From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-ej1-f73.google.com (mail-ej1-f73.google.com [209.85.218.73]) (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 052973932CC for ; Fri, 1 May 2026 11:19:45 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.218.73 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1777634387; cv=none; b=E33nyykF36zdM5+pxwf8ZfcU4P/cjBmHKUYUeFAbiEbi8QBTeAflrgviL8dyJyqr0iae2DQsjaSrrj/k9hssfyjyvCeRpn9NKyv1nvSFKy2JB9pxktetKJ3i6zgx/ZAIkt3KecRNq8IHDZXQqQR18PXFU1ms9zFscwGv6JOxz6k= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1777634387; c=relaxed/simple; bh=a7l0tjjKDWFABSCDQFyGvRK9SXNivegcz+Anu6dErts=; h=Date:Mime-Version:Message-ID:Subject:From:To:Cc:Content-Type; b=ktOQPtg3Nzk6FPv+3etNe7jvxlqQYic7WpnH4f5jQqcKOkvw5OyLArSDHiCCBjuvDN+AbNTgIrlJfqb2L8hWBNPHGy/t4WRaxCOTCzoqKnZMPLpzZxBGj4o1nfoopYgD3vjDSBtcvk+cZCi2tpvONV+blBZ5ZYt47EL2KYI2s6c= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com; spf=pass smtp.mailfrom=flex--smostafa.bounces.google.com; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b=hCsPqm/K; arc=none smtp.client-ip=209.85.218.73 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--smostafa.bounces.google.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="hCsPqm/K" Received: by mail-ej1-f73.google.com with SMTP id a640c23a62f3a-b8ff07df5e4so156340066b.3 for ; Fri, 01 May 2026 04:19:45 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20251104; t=1777634384; x=1778239184; darn=vger.kernel.org; h=content-transfer-encoding:cc:to:from:subject:message-id :mime-version:date:from:to:cc:subject:date:message-id:reply-to; bh=39s63gE+FaRnrwSJ1nF07Frvg2b04cpO6BVgy2OjLZw=; b=hCsPqm/Kj0TShzEPlNJ0152uWklOqIWKrewsylG1Dd3aBi+yNRXWELLmpj1jKOOUE6 pVH6PiCEOeX5tDKLGX2rBLgG1Z+9wknIvbh4tMRqx9ZykQG8avO7b921Q5nlE9R14oot 6O3O75r/Mitnq894TMgwlqUczaKl2q3Hn2V3QXqWv/gM1qJWWjUYInjJ1D9VAtqeUU1d 4l8yj2NQ5OKOk9DsCT7Y2iKbSVg7MVVktUbI42DpEeF4/fYv0CtZf4fS8/JykjdIQ3DX /KcuJNJkdL80XL5rU6Sws4IRnpxdpZRjL+dpoIfzBV5hyVNt3M9ZqDjFgWDDCvjVhwUD OWBg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1777634384; x=1778239184; h=content-transfer-encoding:cc:to:from:subject:message-id :mime-version:date:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to; bh=39s63gE+FaRnrwSJ1nF07Frvg2b04cpO6BVgy2OjLZw=; b=iHXXm6ysQ1r+KkJj1YhoaqwKGHtiuQKYtzACS13cMqnteBHA80legwas+izNzmAhhz 1DR1DQxU7QsBE3obMvWU6W/HN5VltnleX3WmttGxxX9TYBHGRcs8H4DIck3xETszidAY wmS7fGqMyj2TMbOMlfURmL5LDOs7P6ZZhlxPZhjFS9mq+jv7Gyy/43XbtaDoFbk6gqPk CCagTRolA77S2bC27LcxJi0UowdLnp9plLVcPBt99NquZmZ4HsWTxwlgCenrOHvOAwXH unZb2Gc5hnAWRXBsogdz48Ujq/e7dJRQmTxyyQXCim70gu8Gb4+Abo8kDRz/avssREg0 fv4w== X-Forwarded-Encrypted: i=1; AFNElJ9dD1+Msk7zvPvkcs6tIV7d/lVpOkY97xJhEngnYOj3jnit4+kRK2FJDPdZ2GBPvWi+DFaahqzlbVQ2BQA=@vger.kernel.org X-Gm-Message-State: AOJu0YxqquSufFanXdWTnRshJ3AGnOFCSKwG8n88ywzEfeTxzL0IqrXt jX6H8dWO2CHK+7ODVzrYqKv1cqv9J61clJLTmOSSPsFpW7yRXUtw2GFP3oZnDOlpVoDJLjpWHrk kJJXnZLM9+7w7vw== X-Received: from ejcqw28.prod.google.com ([2002:a17:906:6a1c:b0:ba5:6ff2:e915]) (user=smostafa job=prod-delivery.src-stubby-dispatcher) by 2002:a17:907:980d:b0:bab:ba5a:fcb8 with SMTP id a640c23a62f3a-bbac7e8405emr444971866b.32.1777634384133; Fri, 01 May 2026 04:19:44 -0700 (PDT) Date: Fri, 1 May 2026 11:19:02 +0000 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 X-Mailer: git-send-email 2.54.0.545.g6539524ca2-goog Message-ID: <20260501111928.259252-1-smostafa@google.com> Subject: [PATCH v6 00/25] KVM: arm64: SMMUv3 driver for pKVM (trap and emulate) From: Mostafa Saleh To: linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, kvmarm@lists.linux.dev, iommu@lists.linux.dev Cc: