Skip to contentSkip to conversation

KVM: x86: Check irqchip mode before create PIT

PATCH Superseded Brilliant Hanabi 3 participants 9 messages Jan 24, 2024 – Jan 26, 2024

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 KVM: x86: Check irqchip mode before create PIT

As the kvm api(https://docs.kernel.org/virt/kvm/api.html) reads, KVM_CREATE_PIT2 call is only valid after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP.

Without this check, I can create PIT first and enable irqchip-split then, which may cause the PIT invalid because of lacking of in-kernel PIC to inject the interrupt.

---
 arch/x86/kvm/x86.c | 2 ++
 1 file changed, 2 insertions(+)
arch/x86/kvm/x86.c +2 -0
index 27e23714e960..3edc8478310f 100644
@@ -7016,6 +7016,8 @@ int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg)
 		r = -EEXIST;
 		if (kvm->arch.vpit)
 			goto create_pit_unlock;
+		if (!pic_in_kernel(kvm))
+			goto create_pit_unlock;
 		r = -ENOMEM;
 		kvm->arch.vpit = kvm_create_pit(kvm, u.pit_config.flags);
 		if (kvm->arch.vpit)
Signature · 1 line
2.39.3
#1 lore raw
Quoted 7 lines · Brilliant Hanabi

As the kvm api(https://docs.kernel.org/virt/kvm/api.html) reads, KVM_CREATE_PIT2 call is only valid after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP.

Without this check, I can create PIT first and enable irqchip-split then, which may cause the PIT invalid because of lacking of in-kernel PIC to inject the interrupt.

Does this cause actual problems beyond the PIT not working for the guest? E.g. does it put the host kernel at risk? If the only problem is that the PIT doesn't work as expected, I'm tempted to tweak the docs to say that KVM's PIT emulation won't work without an in-kernel I/O APIC. Rejecting the ioctl could theoertically break misconfigured setups that happen to work, e.g. because the guest never uses the PIT.

Quoted 10 lines
---
 arch/x86/kvm/x86.c | 2 ++
 1 file changed, 2 insertions(+)
arch/x86/kvm/x86.c +0 -0
index 27e23714e960..3edc8478310f 100644

arch/x86/kvm/x86.c int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) · line 7020

Quoted 5 lines
arch/x86/kvm/x86.c +2 -0
@@ -7016,6 +7016,8 @@ int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg)
 		r = -EEXIST;
 		if (kvm->arch.vpit)
 			goto create_pit_unlock;
+		if (!pic_in_kernel(kvm))
+			goto create_pit_unlock;

-EEXIST is not an appropriate errno.

Quoted 6 lines
arch/x86/kvm/x86.c +0 -0
 		r = -ENOMEM;
 		kvm->arch.vpit = kvm_create_pit(kvm, u.pit_config.flags);
 		if (kvm->arch.vpit)
Signature · 1 line
2.39.3
author
#2 lore raw
Quoted 9 lines · Brilliant Hanabi
Quoted 7 lines

As the kvm api(https://docs.kernel.org/virt/kvm/api.html) reads, KVM_CREATE_PIT2 call is only valid after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP.

Without this check, I can create PIT first and enable irqchip-split then, which may cause the PIT invalid because of lacking of in-kernel PIC to inject the interrupt.

Quoted 6 lines

Does this cause actual problems beyond the PIT not working for the guest? E.g. does it put the host kernel at risk? If the only problem is that the PIT doesn't work as expected, I'm tempted to tweak the docs to say that KVM's PIT emulation won't work without an in-kernel I/O APIC. Rejecting the ioctl could theoertically break misconfigured setups that happen to work, e.g. because the guest never uses the PIT.

I don't think it will put the host kernel at risk. But that's exactly what kvmtool does: it creates in-kernel PIT first and set KVM_CREATE_IRQCHIP then. I found this problem because I was working on implementing a userspace PIC and PIT in kvmtool. As I planned, I'm going to commit a related patch to kvmtool if this patch will be applied.

