From: Bibo Mao <maobibo@loongson.cn>
Date: Thu, 9 Apr 2026 10:56:36 +0000 (+0800)
Subject: LoongArch: KVM: Move host CSR_EENTRY save and restore in context switch
X-Git-Tag: v7.1-rc1~118^2~12^2~7
X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=f62eb9ca8def410bcef39e8909945409d0968473;p=thirdparty%2Fkernel%2Flinux.git

LoongArch: KVM: Move host CSR_EENTRY save and restore in context switch

CSR register LOONGARCH_CSR_EENTRY is shared between host CPU and guest
vCPU, KVM need save and restore LOONGARCH_CSR_EENTRY register. Here move
LOONGARCH_CSR_EENTRY saving in to context switch function rather than VM
entry.

At most time VM enter/exit is much more frequent than vCPU thread context
switch.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
---

diff --git a/arch/loongarch/kvm/vcpu.c b/arch/loongarch/kvm/vcpu.c
index fc8f7290cfed4..3367a9886b637 100644
--- a/arch/loongarch/kvm/vcpu.c
+++ b/arch/loongarch/kvm/vcpu.c
@@ -320,7 +320,6 @@ static int kvm_pre_enter_guest(struct kvm_vcpu *vcpu)
 		 * and it may also clear KVM_REQ_TLB_FLUSH_GPA pending bit
 		 */
 		kvm_late_check_requests(vcpu);
-		vcpu->arch.host_eentry = csr_read64(LOONGARCH_CSR_EENTRY);
 		/* Clear KVM_LARCH_SWCSR_LATEST as CSR will change when enter guest */
 		vcpu->arch.aux_inuse &= ~KVM_LARCH_SWCSR_LATEST;
 
@@ -1628,9 +1627,11 @@ static int _kvm_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 	 * If not, any old guest state from this vCPU will have been clobbered.
 	 */
 	context = per_cpu_ptr(vcpu->kvm->arch.vmcs, cpu);
-	if (migrated || (context->last_vcpu != vcpu))
+	if (migrated || (context->last_vcpu != vcpu)) {
+		context->last_vcpu = vcpu;
 		vcpu->arch.aux_inuse &= ~KVM_LARCH_HWCSR_USABLE;
-	context->last_vcpu = vcpu;
+		vcpu->arch.host_eentry = csr_read64(LOONGARCH_CSR_EENTRY);
+	}
 
 	/* Restore timer state regardless */
 	kvm_restore_timer(vcpu);