From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-pf1-f202.google.com (mail-pf1-f202.google.com [209.85.210.202]) (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 020C837CD55 for ; Fri, 13 Mar 2026 06:13:53 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.210.202 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1773382441; cv=none; b=WdtSRdrvtn1Kglgv/ElIL/6fr60P8SxvhUesxXJf2OcaPd9X91A6Dmb/Auxuq9Oc41YyAKrOeGsChhQ+KY17R+oSVmQwcySWhAlohAR+ju1z1JD9qePbXonkE0dvSD6KN82OSfo1ODRhTEy0Dq85HBD400ND31IjIHiHRZgbR2k= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1773382441; c=relaxed/simple; bh=PsXVR6/m97u9rpLVI044KHOk18BufZ3/bkmeJAuyM80=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Cc:Content-Type; b=eQGLLgbwayBIs36Q+cm3ORsFRyr3iXxTz9xaSyuYE11FQ+i6DKwWwOQKbJt4xgim7s7OehJWWUCHJfciy1nhl/MvaL8u2ofloLnngMV0HnEBhnODFnivD5S8pVnk3QoyquT4ne9ItUAQ85DYWTcGsK/JNf8CNWxKij3frMV2HF0= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com; spf=pass smtp.mailfrom=flex--ackerleytng.bounces.google.com; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b=eACnccFO; arc=none smtp.client-ip=209.85.210.202 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--ackerleytng.bounces.google.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="eACnccFO" Received: by mail-pf1-f202.google.com with SMTP id d2e1a72fcca58-829942cf9deso1459893b3a.0 for ; Thu, 12 Mar 2026 23:13:53 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20251104; t=1773382433; x=1773987233; 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=Ru4SOG48X5hnADT0eQPIQy13r0oyv8ixWqwkXqYhyy0=; b=eACnccFO/T6kJNzvM6HCM3qj08M2q65ZmvYnhWxX6ZeaqFb90Q9WQ/hlgFG0T6I+zp ic60R5rR+BFkVkk0Md4g8YTlKIEf5SSEiKAmU39CTp9gbyVniVIVa74rTasGPS/3iFXu Uz3AkDK2UqXOBENZt5uCEDPzwzHOJ63Yslc0/41DB9zDmcmMY3N02u6hQwLezEvjhGyR PXTy0u5m7Fxmu4ErUMmkk74qqk9eT3KCzqHHl5mx/s0ZmBkcJHvp2ff7mfsXsVtkg7wp qEkAIVhWdVAz3S0PLK4NaX6saC9IK6mp0Ea18cPF/gaqXFKFU92XIh21SU1Nd7ppb4IZ kiLQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1773382433; x=1773987233; 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=Ru4SOG48X5hnADT0eQPIQy13r0oyv8ixWqwkXqYhyy0=; b=qPXmG2KKBcEvUXxwvyiK6IBn5EMcAi6xU0ANnzhOly9M74cCx6inm1uUYNxilHQjjW OhSi2fN1NsbhN14uVAQN6/AxCEuv4l/4KiXDMQvIAujRQl9U7vlgFSn8lokNanXwBuEP wBBvBGfhAt0d7dOgLeW1KkZCTZdW/z0VGYBnzIfFW0n+mWFd17xUtSBUHIqj7PfpXv6f VStSzhDipgjugKLFIf8g7cfK1rxq7JlQ4CYYppCSkv9DRA7kg19rp87adqv2y5joWMiw 4cSTchBmP1cGkRUsLy/O6oFwTPAWjBkbl2xf/UVG+465Nbq0Q885ryx1jwG/Ise4OBdc Pqcw== X-Forwarded-Encrypted: i=1; AJvYcCVcyj9iqOy0exW+6oaJLLKGcYAH6/RtnTk0TyRttbqVxQmuncf430OKUtCZZTPytBLLPE8ZkBcALBTjnXfUAAA=@vger.kernel.org X-Gm-Message-State: AOJu0YyuW2Qiduf0+nXbLWza5yp9u8g3uzSGyr91tnSiqQStl5xXCoAa PWiwn0EwaEG63yTi5Zs39aUZgFqCQjhPhZ7dQC33JvfT+CI4WWdME/jDsa+wyGRpMWV7jxOh6tN x6AxsYYijDYYlu7NhIMjQn7HX9Q== X-Received: from pfld9.prod.google.com ([2002:a05:6a00:1989:b0:824:a502:8fea]) (user=ackerleytng job=prod-delivery.src-stubby-dispatcher) by 2002:a05:6a00:a113:b0:827:3e19:574b