > 
> 
> Have you actually checked to see what real CPUs do in this case?

I have now, and EMR at least behaves as the SDM describes.  Why do you ask?


kvm_intel: Clearing CR0.PG faulted (vector = 13)


diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index f79604bc0127..f90ad464ab7e 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -8637,6 +8637,23 @@ void vmx_exit(void)
        kvm_x86_vendor_exit();
 }
 
+static noinline void vmx_disable_paging(void)
+{
+       unsigned long cr0 = native_read_cr0();
+       long vector = -1;
+
+       asm volatile("1: mov %1, %%cr0\n\t"
+                    "   mov %2, %%cr0\n\t"
+                    "2:"
+                    _ASM_EXTABLE_FAULT(1b, 2b)
+                    : "+a" (vector)
+                    : "r" (cr0 & ~X86_CR0_PG), "r" (cr0)
+                    : "cc", "memory" );
+
+       pr_warn("Clearing CR0.PG %s (vector = %ld)\n",
+               vector < 0 ? "succeeded" : "faulted", vector);
+}
+
 int __init vmx_init(void)
 {
        int r, cpu;
@@ -8644,6 +8661,8 @@ int __init vmx_init(void)
        if (!kvm_is_vmx_supported())
                return -EOPNOTSUPP;
 
+       vmx_disable_paging();
+
        /*
         * Note, hv_init_evmcs() touches only VMX knobs, i.e. there's nothing
         * to unwind if a later step fails.


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH] KVM: X86: Raise #GP when clearing CR0_PG in 64 bit mode
  2025-06-09 14:23     ` Sean Christopherson
@ 2025-06-09 18:16       ` Jim Mattson
  2025-06-09 18:20         ` Jim Mattson
  0 siblings, 1 reply; 6+ messages in thread
From: Jim Mattson @ 2025-06-09 18:16 UTC (permalink / raw)
  To: Sean Christopherson
  Cc: H. Peter Anvin, Paolo Bonzini, Lai Jiangshan, linux-kernel, kvm,
	Lai Jiangshan, Vitaly Kuznetsov, Wanpeng Li, Joerg Roedel,
	Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, x86

On Mon, Jun 9, 2025 at 7:23 AM Sean Christopherson <seanjc@google.com> wrote:
>
> On Fri, Jun 06, 2025, H. Peter Anvin wrote:
> > On 2021-12-09 09:55, Paolo Bonzini wrote:
> > > On 12/7/21 10:52, Lai Jiangshan wrote:
> > > > From: Lai Jiangshan <laijs@linux.alibaba.com>
> > > >
> > > > In the SDM:
> > > > If the logical processor is in 64-bit mode or if CR4.PCIDE = 1, an
> > > > attempt to clear CR0.PG causes a general-protection exception (#GP).
> > > > Software should transition to compatibility mode and clear CR4.PCIDE
> > > > before attempting to disable paging.
> > > >
> > > > Signed-off-by: Lai Jiangshan <laijs@linux.alibaba.com>
> > > > ---
> > > >   arch/x86/kvm/x86.c | 3 ++-
> > > >   1 file changed, 2 insertions(+), 1 deletion(-)
> > > >
> > > > diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> > > > index 00f5b2b82909..78c40ac3b197 100644
> > > > --- a/arch/x86/kvm/x86.c
> > > > +++ b/arch/x86/kvm/x86.c
> > > > @@ -906,7 +906,8 @@ int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned
> > > > long cr0)
> > > >           !load_pdptrs(vcpu, kvm_read_cr3(vcpu)))
> > > >           return 1;
> > > > -    if (!(cr0 & X86_CR0_PG) && kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE))
> > > > +    if (!(cr0 & X86_CR0_PG) &&
> > > > +        (is_64_bit_mode(vcpu) || kvm_read_cr4_bits(vcpu,
> > > > X86_CR4_PCIDE)))
> > > >           return 1;
> > > >       static_call(kvm_x86_set_cr0)(vcpu, cr0);
> > > >

Isn't this redundant with the "if (cs_l)" check above?

> > > Queued, thanks.
> > >
> >
> > Have you actually checked to see what real CPUs do in this case?
>
> I have now, and EMR at least behaves as the SDM describes.  Why do you ask?
>
>
> kvm_intel: Clearing CR0.PG faulted (vector = 13)
>
>
> diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
> index f79604bc0127..f90ad464ab7e 100644
> --- a/arch/x86/kvm/vmx/vmx.c
> +++ b/arch/x86/kvm/vmx/vmx.c
> @@ -8637,6 +8637,23 @@ void vmx_exit(void)
>         kvm_x86_vendor_exit();
>  }
>
> +static noinline void vmx_disable_paging(void)
> +{
> +       unsigned long cr0 = native_read_cr0();
> +       long vector = -1;
> +
> +       asm volatile("1: mov %1, %%cr0\n\t"
> +                    "   mov %2, %%cr0\n\t"
> +                    "2:"
> +                    _ASM_EXTABLE_FAULT(1b, 2b)
> +                    : "+a" (vector)
> +                    : "r" (cr0 & ~X86_CR0_PG), "r" (cr0)
> +                    : "cc", "memory" );
> +
> +       pr_warn("Clearing CR0.PG %s (vector = %ld)\n",
> +               vector < 0 ? "succeeded" : "faulted", vector);
> +}
> +
>  int __init vmx_init(void)
>  {
>         int r, cpu;
> @@ -8644,6 +8661,8 @@ int __init vmx_init(void)
>         if (!kvm_is_vmx_supported())
>                 return -EOPNOTSUPP;
>
> +       vmx_disable_paging();
> +
>         /*
>          * Note, hv_init_evmcs() touches only VMX knobs, i.e. there's nothing
>          * to unwind if a later step fails.
>

