From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-pl1-f201.google.com (mail-pl1-f201.google.com [209.85.214.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 9E1DC3876C0 for ; Mon, 4 May 2026 22:42:49 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.201 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1777934572; cv=none; b=C7ndZJvkYqpgEwUvHBu0beVClFenQV+5CwohkJ1zIcLEkESayqGVVt3c5YzrQOnkJFDqhx3QBKgDMb1KcLtTp+0Zo7I1N6tggub7NNAd2BJCWne+BSB0uJOZhpJm+ylWROBt9yb64AtW4XadBvndisE1nsQX6Iai8N+9+/Ll78Y= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1777934572; c=relaxed/simple; bh=InGnsmMXyUKk5UR+us+b7DjwbtH0DDKd3A8oEQN+NKI=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Cc:Content-Type; b=J6CHhoB4d8X9JveaH0A1oMqI9k9tFCoNNXF7yZa7mFUDPBan67J7FYS8ZtD2FI+9nPE5OlojDeDwFBiHvjMAeXoqUd+McxdCwwYZnHaMZTrv6tT0DDS0tzIq+ez+TuXfgMPjicmqI+QZ0Mbmj/zAgHz2Y4kN2KJCDNP1HyjmUkg= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com; spf=pass smtp.mailfrom=flex--jthoughton.bounces.google.com; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b=vIb/YPc4; arc=none smtp.client-ip=209.85.214.201 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=flex--jthoughton.bounces.google.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="vIb/YPc4" Received: by mail-pl1-f201.google.com with SMTP id d9443c01a7336-2b258636d16so49473875ad.2 for ; Mon, 04 May 2026 15:42:49 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20251104; t=1777934569; x=1778539369; darn=vger.kernel.org; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=vC+QCMBPXdQWF+dL8GXMYzl+k9Np64sK2GTCcOrjhPY=; b=vIb/YPc4eQY1R+pTc4p70eHlPon7UkycR+sBD/1Q2Ite094bGPk1I+2seC9L6++opt PZJO0MA916CAcPQ2PjbHgCz9VWeFAsHBw/itVy8JQkelSTcZBVg0W8j7q4JvVCVzCBQc IZ+nevL4B8wX/dGI3A4PdW+Sd/e5eZGzr+M/IgO0cYuj1GysDg8GIY6dQkvEVZyF34aI 9XxIZ6D/oNRiX4zQ/LFUM8oGcRGdESwx7v8aSL2zoFNEsjPPtvfHUhnr543oFvT8nB5m R2se9oC+VoaRmckyk6M+86xtJXoLyKwp9Gfwz6HmWcWJI27Ri6NCGg3F9mM9lAXg8vyV m56A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1777934569; x=1778539369; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=vC+QCMBPXdQWF+dL8GXMYzl+k9Np64sK2GTCcOrjhPY=; b=csM9sxJII06zMU/RG0vVuFsjdOGHfGuC2KnFikszP+uoyr54hwvwYzJ6IaAFoycDCL FuePBYv9oTq2kESvdYYBJd/w8h05O6tvawZpGJyGPN63hW5T5BUZ7HDOqUCQadhv42IP SrR9TeKSd6Ife0heLDt/GYq1j3JXURlki84hmy4ouhWMzEo+gAQMepZMjBiCTjc/zj2K aMyU6LNDbeKHNqtjoeP+30zAlMYGTERV7TeIXorj7a6c9KNKMEmCQCZlz3VvJyj72Pxs 5aKiSRlG/zsBvZXzRcZ5YhHS7ZJAcSIWM9K9/pMU/2yWno8ZbYSyc0ac/399mCZJqKXr 7y9Q== X-Forwarded-Encrypted: i=1; AFNElJ897I9H14qAmfhEyWOv5EBz5WXQOfItcyVuTUiJxB1Sts/eqvbSTzcNQZQyL+5xHVXaJwtSzGmWBQ+NJQ4=@vger.kernel.org X-Gm-Message-State: AOJu0YxPY4Dix7w0xLcYXxZ3E0hKUNqBeT5cm0dlOohi7Zu27jpLHqIw