Quoted 5 lines
Quoted 5 lines
---
 arch/x86/kvm/x86.c | 2 ++
 1 file changed, 2 insertions(+)
Quoted 12 lines

arch/x86/kvm/x86.c int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) · line 7020

Quoted 10 lines
arch/x86/kvm/x86.c +2 -0
index 27e23714e960..3edc8478310f 100644
@@ -7016,6 +7016,8 @@ int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg)
 		r = -EEXIST;
 		if (kvm->arch.vpit)
 			goto create_pit_unlock;
+		if (!pic_in_kernel(kvm))
+			goto create_pit_unlock;

-EEXIST is not an appropriate errno.

Which errno do you think is better?

Quoted 6 lines
Quoted 6 lines
arch/x86/kvm/x86.c +0 -0
 		r = -ENOMEM;
 		kvm->arch.vpit = kvm_create_pit(kvm, u.pit_config.flags);
 		if (kvm->arch.vpit)
Signature · 1 line
2.39.3
#3 lore raw
Quoted 16 lines · moehanabi
Quoted 9 lines · Brilliant Hanabi
Quoted 7 lines

As the kvm api(https://docs.kernel.org/virt/kvm/api.html) reads, KVM_CREATE_PIT2 call is only valid after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP.

Without this check, I can create PIT first and enable irqchip-split then, which may cause the PIT invalid because of lacking of in-kernel PIC to inject the interrupt.

Quoted 6 lines

Does this cause actual problems beyond the PIT not working for the guest? E.g. does it put the host kernel at risk? If the only problem is that the PIT doesn't work as expected, I'm tempted to tweak the docs to say that KVM's PIT emulation won't work without an in-kernel I/O APIC. Rejecting the ioctl could theoertically break misconfigured setups that happen to work, e.g. because the guest never uses the PIT.

Quoted 2 lines

I don't think it will put the host kernel at risk. But that's exactly what kvmtool does: it creates in-kernel PIT first and set KVM_CREATE_IRQCHIP then.

Right. My concern, which could be unfounded paranoia, is that rejecting an ioctl() that used to succeed could break existing setups. E.g. if a userspace VMM creates a PIT and checks the ioctl() result, but its guest(s) never actually use the PIT and so don't care that the PIT is busted.

Quoted 22 lines

I found this problem because I was working on implementing a userspace PIC and PIT in kvmtool. As I planned, I'm going to commit a related patch to kvmtool if this patch will be applied.

Quoted 5 lines
Quoted 5 lines
---
 arch/x86/kvm/x86.c | 2 ++
 1 file changed, 2 insertions(+)
Quoted 12 lines

arch/x86/kvm/x86.c int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) · line 7020

Quoted 10 lines
arch/x86/kvm/x86.c +2 -0
index 27e23714e960..3edc8478310f 100644
@@ -7016,6 +7016,8 @@ int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg)
 		r = -EEXIST;
 		if (kvm->arch.vpit)
 			goto create_pit_unlock;
+		if (!pic_in_kernel(kvm))
+			goto create_pit_unlock;

-EEXIST is not an appropriate errno.

Quoted 1 line

Which errno do you think is better?

Maybe ENOENT?

#4 lore raw
Quoted 17 lines · moehanabi
Quoted 16 lines
Quoted 15 lines · Brilliant Hanabi
Quoted 7 lines

As the kvm api(https://docs.kernel.org/virt/kvm/api.html) reads, KVM_CREATE_PIT2 call is only valid after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP.

Without this check, I can create PIT first and enable irqchip-split then, which may cause the PIT invalid because of lacking of in-kernel PIC to inject the interrupt.

Does this cause actual problems beyond the PIT not working for the guest? E.g. does it put the host kernel at risk? If the only problem is that the PIT doesn't work as expected, I'm tempted to tweak the docs to say that KVM's PIT emulation won't work without an in-kernel I/O APIC. Rejecting the ioctl could theoertically break misconfigured setups that happen to work, e.g. because the guest never uses the PIT.