catalin.marinas@arm.com, will@kernel.org, maz@kernel.org, oliver.upton@linux.dev, joey.gouly@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, joro@8bytes.org, jean-philippe@linaro.org, jgg@ziepe.ca, mark.rutland@arm.com, qperret@google.com, tabba@google.com, vdonnefort@google.com, sebastianene@google.com, keirf@google.com, Mostafa Saleh Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable This is v6 of pKVM SMMUv3 support with trap and emulate v1: Implements full fledged pv interface https://lore.kernel.org/kvmarm/20230201125328.2186498-1-jean-philippe@linar= o.org/ v2: Implements full fledged pv interface (+ more features as evtq and s1) https://lore.kernel.org/kvmarm/20241212180423.1578358-1-smostafa@google.com= / v3: Only DMA isolation (using pv) https://lore.kernel.org/kvmarm/20250728175316.3706196-1-smostafa@google.com= / v4: Trap and emulate https://lore.kernel.org/all/20250819215156.2494305-1-smostafa@google.com/ v5: Trap and emulate https://lore.kernel.org/all/20251117184815.1027271-1-smostafa@google.com/ This series is based on the review feedback on v5 + improvements, most notably: - Rebase on ToT which includes the newly merged protected VM support! - Drop non-coherent support to make the patches smaller, this can be added in a later series. - Re-work the io-pgtable-arm split to rely on iommu-pages [Jason] - Use the newly added clock for tracing instead of adding new functions in the hypervisor. - Align nesting support with the upstream driver in terms of supported IPs [Jason] - Keep STE hiltless updating when possible [Jason] - Move some of the refactored code to the c file [Jason] - Add support for evtq and priq tracking - Add extra hardening checks, handle failures and massively reduce the amount of WARN_ONs and other cleanups. - Don=E2=80=99t enforce DMA isolation to not regress pKVM booting. Notes about Sashiko =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D I ran Sashiko locally and it was helpful in discovering problems in the series. However, it still shows large number of critical and high severity issues, I went through them and I believe they are false positive, mainly because (in the order of frequently reported): - It doesn=E2=80=99t understand de-privilege and which data is trusted that the driver populated at boot (keeps complaining about missing checks for zero mmio size..) - It doesn=E2=80=99t understand WARNs are fatal in the hypervisor. - It doesn=E2=80=99t understand that a malicious host can DoS the system an= d pKVM doesn=E2=80=99t guarantee availability - It doesn=E2=80=99t understand the SMMUv3 spec and makes stuff up (eg. ab= out CMD_SYC CS field it makes up an non-existent encoding or wrong semantics for the gbpa register) - It seems to look at one patch at a time and not the whole series, and as the series is written in a way to be bisectable that confuses it. - Sometimes it complains about code which is not related to the change. Fuad is currently working on updating review prompts to make it work better with protected KVM [1] Design: =3D=3D=3D=3D=3D=3D=3D Assumptions: ------------ One of the important points, is that this doesn=E2=80=99t emulate the full SMMUv3 architecture, but only the parts used by Linux kernel, that=E2=80=99s why enablement of this (ARM_SMMU_V3_PKVM) depends on (ARM_SMMU_V3=3Dy) so we are sure of the driver behaviour. Any new change in the driver will likely trigger a WARN_ON ending up in panic, that will require to support also in the hypervisor. Most notable assumptions: - Changing of stream table format/size or l2 pointers is not allowed after initialization. - leaf=3D0 CFGI is not allowed. - CFGI_ALL with