w0JgdLAClHyET3E2rRsYFctM19wKEhSSKa81JB6jiTb432FKttszXOTbvoCBtIA8Ta9g4XGMECW oAh//lCQAinI4qKR1ogD9hQ== X-Received: from plmx11.prod.google.com ([2002:a17:902:fe8b:b0:2ae:3bca:37a5]) (user=jthoughton job=prod-delivery.src-stubby-dispatcher) by 2002:a17:903:4b0d:b0:2b0:6d33:e7c0 with SMTP id d9443c01a7336-2ba535e2df5mr4309355ad.1.1777934568708; Mon, 04 May 2026 15:42:48 -0700 (PDT) Date: Mon, 4 May 2026 22:42:11 +0000 In-Reply-To: <20260504224213.1049426-1-jthoughton@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20260504224213.1049426-1-jthoughton@google.com> X-Mailer: git-send-email 2.54.0.545.g6539524ca2-goog Message-ID: <20260504224213.1049426-5-jthoughton@google.com> Subject: [PATCH 4/5] KVM: Hold MMU lock exclusively when calling kvm_arch_flush_shadow_all() From: James Houghton To: Paolo Bonzini Cc: Marc Zyngier , Oliver Upton , Joey Gouly , Suzuki K Poulose , Zenghui Yu , Sean Christopherson , Gavin Shan , Shaoqin Huang , Ricardo Koller , Tianrui Zhao , Bibo Mao , Huacai Chen , James Hogan , linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, loongarch@lists.linux.dev, linux-mips@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, James Houghton Content-Type: text/plain; charset="UTF-8" All architectures that non-trivially implement this function grab the KVM MMU lock exclusively to prevent things like double-freeing of page table entries and caches. Do so generically to somewhat simplify the architecture-specific logic. Without this change, it is possible for kvm_arch_flush_shadow_all() to be called on the same `kvm` at the same time: if the `kvm`'s mm is destroyed (exit_mm()) at the same time to the final reference to the `kvm` is dropped. Signed-off-by: James Houghton --- arch/arm64/kvm/nested.c | 2 +- arch/loongarch/kvm/mmu.c | 4 +++- arch/mips/kvm/mips.c | 2 +- arch/riscv/kvm/mmu.c | 4 ++-- arch/riscv/kvm/vm.c | 2 ++ arch/x86/kvm/mmu/mmu.c | 4 +--- virt/kvm/kvm_main.c | 3 +++ 7 files changed, 13 insertions(+), 8 deletions(-) diff --git a/arch/arm64/kvm/nested.c b/arch/arm64/kvm/nested.c index 977598bff5e6..ba2e6c98bd45 100644 --- a/arch/arm64/kvm/nested.c +++ b/arch/arm64/kvm/nested.c @@ -1190,7 +1190,7 @@ void kvm_arch_flush_shadow_all(struct kvm *kvm) { int i; - guard(write_lock)(&kvm->mmu_lock); + lockdep_assert_held_write(&kvm->mmu_lock); for (i = 0; i < kvm->arch.nested_mmus_size; i++) { struct kvm_s2_mmu *mmu = &kvm->arch.nested_mmus[i]; diff --git a/arch/loongarch/kvm/mmu.c b/arch/loongarch/kvm/mmu.c index 5dbce9b18e1c..120001da26e4 100644 --- a/arch/loongarch/kvm/mmu.c +++ b/arch/loongarch/kvm/mmu.c @@ -486,7 +486,9 @@ void kvm_arch_commit_memory_region(struct kvm *kvm, void kvm_arch_flush_shadow_all(struct kvm *kvm) { - kvm_flush_range(kvm, 