Linux-RISC-V Archive on lore.kernel.org
 help / color / mirror / Atom feed
* [RFC PATCH] kvm/riscv: Add ctxsstatus and ctxhstatus for migration
@ 2025-09-15 15:27 Jinyu Tang
  2025-09-15 16:52 ` Radim Krčmář
  2025-09-15 17:01 ` Andrew Jones
  0 siblings, 2 replies; 4+ messages in thread
From: Jinyu Tang @ 2025-09-15 15:27 UTC (permalink / raw)
  To: Anup Patel, Atish Patra, Andrew Jones, Conor Dooley,
	Yong-Xuan Wang, Paul Walmsley, Nutty Liu, Radim Krcmar
  Cc: kvm, kvm-riscv, linux-riscv, linux-kernel, Jinyu Tang,
	Tianshun Sun

When migrating a VM which guest running in user mode 
(e.g., executing a while(1) application), the target 
VM fails to run because it loses the information of 
guest_context.hstatus and guest_context.sstatus. The 
VM uses the initialized values instead of the correct ones.

This patch adds two new context registers (ctxsstatus and 
ctxhstatus) to the kvm_vcpu_csr structure and implements 
the necessary KVM get and set logic to preserve these values 
during migration.

QEMU needs to be updated to support these new registers.
See https://github.com/tjy-zhu/qemu
for the corresponding QEMU changes.

I'm not sure if adding these CSR registers is a right way. RISCV
KVM doesn't have API to save these two context csrs now. I will
submit the corresponding QEMU patch to the QEMU community if
KVM has API to get and set them.

Signed-off-by: Jinyu Tang <tjytimi@163.com>
Tested-by: Tianshun Sun <stsmail163@163.com>
---
 arch/riscv/include/asm/kvm_host.h |  2 ++
 arch/riscv/include/uapi/asm/kvm.h |  2 ++
 arch/riscv/kvm/vcpu_onereg.c      | 16 ++++++++++++++++
 3 files changed, 20 insertions(+)

diff --git a/arch/riscv/include/asm/kvm_host.h b/arch/riscv/include/asm/kvm_host.h
index d71d3299a..55604b075 100644
--- a/arch/riscv/include/asm/kvm_host.h
+++ b/arch/riscv/include/asm/kvm_host.h
@@ -161,6 +161,8 @@ struct kvm_vcpu_csr {
 	unsigned long vsatp;
 	unsigned long scounteren;
 	unsigned long senvcfg;
+	unsigned long ctxsstatus;
+	unsigned long ctxhstatus;
 };
 
 struct kvm_vcpu_config {
diff --git a/arch/riscv/include/uapi/asm/kvm.h b/arch/riscv/include/uapi/asm/kvm.h
index ef27d4289..cd7d7087f 100644
--- a/arch/riscv/include/uapi/asm/kvm.h
+++ b/arch/riscv/include/uapi/asm/kvm.h
@@ -81,6 +81,8 @@ struct kvm_riscv_csr {
 	unsigned long satp;
 	unsigned long scounteren;
 	unsigned long senvcfg;
+	unsigned long ctxsstatus;
+	unsigned long ctxhstatus;
 };
 
 /* AIA CSR registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
diff --git a/arch/riscv/kvm/vcpu_onereg.c b/arch/riscv/kvm/vcpu_onereg.c
index cce6a38ea..284ee6e81 100644
--- a/arch/riscv/kvm/vcpu_onereg.c
+++ b/arch/riscv/kvm/vcpu_onereg.c
@@ -489,6 +489,12 @@ static int kvm_riscv_vcpu_general_get_csr(struct kvm_vcpu *vcpu,
 	if (reg_num >= sizeof(struct kvm_riscv_csr) / sizeof(unsigned long))
 		return -ENOENT;
 