any value but 31 is not allowed. - Some commands which are not used are not allowed. - Values set in ARM_SMMU_CR1 are hardcoded and don't change. Emulation logic mainly targets: 1) Command Queue ---------------- At boot time, the hypervisor will allocate a shadow command queue (doesn=E2=80=99t need to match the host size) which then sets up in HW, the= n it will trap access to i) ARM_SMMU_CMDQ_BASE That can only be written when the cmdq is disabled. Then on enable, the hypervisor will put the host command queue in a shared state to avoid transition into the hypervisor or VMs. It will be unshared with the cmdq is disabled ii) ARM_SMMU_CMDQ_PROD Trigger emulation code, where the hypervisor will copy the commands between cons and prod, of the host queue and sanitise them (mostly WARNs if the host is malicious and issuing commands it shouldn=E2=80=99t) then eagerly consume them, updating the host cons. iii) ARM_SMMU_CMDQ_CONS No much logic, just return the emulated cons + error bits. 2) Stream table --------------- Similar to the command queue, the first level is allocated at boot with max possible size, then the hypervisor will trap access to: i) ARM_SMMU_STRTAB_BASE/ARM_SMMU_STRTAB_BASE_CFG: Keep track of the stream table to put it in a shared state. On CFGI_STE, the hypervisor will read the STE in scope from the host copy, shadow L2 pointers if needed and attach stage-2. 3) GBPA ------- The hypervisor will set GBPA to abort at boot, then any read from the host will return ABORT and writes are ignored. If the host tries to clear GBPA, it will look like GBPA is refusing to update and time out. 4) EVTQ and PRIDQ No shadowing needed for those queues, but the hypervisor needs to keep track of them to put them in a shared state so they can=E2=80=99t be used b= y the host or the hypervisor. Bisectibility: =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D I wrote the patches where most of them are bisectable at run time (so we can run with a prefix of the series till MMIO emulation, cmdq emulation, STE or full nested) that was very helpful in debugging, and I kept it like this to make debugging easier. Constraints: =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D 1) Discovery: ------------- Only device trees are supported at the moment. I don=E2=80=99t usually use ACPI, but I can look into adding that later. (not make this series bigger) 1) Shadow page table -------------------- Uses page granularity (leaf) for memory, that=E2=80=99s because of the lack of split_block_unmap() logic. I am currently looking into the possibility of sharing page tables, if that turned complicated (as expected) it might be worth to re-add this logic Boot and Probe ordering: =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D The main SMMUv3 MUST be only bound/probed after KVM fully initialises so it can set up the MMIO emulation. The KVM SMMUv3 driver is loaded early before KVM init so it can register itself, during that point it will probe all the SMMUs from the platform bus and bind them to the driver. Then at a later init call it will create an auxiliary device per SMMU, that the main driver will probe. The main driver still relies on this device(parent) for all driver activity. (Check comment in patch 14. Future work =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D 1) Sharing page tables will be an interesting optimization, but requires dealing with stage-2 page faults (which are handled by the kernel), BBM and possibly more complexity. 2) There is currently ongoing work to enable RPM, that will possibly enable/disable the SMMU frequently, we might need some optimizations to avoid re-shadowing the CMDQ/STE unnecessarily. 