* Re: [PATCH v3 2/2] KVM: X86: Add Force Emulation Prefix for "emulate the next instruction"
@ 2018-03-28 6:55 Liran Alon
0 siblings, 0 replies; 4+ messages in thread
From: Liran Alon @ 2018-03-28 6:55 UTC (permalink / raw)
To: kernellwp
Cc: konrad.wilk, rkrcmar, pbonzini, linux-kernel, andrew.cooper3, kvm
----- kernellwp@gmail.com wrote:
> From: Wanpeng Li <wanpengli@tencent.com>
>
> There is no easy way to force KVM to run an instruction through the
> emulator
> (by design as that will expose the x86 emulator as a significant
> attack-surface).
> However, we do wish to expose the x86 emulator in case we are testing
> it
> (e.g. via kvm-unit-tests). Therefore, this patch adds a "force
> emulation prefix"
> that is designed to raise #UD which KVM will trap and it's #UD
> exit-handler will
> match "force emulation prefix" to run instruction after prefix by the
> x86 emulator.
> To not expose the x86 emulator by default, we add a module parameter
> that should
> be off by default.
>
> A simple testcase here:
>
> #include <stdio.h>
> #include <string.h>
>
> #define HYPERVISOR_INFO 0x40000000
>
> #define CPUID(idx, eax, ebx, ecx, edx) \
> asm volatile (\
> "ud2a; .ascii \"kvm\"; cpuid" \
> :"=b" (*ebx), "=a" (*eax), "=c" (*ecx), "=d" (*edx) \
> :"0"(idx) );
>
> void main()
> {
> unsigned int eax, ebx, ecx, edx;
> char string[13];
>
> CPUID(HYPERVISOR_INFO, &eax, &ebx, &ecx, &edx);
> *(unsigned int *)(string + 0) = ebx;
> *(unsigned int *)(string + 4) = ecx;
> *(unsigned int *)(string + 8) = edx;
>
> string[12] = 0;
> if (strncmp(string, "KVMKVMKVM\0\0\0", 12) == 0)
> printf("kvm guest\n");
> else
> printf("bare hardware\n");
> }
>
> Suggested-by: Andrew Cooper <andrew.cooper3@citrix.com>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Radim Krčmář <rkrcmar@redhat.com>
> Cc: Andrew Cooper <andrew.cooper3@citrix.com>
> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
> Cc: Liran Alon <liran.alon@oracle.com>
> Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
> ---
> arch/x86/kvm/x86.c | 18 +++++++++++++++++-
> 1 file changed, 17 insertions(+), 1 deletion(-)
>
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index e3a60ab..40e2f78 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -146,6 +146,9 @@ bool __read_mostly enable_vmware_backdoor =
> false;
> module_param(enable_vmware_backdoor, bool, S_IRUGO);
> EXPORT_SYMBOL_GPL(enable_vmware_backdoor);
>
> +static bool __read_mostly force_emulation_prefix = false;
> +module_param(force_emulation_prefix, bool, S_IRUGO);
> +
> #define KVM_NR_SHARED_MSRS 16
>
> struct kvm_shared_msrs_global {
> @@ -4843,8 +4846,21 @@
> EXPORT_SYMBOL_GPL(kvm_write_guest_virt_system);
> int handle_ud(struct kvm_vcpu *vcpu)
> {
> enum emulation_result er;
> + int emulation_type = EMULTYPE_TRAP_UD;
> +
> + if (force_emulation_prefix) {
> + char sig[5]; /* ud2; .ascii "kvm" */
> + struct x86_exception e;
> +
> + kvm_read_guest_virt(&vcpu->arch.emulate_ctxt,
> + kvm_get_linear_rip(vcpu), sig, sizeof(sig), &e);
> + if (memcmp(sig, "\xf\xbkvm", sizeof(sig)) == 0) {
> + emulation_type = 0;
> + kvm_rip_write(vcpu, kvm_rip_read(vcpu) + sizeof(sig));
> + }
> + }
>
> - er = emulate_instruction(vcpu, EMULTYPE_TRAP_UD);
> + er = emulate_instruction(vcpu, emulation_type);
> if (er == EMULATE_USER_EXIT)
> return 0;
> if (er != EMULATE_DONE)
> --
> 2.7.4
Reviewed-by: Liran Alon <liran.alon@oracle.com>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH v3 2/2] KVM: X86: Add Force Emulation Prefix for "emulate the next instruction"
2018-03-29 21:29 ` Radim Krčmář
@ 2018-04-01 11:12 ` Wanpeng Li
0 siblings, 0 replies; 4+ messages in thread
From: Wanpeng Li @ 2018-04-01 11:12 UTC (permalink / raw)
To: Radim Krčmář
Cc: LKML, kvm, Paolo Bonzini, Andrew Cooper, Konrad Rzeszutek Wilk,
Liran Alon
2018-03-30 5:29 GMT+08:00 Radim Krčmář <rkrcmar@redhat.com>:
> 2018-03-27 17:48-0700, Wanpeng Li:
>> From: Wanpeng Li <wanpengli@tencent.com>
>>
>> There is no easy way to force KVM to run an instruction through the emulator
>> (by design as that will expose the x86 emulator as a significant attack-surface).
