On Tue, Feb 25, 2025, Yan Zhao wrote:

On Mon, Feb 24, 2025 at 03:55:39PM -0800, Sean Christopherson wrote:

quoted

Don't load (and then put) a vCPU when unloading its MMU during VM
destruction, as nothing in kvm_mmu_unload() accesses vCPU state beyond the
root page/address of each MMU, i.e. can't possible need to run with the
vCPU loaded.

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 arch/x86/kvm/x86.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 045c61cc7e54..9978ed4c0917 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c

@@ -12767,13 +12767,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	return ret;
 }
 
-static void kvm_unload_vcpu_mmu(struct kvm_vcpu *vcpu)
-{
-	vcpu_load(vcpu);
-	kvm_mmu_unload(vcpu);
-	vcpu_put(vcpu);
-}
-
 static void kvm_unload_vcpu_mmus(struct kvm *kvm)
 {
 	unsigned long i;

@@ -12781,7 +12774,7 @@ static void kvm_unload_vcpu_mmus(struct kvm *kvm)
 
 	kvm_for_each_vcpu(i, vcpu, kvm) {
 		kvm_clear_async_pf_completion_queue(vcpu);
-		kvm_unload_vcpu_mmu(vcpu);
+		kvm_mmu_unload(vcpu);

What about just dropping kvm_unload_vcpu_mmu() here?
kvm_mmu_unload() will be invoked again in kvm_mmu_destroy().

kvm_arch_vcpu_destroy() --> kvm_mmu_destroy() --> kvm_mmu_unload().

Ugh, I missed that there's yet another call to kvm_mmu_unload().  I definitely
agree with dropping the first kvm_mmu_load(), but I'll do it in a follow-up patch
so that all three changes are isolated (not doing the load/put, doing unload as
part of vCPU destruction, doing unload only once at the end).

And looking at both calls to kvm_mmu_unload(), I suspect that grabbing kvm->srcu
around kvm_mmu_destroy() is unnecessary.  I'll try cleaning that up as well.