[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20240624071422.3473789-5-maobibo@loongson.cn>
Date: Mon, 24 Jun 2024 15:14:19 +0800
From: Bibo Mao <maobibo@...ngson.cn>
To: Tianrui Zhao <zhaotianrui@...ngson.cn>,
Huacai Chen <chenhuacai@...nel.org>
Cc: WANG Xuerui <kernel@...0n.name>,
Sean Christopherson <seanjc@...gle.com>,
kvm@...r.kernel.org,
loongarch@...ts.linux.dev,
linux-kernel@...r.kernel.org,
WANG Rui <wangrui@...ngson.cn>
Subject: [PATCH v3 4/7] LoongArch: KVM: Add memory barrier before update pmd entry
When updating pmd entry such as allocating new pmd page or splitting
huge page into normal page, it is necessary to firstly update all pte
entries, and then update pmd entry.
It is weak order with LoongArch system, there will be problem if other
vcpus sees pmd update firstly however pte is not updated. Here smp_wmb()
is added to assure this.
Signed-off-by: Bibo Mao <maobibo@...ngson.cn>
---
arch/loongarch/kvm/mmu.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/loongarch/kvm/mmu.c b/arch/loongarch/kvm/mmu.c
index 1690828bd44b..092c063e4e56 100644
--- a/arch/loongarch/kvm/mmu.c
+++ b/arch/loongarch/kvm/mmu.c
@@ -163,6 +163,7 @@ static kvm_pte_t *kvm_populate_gpa(struct kvm *kvm,
child = kvm_mmu_memory_cache_alloc(cache);
_kvm_pte_init(child, ctx.invalid_ptes[ctx.level - 1]);
+ smp_wmb(); /* make pte visible before pmd */
kvm_set_pte(entry, __pa(child));
} else if (kvm_pte_huge(*entry)) {
return entry;
@@ -746,6 +747,7 @@ static kvm_pte_t *kvm_split_huge(struct kvm_vcpu *vcpu, kvm_pte_t *ptep, gfn_t g
val += PAGE_SIZE;
}
+ smp_wmb(); /* make pte visible before pmd */
/* The later kvm_flush_tlb_gpa() will flush hugepage tlb */
kvm_set_pte(ptep, __pa(child));
--
2.39.3
Powered by blists - more mailing lists