+	if (reg_num == KVM_REG_RISCV_CSR_REG(ctxsstatus))
+		csr->ctxsstatus = vcpu->arch.guest_context.sstatus;
+
+	if (reg_num == KVM_REG_RISCV_CSR_REG(ctxhstatus))
+		csr->ctxhstatus = vcpu->arch.guest_context.hstatus;
+
 	if (reg_num == KVM_REG_RISCV_CSR_REG(sip)) {
 		kvm_riscv_vcpu_flush_interrupts(vcpu);
 		*out_val = (csr->hvip >> VSIP_TO_HVIP_SHIFT) & VSIP_VALID_MASK;
@@ -515,6 +521,16 @@ static int kvm_riscv_vcpu_general_set_csr(struct kvm_vcpu *vcpu,
 
 	((unsigned long *)csr)[reg_num] = reg_val;
 
+	if (reg_num == KVM_REG_RISCV_CSR_REG(ctxsstatus)) {
+		if (csr->ctxsstatus != 0)
+			vcpu->arch.guest_context.sstatus = csr->ctxsstatus;
+	}
+
+	if (reg_num == KVM_REG_RISCV_CSR_REG(ctxhstatus)) {
+		if (csr->ctxhstatus != 0)
+			vcpu->arch.guest_context.hstatus = csr->ctxhstatus;
+	}
+
 	if (reg_num == KVM_REG_RISCV_CSR_REG(sip))
 		WRITE_ONCE(vcpu->arch.irqs_pending_mask[0], 0);
 
-- 
2.43.0


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

^ permalink raw reply related	[flat|nested] 4+ messages in thread
* Re: [RFC PATCH] kvm/riscv: Add ctxsstatus and ctxhstatus for migration
  2025-09-15 15:27 [RFC PATCH] kvm/riscv: Add ctxsstatus and ctxhstatus for migration Jinyu Tang
@ 2025-09-15 16:52 ` Radim Krčmář
  2025-09-15 17:01   ` Radim Krčmář
  2025-09-15 17:01 ` Andrew Jones
  1 sibling, 1 reply; 4+ messages in thread
From: Radim Krčmář @ 2025-09-15 16:52 UTC (permalink / raw)
  To: Jinyu Tang, Anup Patel, Atish Patra, Andrew Jones, Conor Dooley,
	Yong-Xuan Wang, Paul Walmsley, Nutty Liu
  Cc: kvm, kvm-riscv, linux-riscv, linux-kernel, Tianshun Sun

2025-09-15T23:27:31+08:00, Jinyu Tang <tjytimi@163.com>:
> When migrating a VM which guest running in user mode 
> (e.g., executing a while(1) application), the target 
> VM fails to run because it loses the information of 
> guest_context.hstatus and guest_context.sstatus. The 
> VM uses the initialized values instead of the correct ones.
>
> This patch adds two new context registers (ctxsstatus and 
> ctxhstatus) to the kvm_vcpu_csr structure and implements 
> the necessary KVM get and set logic to preserve these values 
> during migration.
>
> QEMU needs to be updated to support these new registers.
> See https://github.com/tjy-zhu/qemu
> for the corresponding QEMU changes.
>
> I'm not sure if adding these CSR registers is a right way. RISCV
> KVM doesn't have API to save these two context csrs now. I will
> submit the corresponding QEMU patch to the QEMU community if
> KVM has API to get and set them.

I don't think it is...

> Signed-off-by: Jinyu Tang <tjytimi@163.com>
> Tested-by: Tianshun Sun <stsmail163@163.com>
> ---
> diff --git a/arch/riscv/kvm/vcpu_onereg.c b/arch/riscv/kvm/vcpu_onereg.c
> @@ -489,6 +489,12 @@ static int kvm_riscv_vcpu_general_get_csr(struct kvm_vcpu *vcpu,
>  	if (reg_num >= sizeof(struct kvm_riscv_csr) / sizeof(unsigned long))
>  		return -ENOENT;
>  
> +	if (reg_num == KVM_REG_RISCV_CSR_REG(ctxsstatus))
> +		csr->ctxsstatus = vcpu->arch.guest_context.sstatus;

Userspace can set guest privilege mode via KVM_REG_RISCV_CORE_REG(mode).

This does propagate to guest_context.sstatus, which is otherwise
internal KVM state, so we definitely shouldn't expose it directly.