with SMTP id d2e1a72fcca58-82a198ad1bcmr1863557b3a.43.1773382432573; Thu, 12 Mar 2026 23:13:52 -0700 (PDT) Date: Fri, 13 Mar 2026 06:13:18 +0000 In-Reply-To: <20260313-gmem-inplace-conversion-v3-0-5fc12a70ec89@google.com> Precedence: bulk X-Mailing-List: linux-kselftest@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20260313-gmem-inplace-conversion-v3-0-5fc12a70ec89@google.com> X-Developer-Key: i=ackerleytng@google.com; a=ed25519; pk=sAZDYXdm6Iz8FHitpHeFlCMXwabodTm7p8/3/8xUxuU= X-Developer-Signature: v=1; a=ed25519-sha256; t=1773382364; l=7369; i=ackerleytng@google.com; s=20260225; h=from:subject:message-id; bh=PsXVR6/m97u9rpLVI044KHOk18BufZ3/bkmeJAuyM80=; b=owLHrJvPH+l1S7PSWEo0c0posSa5NI9r+gR/dXJ1kxWlqXVcWCE3yqX5Fz/J11W0Cplhquj3O WVU750zLuu/BG4RBzBcDZY3JbDsFeKGGQXkwf3B0FdWOS3B1Id6bQKC X-Mailer: b4 0.14.3 Message-ID: <20260313-gmem-inplace-conversion-v3-39-5fc12a70ec89@google.com> Subject: [PATCH RFC v3 39/43] KVM: guest_memfd: Apply content modes while setting memory attributes From: Ackerley Tng To: aik@amd.com, andrew.jones@linux.dev, binbin.wu@linux.intel.com, brauner@kernel.org, chao.p.peng@linux.intel.com, david@kernel.org, ira.weiny@intel.com, jmattson@google.com, jroedel@suse.de, jthoughton@google.com, michael.roth@amd.com, oupton@kernel.org, pankaj.gupta@amd.com, qperret@google.com, rick.p.edgecombe@intel.com, rientjes@google.com, shivankg@amd.com, steven.price@arm.com, tabba@google.com, willy@infradead.org, wyihan@google.com, yan.y.zhao@intel.com, forkloop@google.com, pratyush@kernel.org, suzuki.poulose@arm.com, aneesh.kumar@kernel.org, Paolo Bonzini , Sean Christopherson , Thomas Gleixner , Ingo Molnar , Borislav Petkov , Dave Hansen , x86@kernel.org, "H. Peter Anvin" , Steven Rostedt , Masami Hiramatsu , Mathieu Desnoyers , Jonathan Corbet , Shuah Khan , Shuah Khan , Vishal Annapurve , Jason Gunthorpe , Vlastimil Babka Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, linux-trace-kernel@vger.kernel.org, linux-doc@vger.kernel.org, linux-kselftest@vger.kernel.org, Ackerley Tng Content-Type: text/plain; charset="utf-8" Provide defined memory content modes so that KVM can make guarantees about memory content after setting memory attributes, according to userspace requests. Suggested-by: Sean Christoperson Signed-off-by: Ackerley Tng --- Documentation/virt/kvm/api.rst | 40 +++++++++++++++++++++++++++++ include/uapi/linux/kvm.h | 4 +++ virt/kvm/guest_memfd.c | 58 ++++++++++++++++++++++++++++++++++++++++-- 3 files changed, 100 insertions(+), 2 deletions(-) diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst index 15148c80cfdb6..3ec92f8606099 100644 --- a/Documentation/virt/kvm/api.rst +++ b/Documentation/virt/kvm/api.rst @@ -6571,6 +6571,8 @@ Errors: EAGAIN Some page within requested range had unexpected refcounts. The offset of the page will be returned in `error_offset`. ENOMEM Ran out of memory trying to track private/shared state + EOPNOTSUPP There is no way for KVM to