0, kvm->arch.gpa_size >> PAGE_SHIFT, 1); + lockdep_assert_held(&kvm->mmu_lock); + + kvm_flush_range(kvm, 0, kvm->arch.gpa_size >> PAGE_SHIFT, 0); } void kvm_arch_flush_shadow_memslot(struct kvm *kvm, struct kvm_memory_slot *slot) diff --git a/arch/mips/kvm/mips.c b/arch/mips/kvm/mips.c index 463b6c4aa62c..4ad9e21a3321 100644 --- a/arch/mips/kvm/mips.c +++ b/arch/mips/kvm/mips.c @@ -180,7 +180,7 @@ long kvm_arch_dev_ioctl(struct file *filp, unsigned int ioctl, void kvm_arch_flush_shadow_all(struct kvm *kvm) { - guard(spinlock)(&kvm->mmu_lock); + lockdep_assert_held(&kvm->mmu_lock); /* Flush whole GPA */ kvm_mips_flush_gpa_pt(kvm, 0, ~0); diff --git a/arch/riscv/kvm/mmu.c b/arch/riscv/kvm/mmu.c index 2d3def024270..c1b9333f17eb 100644 --- a/arch/riscv/kvm/mmu.c +++ b/arch/riscv/kvm/mmu.c @@ -558,7 +558,8 @@ void kvm_riscv_mmu_free_pgd(struct kvm *kvm) struct kvm_gstage gstage; void *pgd = NULL; - spin_lock(&kvm->mmu_lock); + lockdep_assert_held(&kvm->mmu_lock); + if (kvm->arch.pgd) { kvm_riscv_gstage_init(&gstage, kvm); kvm_riscv_gstage_unmap_range(&gstage, 0UL, @@ -568,7 +569,6 @@ void kvm_riscv_mmu_free_pgd(struct kvm *kvm) kvm->arch.pgd_phys = 0; kvm->arch.pgd_levels = 0; } - spin_unlock(&kvm->mmu_lock); if (pgd) free_pages((unsigned long)pgd, get_order(kvm_riscv_gstage_pgd_size)); diff --git a/arch/riscv/kvm/vm.c b/arch/riscv/kvm/vm.c index a9f083feeb76..f704a64bfc48 100644 --- a/arch/riscv/kvm/vm.c +++ b/arch/riscv/kvm/vm.c @@ -38,7 +38,9 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) r = kvm_riscv_gstage_vmid_init(kvm); if (r) { + spin_lock(&kvm->mmu_lock); kvm_riscv_mmu_free_pgd(kvm); + spin_unlock(&kvm->mmu_lock); return r; } diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index 892246204435..6e6f94046b3f 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -7352,7 +7352,7 @@ static void kvm_mmu_zap_all(struct kvm *kvm) LIST_HEAD(invalid_list); int ign; - write_lock(&kvm->mmu_lock); + lockdep_assert_held_write(&kvm->mmu_lock); restart: list_for_each_entry_safe(sp, node, &kvm->arch.active_mmu_pages, link) { if (WARN_ON_ONCE(sp->role.invalid)) @@ -7367,8 +7367,6 @@ static void kvm_mmu_zap_all(struct kvm *kvm) if (tdp_mmu_enabled) kvm_tdp_mmu_zap_all(kvm); - - write_unlock(&kvm->mmu_lock); } void kvm_arch_flush_shadow_all(struct kvm *kvm) diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 89489996fbc1..f5affd3bfda8 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -340,7 +340,10 @@ void kvm_flush_remote_tlbs_memslot(struct kvm *kvm, static void kvm_flush_shadow_all(struct kvm *kvm) { + KVM_MMU_LOCK(kvm); kvm_arch_flush_shadow_all(kvm); + KVM_MMU_UNLOCK(kvm); + kvm_arch_guest_memory_reclaimed(kvm); } -- 2.54.0.545.g6539524ca2-goog