>> However, we do wish to expose the x86 emulator in case we are testing it
>> (e.g. via kvm-unit-tests). Therefore, this patch adds a "force emulation prefix"
>> that is designed to raise #UD which KVM will trap and it's #UD exit-handler will
>> match "force emulation prefix" to run instruction after prefix by the x86 emulator.
>> To not expose the x86 emulator by default, we add a module parameter that should
>> be off by default.
>>
>> A simple testcase here:
>>
>> #include <stdio.h>
>> #include <string.h>
>>
>> #define HYPERVISOR_INFO 0x40000000
>>
>> #define CPUID(idx, eax, ebx, ecx, edx) \
>> asm volatile (\
>> "ud2a; .ascii \"kvm\"; cpuid" \
>> :"=b" (*ebx), "=a" (*eax), "=c" (*ecx), "=d" (*edx) \
>> :"0"(idx) );
>>
>> void main()
>> {
>> unsigned int eax, ebx, ecx, edx;
>> char string[13];
>>
>> CPUID(HYPERVISOR_INFO, &eax, &ebx, &ecx, &edx);
>> *(unsigned int *)(string + 0) = ebx;
>> *(unsigned int *)(string + 4) = ecx;
>> *(unsigned int *)(string + 8) = edx;
>>
>> string[12] = 0;
>> if (strncmp(string, "KVMKVMKVM\0\0\0", 12) == 0)
>> printf("kvm guest\n");
>> else
>> printf("bare hardware\n");
>> }
>>
>> Suggested-by: Andrew Cooper <andrew.cooper3@citrix.com>
>> Cc: Paolo Bonzini <pbonzini@redhat.com>
>> Cc: Radim Krčmář <rkrcmar@redhat.com>
>> Cc: Andrew Cooper <andrew.cooper3@citrix.com>
>> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
>> Cc: Liran Alon <liran.alon@oracle.com>
>> Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
>> ---
>> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
>> @@ -146,6 +146,9 @@ bool __read_mostly enable_vmware_backdoor = false;
>> module_param(enable_vmware_backdoor, bool, S_IRUGO);
>> EXPORT_SYMBOL_GPL(enable_vmware_backdoor);
>>
>> +static bool __read_mostly force_emulation_prefix = false;
>> +module_param(force_emulation_prefix, bool, S_IRUGO);
>> +
>> #define KVM_NR_SHARED_MSRS 16
>>
>> struct kvm_shared_msrs_global {
>> @@ -4843,8 +4846,21 @@ EXPORT_SYMBOL_GPL(kvm_write_guest_virt_system);
>> int handle_ud(struct kvm_vcpu *vcpu)
>> {
>> enum emulation_result er;
>> + int emulation_type = EMULTYPE_TRAP_UD;
>> +
>> + if (force_emulation_prefix) {
>> + char sig[5]; /* ud2; .ascii "kvm" */
>> + struct x86_exception e;
>> +
>> + kvm_read_guest_virt(&vcpu->arch.emulate_ctxt,
>> + kvm_get_linear_rip(vcpu), sig, sizeof(sig), &e);
>> + if (memcmp(sig, "\xf\xbkvm", sizeof(sig)) == 0) {
>> + emulation_type = 0;
>> + kvm_rip_write(vcpu, kvm_rip_read(vcpu) + sizeof(sig));
>> + }
>> + }
>>
>> - er = emulate_instruction(vcpu, EMULTYPE_TRAP_UD);
>> + er = emulate_instruction(vcpu, emulation_type);
>> if (er == EMULATE_USER_EXIT)
>> return 0;
>> if (er != EMULATE_DONE)
>
> The code afterwards is going to inject an #UD if the emulation failed.
> I think that preserving the cpu state and forwarding the emulation
> failure to userspace would be more useful. The change would probably be
> best as:
>
> if (memcmp(sig, "\xf\xbkvm", sizeof(sig)) == 0) {
> kvm_rip_write(vcpu, kvm_rip_read(vcpu) + sizeof(sig));
> return emulate_instruction(vcpu, 0) == EMULATE_DONE;
> }
>
> Looks great otherwise, thanks.