guarantee in-memory contents as + requested. ========== =============================================================== KVM_SET_MEMORY_ATTRIBUTES2 is an extension to @@ -6619,6 +6621,44 @@ on the shared pages, such as refcounts taken by get_user_pages(), and try the ioctl again. A possible source of these long term refcounts is if the guest_memfd memory was pinned in IOMMU page tables. +By default, KVM makes no guarantees about the in-memory values after +memory is convert to/from shared/private. Optionally, userspace may +instruct KVM to ensure the contents of memory are zeroed or preserved, +e.g. to enable in-place sharing of data, or as an optimization to +avoid having to re-zero memory when userspace could have relied on the +trusted entity to guarantee the memory will be zeroed as part of the +entire conversion process. + +The content modes available are as follows: + +``KVM_SET_MEMORY_ATTRIBUTES2_ZERO`` + + On conversion, KVM guarantees all entities that have "allowed" + access to the memory will read zeros. E.g. on private to shared + conversion, both trusted and untrusted code will read zeros. + + Zeroing is currently only guaranteed for private-to-shared + conversions, as KVM in general is untrusted and thus cannot + guarantee the guest (or any trusted entity) will read zeros after + conversion. Note, some CoCo implementations do zero memory contents + such that the guest reads zeros after conversion, and the guest may + choose to rely on that behavior. However, that's a contract between + the trusted CoCo entity and the guest, not between KVM and the + guest. + +``KVM_SET_MEMORY_ATTRIBUTES2_PRESERVE`` + + On conversion, KVM guarantees memory contents will be preserved with + respect to the last written unencrypted value. As a concrete + example, if the host writes ``0xbeef`` to shared memory and converts + the memory to private, the guest will also read ``0xbeef``, even if + the in-memory data is encrypted as part of the conversion. And vice + versa, if the guest writes ``0xbeef`` to private memory and then + converts the memory to shared, the host (and guest) will read + ``0xbeef`` (if the memory is accessible). + +TODO: Document CAP after CAP discussion. + See also: :ref: `KVM_SET_MEMORY_ATTRIBUTES`. .. _kvm_run: diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index 29baaa60de35a..0fc9ad4ea0d93 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -1642,6 +1642,10 @@ struct kvm_memory_attributes { /* Available with KVM_CAP_MEMORY_ATTRIBUTES2 */ #define KVM_SET_MEMORY_ATTRIBUTES2 _IOWR(KVMIO, 0xd2, struct kvm_memory_attributes2) +#define KVM_SET_MEMORY_ATTRIBUTES2_MODE_UNSPECIFIED 0 +#define KVM_SET_MEMORY_ATTRIBUTES2_ZERO (1ULL << 0) +#define KVM_SET_MEMORY_ATTRIBUTES2_PRESERVE (1ULL << 1) + struct kvm_memory_attributes2 { union { __u64 address; diff --git a/virt/kvm/guest_memfd.c b/virt/kvm/guest_memfd.c index f23acbca28e54..6bfcb2ed12c61 100644 --- a/virt/kvm/guest_memfd.c +++ b/virt/kvm/guest_memfd.c @@ -674,8 +674,50 @@ int __weak kvm_arch_gmem_apply_content_mode_preserve(struct kvm *kvm, return -EOPNOTSUPP; } +static int kvm_gmem_apply_content_mode_folio(struct kvm *kvm, + struct folio *folio, + uint64_t content_mode) +{ + switch (content_mode) { + case KVM_SET_MEMORY_ATTRIBUTES2_MODE_UNSPECIFIED: + return kvm_arch_gmem_apply_content_mode_unspecified(kvm, folio); + case KVM_SET_MEMORY_ATTRIBUTES2_ZERO: + return kvm_arch_gmem_apply_content_mode_zero(kvm, folio); + case KVM_SET_MEMORY_ATTRIBUTES2_PRESERVE: + return kvm_arch_gmem_apply_content_mode_preserve(kvm, folio); + default: + WARN_ONCE(1, "Unexpected policy requested."); + return -EOPNOTSUPP; + } +} + +static void kvm_gmem_apply_content_mode(struct inode *inode, pgoff_t start, + pgoff_t end, struct kvm *kvm, + uint64_t content_mode) +{ + struct address_space *mapping = inode->i_mapping; + struct folio_batch fbatch; + int i; + + folio_batch_init(&fbatch); + while (filemap_get_folios(mapping, &start, end - 1, &fbatch)) { + + for (i = 0; i < folio_batch_count(&fbatch); ++i) { + struct folio *folio = fbatch.folios[i]; + int ret; + + ret = kvm_gmem_apply_content_mode_folio(kvm, folio, + content_mode); + WARN_ON_ONCE(ret); + } + + folio_batch_release(&fbatch); + } +} + static int __kvm_gmem_set_attributes(struct inode *inode, pgoff_t start, size_t nr_pages, uint64_t attrs, + struct kvm *kvm, uint64_t content_mode, pgoff_t *err_index) { struct address_space *mapping = inode->i_mapping; @@ -689,6 +731,12 @@ static int __kvm_gmem_set_attributes(struct inode *inode, pgoff_t start, filemap_invalidate_lock(mapping); + if (content_mode && + !(kvm_gmem_supported_content_modes(kvm) & content_mode)) { + r = -EOPNOTSUPP; + goto out; + } + mas_init(&mas, mt, start); if (kvm_gmem_range_has_attributes(mt, start, nr_pages, attrs)) { @@ -715,6 +763,8 @@ static int __kvm_gmem_set_attributes(struct inode *inode, pgoff_t start, kvm_gmem_invalidate_begin(inode, start, end); + kvm_gmem_apply_content_mode(inode, start, end, kvm, content_mode); + mas_store_prealloc(&mas, xa_mk_value(attrs)); kvm_gmem_invalidate_end(inode, start, end); @@ -736,7 +786,11 @@ static long kvm_gmem_set_attributes(struct file *file, void __user *argp) if (copy_from_user(&attrs, argp, sizeof(attrs))) return -EFAULT; - if (attrs.flags) + if (attrs.flags & ~(KVM_SET_MEMORY_ATTRIBUTES2_ZERO | + KVM_SET_MEMORY_ATTRIBUTES2_PRESERVE)) + return -EINVAL; + if ((attrs.flags & KVM_SET_MEMORY_ATTRIBUTES2_ZERO) && + (attrs.flags & KVM_SET_MEMORY_ATTRIBUTES2_PRESERVE)) return -EINVAL; if (attrs.error_offset) return -EINVAL; @@ -758,7 +812,7 @@ static long kvm_gmem_set_attributes(struct file *file, void __user *argp) nr_pages = attrs.size >> PAGE_SHIFT; index = attrs.offset >> PAGE_SHIFT; r = __kvm_gmem_set_attributes(inode, index, nr_pages, attrs.attributes, - &err_index); + f->kvm, attrs.flags, &err_index); if (r) { attrs.error_offset = err_index << PAGE_SHIFT; -- 2.53.0.851.ga537e3e6e9-goog