3) Add support for non-coherent SMMUs 4) Optimizations (as using block mappings for memory) Patches overview =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D The patches are split as follows: Patches 01-02: Core hypervisor: Dealing with MMIO and timers. Patches 04-06: Refactoring of io-pgtable-arm and SMMUv3 driver. Patches 07-10: Hypervisor IOMMU core: pagetable management, dabts. Patches 11-25: KVM SMMUv3 code. Tested on Lenovo IdeaCentre mini X and Qemu. A development branch can be found at [2] [1] https://github.com/ftabba/review-prompts/commits/local-arm64-kvm/ [2] https://android-kvm.googlesource.com/linux/+/refs/heads/pkvm-smmu-v6 Jean-Philippe Brucker (1): iommu/arm-smmu-v3-kvm: Add SMMUv3 driver Mostafa Saleh (24): KVM: arm64: Generalize trace clock KVM: arm64: Donate MMIO to the hypervisor iommu/arm-smmu-v3: Split code with hyp iommu/arm-smmu-v3: Move TLB range invalidation into common code iommu/arm-smmu-v3: Move IDR parsing to common functions iommu/io-pgtable-arm: Rework to use the iommu-pages API KVM: arm64: iommu: Introduce IOMMU driver infrastructure KVM: arm64: iommu: Shadow host stage-2 page table KVM: arm64: iommu: Add memory pool KVM: arm64: iommu: Support DABT for IOMMU iommu/arm-smmu-v3-kvm: Add the kernel driver iommu/arm-smmu-v3-kvm: Probe SMMU HW iommu/arm-smmu-v3-kvm: Add MMIO emulation iommu/arm-smmu-v3-kvm: Shadow the command queue iommu/arm-smmu-v3-kvm: Add CMDQ functions iommu/arm-smmu-v3-kvm: Emulate CMDQ for host iommu/arm-smmu-v3-kvm: Shadow stream table iommu/arm-smmu-v3-kvm: Shadow STEs iommu/arm-smmu-v3-kvm: Share other queues iommu/arm-smmu-v3-kvm: Emulate GBPA iommu/io-pgtable-arm: Support io-pgtable-arm in the hypervisor iommu/arm-smmu-v3-kvm: Shadow the CPU stage-2 page table iommu/arm-smmu-v3-kvm: Enable nesting KVM: arm64: Add documentation for pKVM DMA isolation .../admin-guide/kernel-parameters.txt | 4 + Documentation/virt/kvm/arm/pkvm.rst | 19 +- arch/arm64/include/asm/kvm_host.h | 6 + arch/arm64/kvm/Makefile | 2 +- arch/arm64/kvm/hyp/include/nvhe/clock.h | 11 +- arch/arm64/kvm/hyp/include/nvhe/iommu.h | 23 + arch/arm64/kvm/hyp/include/nvhe/mem_protect.h | 4 + arch/arm64/kvm/hyp/nvhe/Makefile | 13 +- arch/arm64/kvm/hyp/nvhe/clock.c | 44 +- arch/arm64/kvm/hyp/nvhe/iommu/iommu.c | 156 ++ arch/arm64/kvm/hyp/nvhe/mem_protect.c | 169 ++- arch/arm64/kvm/hyp/nvhe/setup.c | 20 + arch/arm64/kvm/hyp/nvhe/trace.c | 4 +- arch/arm64/kvm/hyp/pgtable.c | 9 +- arch/arm64/kvm/iommu.c | 57 + arch/arm64/kvm/pkvm.c | 1 + drivers/iommu/arm/Kconfig | 9 + drivers/iommu/arm/arm-smmu-v3/Makefile | 3 +- .../arm/arm-smmu-v3/arm-smmu-v3-common-lib.c | 224 +++ .../iommu/arm/arm-smmu-v3/arm-smmu-v3-kvm.c | 232 +++ drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 387 +---- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h | 150 ++ .../iommu/arm/arm-smmu-v3/pkvm/arm-smmu-v3.c | 1250 +++++++++++++++++ .../iommu/arm/arm-smmu-v3/pkvm/arm_smmu_v3.h | 67 + drivers/iommu/io-pgtable-arm.c | 68 +- drivers/iommu/io-pgtable-arm.h | 6 + drivers/iommu/iommu-pages.h | 99 ++ 27 files changed, 2668 insertions(+), 369 deletions(-) create mode 100644 arch/arm64/kvm/hyp/include/nvhe/iommu.h create mode 100644 arch/arm64/kvm/hyp/nvhe/iommu/iommu.c create mode 100644 arch/arm64/kvm/iommu.c create mode 100644 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-common-lib.c create mode 100644 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-kvm.c create mode 100644 drivers/iommu/arm/arm-smmu-v3/pkvm/arm-smmu-v3.c create mode 100644 drivers/iommu/arm/arm-smmu-v3/pkvm/arm_smmu_v3.h --=20 2.54.0.545.g6539524ca2-goog