Do it in v4. :)
Regards,
Wanpeng Li
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH v3 2/2] KVM: X86: Add Force Emulation Prefix for "emulate the next instruction"
2018-03-28 0:48 ` [PATCH v3 2/2] " Wanpeng Li
@ 2018-03-29 21:29 ` Radim Krčmář
2018-04-01 11:12 ` Wanpeng Li
0 siblings, 1 reply; 4+ messages in thread
From: Radim Krčmář @ 2018-03-29 21:29 UTC (permalink / raw)
To: Wanpeng Li
Cc: linux-kernel, kvm, Paolo Bonzini, Andrew Cooper,
Konrad Rzeszutek Wilk, Liran Alon
2018-03-27 17:48-0700, Wanpeng Li:
> From: Wanpeng Li <wanpengli@tencent.com>
>
> There is no easy way to force KVM to run an instruction through the emulator
> (by design as that will expose the x86 emulator as a significant attack-surface).
> However, we do wish to expose the x86 emulator in case we are testing it
> (e.g. via kvm-unit-tests). Therefore, this patch adds a "force emulation prefix"
> that is designed to raise #UD which KVM will trap and it's #UD exit-handler will
> match "force emulation prefix" to run instruction after prefix by the x86 emulator.
> To not expose the x86 emulator by default, we add a module parameter that should
> be off by default.
>
> A simple testcase here:
>
> #include <stdio.h>
> #include <string.h>
>
> #define HYPERVISOR_INFO 0x40000000
>
> #define CPUID(idx, eax, ebx, ecx, edx) \
> asm volatile (\
> "ud2a; .ascii \"kvm\"; cpuid" \
> :"=b" (*ebx), "=a" (*eax), "=c" (*ecx), "=d" (*edx) \
> :"0"(idx) );
>
> void main()
> {
> unsigned int eax, ebx, ecx, edx;
> char string[13];
>
> CPUID(HYPERVISOR_INFO, &eax, &ebx, &ecx, &edx);
> *(unsigned int *)(string + 0) = ebx;
> *(unsigned int *)(string + 4) = ecx;
> *(unsigned int *)(string + 8) = edx;
>
> string[12] = 0;
> if (strncmp(string, "KVMKVMKVM\0\0\0", 12) == 0)
> printf("kvm guest\n");
> else
> printf("bare hardware\n");
> }
>
> Suggested-by: Andrew Cooper <andrew.cooper3@citrix.com>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Radim Krčmář <rkrcmar@redhat.com>
> Cc: Andrew Cooper <andrew.cooper3@citrix.com>
> Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
> Cc: Liran Alon <liran.alon@oracle.com>
> Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
> ---
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> @@ -146,6 +146,9 @@ bool __read_mostly enable_vmware_backdoor = false;
> module_param(enable_vmware_backdoor, bool, S_IRUGO);
> EXPORT_SYMBOL_GPL(enable_vmware_backdoor);
>
> +static bool __read_mostly force_emulation_prefix = false;
> +module_param(force_emulation_prefix, bool, S_IRUGO);
> +
> #define KVM_NR_SHARED_MSRS 16
>
> struct kvm_shared_msrs_global {
> @@ -4843,8 +4846,21 @@ EXPORT_SYMBOL_GPL(kvm_write_guest_virt_system);
> int handle_ud(struct kvm_vcpu *vcpu)
> {
> enum emulation_result er;
> + int emulation_type = EMULTYPE_TRAP_UD;
> +
> + if (force_emulation_prefix) {
> + char sig[5]; /* ud2; .ascii "kvm" */
> + struct x86_exception e;
> +
> + kvm_read_guest_virt(&vcpu->arch.emulate_ctxt,
> + kvm_get_linear_rip(vcpu), sig, sizeof(sig), &e);
> + if (memcmp(sig, "\xf\xbkvm", sizeof(sig)) == 0) {
> + emulation_type = 0;
> + kvm_rip_write(vcpu, kvm_rip_read(vcpu) + sizeof(sig));
> + }
> + }
>
> - er = emulate_instruction(vcpu, EMULTYPE_TRAP_UD);
> + er = emulate_instruction(vcpu, emulation_type);
> if (er == EMULATE_USER_EXIT)
> return 0;
> if (er != EMULATE_DONE)
The code afterwards is going to inject an #UD if the emulation failed.
I think that preserving the cpu state and forwarding the emulation
failure to userspace would be more useful. The change would probably be
best as:
if (memcmp(sig, "\xf\xbkvm", sizeof(sig)) == 0) {
kvm_rip_write(vcpu, kvm_rip_read(vcpu) + sizeof(sig));
return emulate_instruction(vcpu, 0) == EMULATE_DONE;
}
Looks great otherwise, thanks.