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] KVM: X86: Raise #GP when clearing CR0_PG in 64 bit mode
  2025-06-09 18:16       ` Jim Mattson
@ 2025-06-09 18:20         ` Jim Mattson
  0 siblings, 0 replies; 6+ messages in thread
From: Jim Mattson @ 2025-06-09 18:20 UTC (permalink / raw)
  To: Sean Christopherson
  Cc: H. Peter Anvin, Paolo Bonzini, Lai Jiangshan, linux-kernel, kvm,
	Lai Jiangshan, Vitaly Kuznetsov, Wanpeng Li, Joerg Roedel,
	Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, x86

On Mon, Jun 9, 2025 at 11:16 AM Jim Mattson <jmattson@google.com> wrote:
>
> On Mon, Jun 9, 2025 at 7:23 AM Sean Christopherson <seanjc@google.com> wrote:
> >
> > On Fri, Jun 06, 2025, H. Peter Anvin wrote:
> > > On 2021-12-09 09:55, Paolo Bonzini wrote:
> > > > On 12/7/21 10:52, Lai Jiangshan wrote:
> > > > > From: Lai Jiangshan <laijs@linux.alibaba.com>
> > > > >
> > > > > In the SDM:
> > > > > If the logical processor is in 64-bit mode or if CR4.PCIDE = 1, an
> > > > > attempt to clear CR0.PG causes a general-protection exception (#GP).
> > > > > Software should transition to compatibility mode and clear CR4.PCIDE
> > > > > before attempting to disable paging.
> > > > >
> > > > > Signed-off-by: Lai Jiangshan <laijs@linux.alibaba.com>
> > > > > ---
> > > > >   arch/x86/kvm/x86.c | 3 ++-
> > > > >   1 file changed, 2 insertions(+), 1 deletion(-)
> > > > >
> > > > > diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> > > > > index 00f5b2b82909..78c40ac3b197 100644
> > > > > --- a/arch/x86/kvm/x86.c
> > > > > +++ b/arch/x86/kvm/x86.c
> > > > > @@ -906,7 +906,8 @@ int kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned
> > > > > long cr0)
> > > > >           !load_pdptrs(vcpu, kvm_read_cr3(vcpu)))
> > > > >           return 1;
> > > > > -    if (!(cr0 & X86_CR0_PG) && kvm_read_cr4_bits(vcpu, X86_CR4_PCIDE))
> > > > > +    if (!(cr0 & X86_CR0_PG) &&
> > > > > +        (is_64_bit_mode(vcpu) || kvm_read_cr4_bits(vcpu,
> > > > > X86_CR4_PCIDE)))
> > > > >           return 1;
> > > > >       static_call(kvm_x86_set_cr0)(vcpu, cr0);
> > > > >
>
> Isn't this redundant with the "if (cs_l)" check above?

Never mind. That's an attempt to set CR0.PG, not to clear it.

> > > > Queued, thanks.
> > > >
> > >
> > > Have you actually checked to see what real CPUs do in this case?
> >
> > I have now, and EMR at least behaves as the SDM describes.  Why do you ask?
> >
> >
> > kvm_intel: Clearing CR0.PG faulted (vector = 13)
> >
> >
> > diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
> > index f79604bc0127..f90ad464ab7e 100644
> > --- a/arch/x86/kvm/vmx/vmx.c
> > +++ b/arch/x86/kvm/vmx/vmx.c
> > @@ -8637,6 +8637,23 @@ void vmx_exit(void)
> >         kvm_x86_vendor_exit();
> >  }
> >
> > +static noinline void vmx_disable_paging(void)
> > +{
> > +       unsigned long cr0 = native_read_cr0();
> > +       long vector = -1;
> > +
> > +       asm volatile("1: mov %1, %%cr0\n\t"
> > +                    "   mov %2, %%cr0\n\t"
> > +                    "2:"
> > +                    _ASM_EXTABLE_FAULT(1b, 2b)
> > +                    : "+a" (vector)
> > +                    : "r" (cr0 & ~X86_CR0_PG), "r" (cr0)
> > +                    : "cc", "memory" );
> > +
> > +       pr_warn("Clearing CR0.PG %s (vector = %ld)\n",
> > +               vector < 0 ? "succeeded" : "faulted", vector);
> > +}
> > +
> >  int __init vmx_init(void)
> >  {
> >         int r, cpu;
> > @@ -8644,6 +8661,8 @@ int __init vmx_init(void)
> >         if (!kvm_is_vmx_supported())
> >                 return -EOPNOTSUPP;
> >
> > +       vmx_disable_paging();
> > +
> >         /*
> >          * Note, hv_init_evmcs() touches only VMX knobs, i.e. there's nothing
> >          * to unwind if a later step fails.
> >

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2025-06-09 18:20 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2021-12-07  9:52 [PATCH] KVM: X86: Raise #GP when clearing CR0_PG in 64 bit mode Lai Jiangshan
2021-12-09 17:55 ` Paolo Bonzini
2025-06-07  0:58   ` H. Peter Anvin
2025-06-09 14:23     ` Sean Christopherson
2025-06-09 18:16       ` Jim Mattson
2025-06-09 18:20         ` Jim Mattson

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).