Quoted 7 lines
Quoted 2 lines

I don't think it will put the host kernel at risk. But that's exactly what kvmtool does: it creates in-kernel PIT first and set KVM_CREATE_IRQCHIP then.

Right. My concern, which could be unfounded paranoia, is that rejecting an ioctl() that used to succeed could break existing setups. E.g. if a userspace VMM creates a PIT and checks the ioctl() result, but its guest(s) never actually use the PIT and so don't care that the PIT is busted.

Thanks for your review. In my opinion, it is better to avoid potential bugs which is difficult to detect, as long as you can return errors to let developers know about them in advance, although the kernel is not to blame for this bug.

Quoted 22 lines
Quoted 22 lines

I found this problem because I was working on implementing a userspace PIC and PIT in kvmtool. As I planned, I'm going to commit a related patch to kvmtool if this patch will be applied.

Quoted 17 lines
Quoted 10 lines
---
 arch/x86/kvm/x86.c | 2 ++
 1 file changed, 2 insertions(+)
arch/x86/kvm/x86.c +0 -0
index 27e23714e960..3edc8478310f 100644

arch/x86/kvm/x86.c int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) · line 7020

Quoted 5 lines
arch/x86/kvm/x86.c +2 -0
@@ -7016,6 +7016,8 @@ int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg)
 		r = -EEXIST;
 		if (kvm->arch.vpit)
 			goto create_pit_unlock;
+		if (!pic_in_kernel(kvm))
+			goto create_pit_unlock;

-EEXIST is not an appropriate errno.

Quoted 4 lines
Quoted 1 line

Which errno do you think is better?

Maybe ENOENT?

I'm glad to send a new version patch if you're willing to accept the patch.

#5 lore raw
Quoted 25 lines · Brilliant Hanabi
Quoted 17 lines · moehanabi
Quoted 16 lines
Quoted 15 lines · Brilliant Hanabi
Quoted 7 lines

As the kvm api(https://docs.kernel.org/virt/kvm/api.html) reads, KVM_CREATE_PIT2 call is only valid after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP.

Without this check, I can create PIT first and enable irqchip-split then, which may cause the PIT invalid because of lacking of in-kernel PIC to inject the interrupt.

Does this cause actual problems beyond the PIT not working for the guest? E.g. does it put the host kernel at risk? If the only problem is that the PIT doesn't work as expected, I'm tempted to tweak the docs to say that KVM's PIT emulation won't work without an in-kernel I/O APIC. Rejecting the ioctl could theoertically break misconfigured setups that happen to work, e.g. because the guest never uses the PIT.

Quoted 7 lines
Quoted 2 lines

I don't think it will put the host kernel at risk. But that's exactly what kvmtool does: it creates in-kernel PIT first and set KVM_CREATE_IRQCHIP then.

Right. My concern, which could be unfounded paranoia, is that rejecting an ioctl() that used to succeed could break existing setups. E.g. if a userspace VMM creates a PIT and checks the ioctl() result, but its guest(s) never actually use the PIT and so don't care that the PIT is busted.

Quoted 4 lines

Thanks for your review. In my opinion, it is better to avoid potential bugs which is difficult to detect, as long as you can return errors to let developers know about them in advance, although the kernel is not to blame for this bug.

Oh, I completely agree that explict errors are far better. My only concern is that there's a teeny tiny chance that rejecting an ioctl() that used to work could break userspace.

Quoted 22 lines
Quoted 22 lines
Quoted 22 lines

I found this problem because I was working on implementing a userspace PIC and PIT in kvmtool. As I planned, I'm going to commit a related patch to kvmtool if this patch will be applied.

Quoted 17 lines
Quoted 10 lines
---
 arch/x86/kvm/x86.c | 2 ++
 1 file changed, 2 insertions(+)
arch/x86/kvm/x86.c +0 -0
index 27e23714e960..3edc8478310f 100644

arch/x86/kvm/x86.c int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) · line 7020

Quoted 5 lines
arch/x86/kvm/x86.c +2 -0
@@ -7016,6 +7016,8 @@ int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg)
 		r = -EEXIST;
 		if (kvm->arch.vpit)
 			goto create_pit_unlock;
+		if (!pic_in_kernel(kvm))
+			goto create_pit_unlock;

-EEXIST is not an appropriate errno.

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

Which errno do you think is better?

Maybe ENOENT?

I'm glad to send a new version patch if you're willing to accept the patch.

Go ahead and send v2. I'll get Paolo's thoughts on whether or not this is likely to break userspace and we can go from there.

author
#6 lore raw

PATCH v2 1/1 KVM: x86: Check irqchip mode before create PIT

As the kvm api(https://docs.kernel.org/virt/kvm/api.html) reads, KVM_CREATE_PIT2 call is only valid after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP.

Without this check, I can create PIT first and enable irqchip-split then, which may cause the PIT invalid because of lacking of in-kernel PIC to inject the interrupt.

---
v1 -> v2: Change errno from -EEXIST to -ENOENT.
v1 link: https://lore.kernel.org/lkml/[email protected]/

 arch/x86/kvm/x86.c | 3 +++
 1 file changed, 3 insertions(+)
arch/x86/kvm/x86.c +3 -0
index 27e23714e960..c1e3aecd627f 100644
@@ -7016,6 +7016,9 @@ int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg)
 		r = -EEXIST;
 		if (kvm->arch.vpit)
 			goto create_pit_unlock;
+		r = -ENOENT;
+		if (!pic_in_kernel(kvm))
+			goto create_pit_unlock;
 		r = -ENOMEM;
 		kvm->arch.vpit = kvm_create_pit(kvm, u.pit_config.flags);
 		if (kvm->arch.vpit)
Signature · 1 line
2.39.3
maintainer Applied
#7 lore raw
Quoted 10 lines · Tengfei Yu

As the kvm api(https://docs.kernel.org/virt/kvm/api.html) reads, KVM_CREATE_PIT2 call is only valid after enabling in-kernel irqchip support via KVM_CREATE_IRQCHIP.

Without this check, I can create PIT first and enable irqchip-split then, which may cause the PIT invalid because of lacking of in-kernel PIC to inject the interrupt.

Queued, thanks.

Paolo

Quoted 24 lines

---
v1 -> v2: Change errno from -EEXIST to -ENOENT.
v1 link: https://lore.kernel.org/lkml/[email protected]/

 arch/x86/kvm/x86.c | 3 +++
 1 file changed, 3 insertions(+)
arch/x86/kvm/x86.c +3 -1
index 27e23714e960..c1e3aecd627f 100644
@@ -7016,6 +7016,9 @@ int kvm_arch_vm_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg)
                r = -EEXIST;
                if (kvm->arch.vpit)
                        goto create_pit_unlock;
+               r = -ENOENT;
+               if (!pic_in_kernel(kvm))
+                       goto create_pit_unlock;
                r = -ENOMEM;
                kvm->arch.vpit = kvm_create_pit(kvm, u.pit_config.flags);
                if (kvm->arch.vpit)
--

2.39.3

maintainer
#8 lore raw
In reply to Sean Christopherson
Quoted 6 lines · Sean Christopherson
Quoted 4 lines · Brilliant Hanabi

Thanks for your review. In my opinion, it is better to avoid potential bugs which is difficult to detect, as long as you can return errors to let developers know about them in advance, although the kernel is not to blame for this bug.

Quoted 6 lines

Oh, I completely agree that explict errors are far better. My only concern is that there's a teeny tiny chance that rejecting an ioctl() that used to work could break userspace.

Go ahead and send v2. I'll get Paolo's thoughts on whether or not this is likely to break userspace and we can go from there.

I share the same worry but I agree it's quite unlikely. Let's just do it, and if someone complains we'll revert it.

Paolo