(We want to use this in emulate.c kvm-unit-test that currently fails
because of a hack that doesn't work anymore.)
^ permalink raw reply [flat|nested] 4+ messages in thread
* [PATCH v3 2/2] KVM: X86: Add Force Emulation Prefix for "emulate the next instruction"
2018-03-28 0:48 [PATCH v3 0/2] " Wanpeng Li
@ 2018-03-28 0:48 ` Wanpeng Li
2018-03-29 21:29 ` Radim Krčmář
0 siblings, 1 reply; 4+ messages in thread
From: Wanpeng Li @ 2018-03-28 0:48 UTC (permalink / raw)
To: linux-kernel, kvm
Cc: Paolo Bonzini, Radim Krčmář,
Andrew Cooper, Konrad Rzeszutek Wilk, Liran Alon
From: Wanpeng Li <wanpengli@tencent.com>
There is no easy way to force KVM to run an instruction through the emulator
(by design as that will expose the x86 emulator as a significant attack-surface).
However, we do wish to expose the x86 emulator in case we are testing it
(e.g. via kvm-unit-tests). Therefore, this patch adds a "force emulation prefix"
that is designed to raise #UD which KVM will trap and it's #UD exit-handler will
match "force emulation prefix" to run instruction after prefix by the x86 emulator.
To not expose the x86 emulator by default, we add a module parameter that should
be off by default.
A simple testcase here:
#include <stdio.h>
#include <string.h>
#define HYPERVISOR_INFO 0x40000000
#define CPUID(idx, eax, ebx, ecx, edx) \
asm volatile (\
"ud2a; .ascii \"kvm\"; cpuid" \
:"=b" (*ebx), "=a" (*eax), "=c" (*ecx), "=d" (*edx) \
:"0"(idx) );
void main()
{
unsigned int eax, ebx, ecx, edx;
char string[13];
CPUID(HYPERVISOR_INFO, &eax, &ebx, &ecx, &edx);
*(unsigned int *)(string + 0) = ebx;
*(unsigned int *)(string + 4) = ecx;
*(unsigned int *)(string + 8) = edx;
string[12] = 0;
if (strncmp(string, "KVMKVMKVM\0\0\0", 12) == 0)
printf("kvm guest\n");
else
printf("bare hardware\n");
}
Suggested-by: Andrew Cooper <andrew.cooper3@citrix.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Liran Alon <liran.alon@oracle.com>
Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
---
arch/x86/kvm/x86.c | 18 +++++++++++++++++-
1 file changed, 17 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index e3a60ab..40e2f78 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -146,6 +146,9 @@ bool __read_mostly enable_vmware_backdoor = false;
module_param(enable_vmware_backdoor, bool, S_IRUGO);
EXPORT_SYMBOL_GPL(enable_vmware_backdoor);
+static bool __read_mostly force_emulation_prefix = false;
+module_param(force_emulation_prefix, bool, S_IRUGO);
+
#define KVM_NR_SHARED_MSRS 16
struct kvm_shared_msrs_global {
@@ -4843,8 +4846,21 @@ EXPORT_SYMBOL_GPL(kvm_write_guest_virt_system);
int handle_ud(struct kvm_vcpu *vcpu)
{
enum emulation_result er;
+ int emulation_type = EMULTYPE_TRAP_UD;
+
+ if (force_emulation_prefix) {
+ char sig[5]; /* ud2; .ascii "kvm" */
+ struct x86_exception e;
+
+ kvm_read_guest_virt(&vcpu->arch.emulate_ctxt,
+ kvm_get_linear_rip(vcpu), sig, sizeof(sig), &e);
+ if (memcmp(sig, "\xf\xbkvm", sizeof(sig)) == 0) {
+ emulation_type = 0;
+ kvm_rip_write(vcpu, kvm_rip_read(vcpu) + sizeof(sig));
+ }
+ }
- er = emulate_instruction(vcpu, EMULTYPE_TRAP_UD);
+ er = emulate_instruction(vcpu, emulation_type);
if (er == EMULATE_USER_EXIT)
return 0;
if (er != EMULATE_DONE)
--
2.7.4
^ permalink raw reply related [flat|nested] 4+ messages in thread
end of thread, other threads:[~2018-04-01 11:12 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-03-28 6:55 [PATCH v3 2/2] KVM: X86: Add Force Emulation Prefix for "emulate the next instruction" Liran Alon
-- strict thread matches above, loose matches on Subject: below --
2018-03-28 0:48 [PATCH v3 0/2] " Wanpeng Li
2018-03-28 0:48 ` [PATCH v3 2/2] " Wanpeng Li
2018-03-29 21:29 ` Radim Krčmář
2018-04-01 11:12 ` Wanpeng Li
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.