> +
> +	if (reg_num == KVM_REG_RISCV_CSR_REG(ctxhstatus))
> +		csr->ctxhstatus = vcpu->arch.guest_context.hstatus;

Neither should userspace be able to directly set hstatus.
KVM should derive it from other userspace configuration.

What isn't correctly reflected in hstatus?

Thanks.

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

^ permalink raw reply	[flat|nested] 4+ messages in thread
* Re: [RFC PATCH] kvm/riscv: Add ctxsstatus and ctxhstatus for migration
  2025-09-15 16:52 ` Radim Krčmář
@ 2025-09-15 17:01   ` Radim Krčmář
  0 siblings, 0 replies; 4+ messages in thread
From: Radim Krčmář @ 2025-09-15 17:01 UTC (permalink / raw)
  To: Jinyu Tang, Anup Patel, Atish Patra, Andrew Jones, Conor Dooley,
	Yong-Xuan Wang, Paul Walmsley, Nutty Liu
  Cc: kvm, kvm-riscv, linux-riscv, linux-kernel, Tianshun Sun

2025-09-15T18:52:56+02:00, Radim Krčmář <rkrcmar@ventanamicro.com>:
> 2025-09-15T23:27:31+08:00, Jinyu Tang <tjytimi@163.com>:
>> +	if (reg_num == KVM_REG_RISCV_CSR_REG(ctxhstatus))
>> +		csr->ctxhstatus = vcpu->arch.guest_context.hstatus;
>
> Neither should userspace be able to directly set hstatus.
> KVM should derive it from other userspace configuration.
>
> What isn't correctly reflected in hstatus?

Ah, we don't set hstatus.SPVP when setting KVM_REG_RISCV_CORE_REG(mode).

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

^ permalink raw reply	[flat|nested] 4+ messages in thread
* Re: [RFC PATCH] kvm/riscv: Add ctxsstatus and ctxhstatus for migration
  2025-09-15 15:27 [RFC PATCH] kvm/riscv: Add ctxsstatus and ctxhstatus for migration Jinyu Tang
  2025-09-15 16:52 ` Radim Krčmář
@ 2025-09-15 17:01 ` Andrew Jones
  1 sibling, 0 replies; 4+ messages in thread
From: Andrew Jones @ 2025-09-15 17:01 UTC (permalink / raw)
  To: Jinyu Tang
  Cc: Anup Patel, Atish Patra, Conor Dooley, Yong-Xuan Wang,
	Paul Walmsley, Nutty Liu, Radim Krcmar, kvm, kvm-riscv,
	linux-riscv, linux-kernel, Tianshun Sun

On Mon, Sep 15, 2025 at 11:27:31PM +0800, Jinyu Tang wrote:
> When migrating a VM which guest running in user mode 
> (e.g., executing a while(1) application), the target 
> VM fails to run because it loses the information of 
> guest_context.hstatus and guest_context.sstatus. The 
> VM uses the initialized values instead of the correct ones.

Does https://lore.kernel.org/all/20250915070811.3422578-1-xb@ultrarisc.com/
fix this?

Thanks,
drew

> 
> This patch adds two new context registers (ctxsstatus and 
> ctxhstatus) to the kvm_vcpu_csr structure and implements 
> the necessary KVM get and set logic to preserve these values 
> during migration.
> 
> QEMU needs to be updated to support these new registers.
> See https://github.com/tjy-zhu/qemu
> for the corresponding QEMU changes.
> 
> I'm not sure if adding these CSR registers is a right way. RISCV
> KVM doesn't have API to save these two context csrs now. I will
> submit the corresponding QEMU patch to the QEMU community if
> KVM has API to get and set them.
> 
> Signed-off-by: Jinyu Tang <tjytimi@163.com>
> Tested-by: Tianshun Sun <stsmail163@163.com>
> ---
>  arch/riscv/include/asm/kvm_host.h |  2 ++
>  arch/riscv/include/uapi/asm/kvm.h |  2 ++
>  arch/riscv/kvm/vcpu_onereg.c      | 16 ++++++++++++++++
>  3 files changed, 20 insertions(+)
> 
> diff --git a/arch/riscv/include/asm/kvm_host.h b/arch/riscv/include/asm/kvm_host.h
> index d71d3299a..55604b075 100644
> --- a/arch/riscv/include/asm/kvm_host.h
> +++ b/arch/riscv/include/asm/kvm_host.h
> @@ -161,6 +161,8 @@ struct kvm_vcpu_csr {
>  	unsigned long vsatp;
>  	unsigned long scounteren;
>  	unsigned long senvcfg;
> +	unsigned long ctxsstatus;
> +	unsigned long ctxhstatus;
>  };
>  
>  struct kvm_vcpu_config {
> diff --git a/arch/riscv/include/uapi/asm/kvm.h b/arch/riscv/include/uapi/asm/kvm.h
> index ef27d4289..cd7d7087f 100644
> --- a/arch/riscv/include/uapi/asm/kvm.h
> +++ b/arch/riscv/include/uapi/asm/kvm.h
> @@ -81,6 +81,8 @@ struct kvm_riscv_csr {
>  	unsigned long satp;
>  	unsigned long scounteren;
>  	unsigned long senvcfg;
> +	unsigned long ctxsstatus;
> +	unsigned long ctxhstatus;
>  };
>  
>  /* AIA CSR registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
> diff --git a/arch/riscv/kvm/vcpu_onereg.c b/arch/riscv/kvm/vcpu_onereg.c
> index cce6a38ea..284ee6e81 100644
> --- a/arch/riscv/kvm/vcpu_onereg.c
> +++ b/arch/riscv/kvm/vcpu_onereg.c
> @@ -489,6 +489,12 @@ static int kvm_riscv_vcpu_general_get_csr(struct kvm_vcpu *vcpu,
>  	if (reg_num >= sizeof(struct kvm_riscv_csr) / sizeof(unsigned long))
>  		return -ENOENT;
>  
> +	if (reg_num == KVM_REG_RISCV_CSR_REG(ctxsstatus))
> +		csr->ctxsstatus = vcpu->arch.guest_context.sstatus;
> +
> +	if (reg_num == KVM_REG_RISCV_CSR_REG(ctxhstatus))
> +		csr->ctxhstatus = vcpu->arch.guest_context.hstatus;
> +
>  	if (reg_num == KVM_REG_RISCV_CSR_REG(sip)) {
>  		kvm_riscv_vcpu_flush_interrupts(vcpu);
>  		*out_val = (csr->hvip >> VSIP_TO_HVIP_SHIFT) & VSIP_VALID_MASK;
> @@ -515,6 +521,16 @@ static int kvm_riscv_vcpu_general_set_csr(struct kvm_vcpu *vcpu,
>  
>  	((unsigned long *)csr)[reg_num] = reg_val;
>  
> +	if (reg_num == KVM_REG_RISCV_CSR_REG(ctxsstatus)) {
> +		if (csr->ctxsstatus != 0)
> +			vcpu->arch.guest_context.sstatus = csr->ctxsstatus;
> +	}
> +
> +	if (reg_num == KVM_REG_RISCV_CSR_REG(ctxhstatus)) {
> +		if (csr->ctxhstatus != 0)
> +			vcpu->arch.guest_context.hstatus = csr->ctxhstatus;
> +	}
> +
>  	if (reg_num == KVM_REG_RISCV_CSR_REG(sip))
>  		WRITE_ONCE(vcpu->arch.irqs_pending_mask[0], 0);
>  
> -- 
> 2.43.0
> 
> 
> -- 
> kvm-riscv mailing list
> kvm-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/kvm-riscv

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

^ permalink raw reply	[flat|nested] 4+ messages in thread
end of thread, other threads:[~2025-09-15 17:02 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-09-15 15:27 [RFC PATCH] kvm/riscv: Add ctxsstatus and ctxhstatus for migration Jinyu Tang
2025-09-15 16:52 ` Radim Krčmář
2025-09-15 17:01   ` Radim Krčmář
2025-09-15 17:01 ` Andrew Jones

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox