From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-wr1-f74.google.com (mail-wr1-f74.google.com [209.85.221.74]) (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 04A3D442124 for ; Wed, 15 Jul 2026 11:59:16 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.221.74 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784116759; cv=none; b=j6l/BQp3E/QZU8E2NMlTXKrn+bJqkhYntaHV7wUU1qXC05kMeN0owJ936CfQnQLykDyYgeVKB7TNCwMI+ltqkyuyNbaBUZVP4mqtA/x14CG3VpfnfWR9VksyQI3/bNu11sDYZthVfladneg/iGivOpD2krL1mBDFYNbY+gRq8RA= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1784116759; c=relaxed/simple; bh=Ep2a0G/52eGsYrTUgu5RVlmketd92FEfjFvjwS2XdcA=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Cc:Content-Type; b=OEmO+IUVvVvrPOTjQhhGgONVIztzEdREWNXlIUWuULKpyVllJSsYqqp7hMbJpSnjPPEXSCP9krBNfc6xVAoUiE54+NKsOufgT+uGLlQztiXjktchvViWn4W5fzy/gLv18MvZz8EpaVwsdYeAa+HzrnjhFRgVvGPYkXB4DwJaOCU= 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=KAnAtEoe; arc=none smtp.client-ip=209.85.221.74 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="KAnAtEoe" Received: by mail-wr1-f74.google.com with SMTP id ffacd0b85a97d-47f3e6cc5c4so1198828f8f.3 for ; Wed, 15 Jul 2026 04:59:16 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20251104; t=1784116755; x=1784721555; darn=lists.linux.dev; h=content-transfer-encoding:content-type:cc:to:from:subject :message-id:references:mime-version:in-reply-to:date:from:to:cc :subject:date:message-id:reply-to:content-type; bh=M3eZs1fgfcJqE0urG21tDoGKs0yo1icKieB+aaaZBxk=; b=KAnAtEoe3e+YV4/sZS2A1isiZ5UjVOPGlkaqdt2OdX+icS5/kyDWpajKgORfQElNuW 7l+dJlkWqq6xJTR1GFGLMzNZJTM65ctWQ4/+AVZrXbJ38qQ7iVWg65MObXTSBGMxQskk fmCg2599lL6jQ20r99+6h09U5it+bf/eDTCLryIYZkjl7k3Fi2xdenwZJf/VGW4bHMxn ksZQ8ZbVOyjE9AxIrnGHK2k9b5mc0PJ7zyi6DfoEuxWOqyOGanhiP000Zj1NQ3uvXNuX 0fG2JUiYq1/3c4Dx41r0UXb1PHkakTDIqpeNYcIXvxvcIcW42W56Iq9UKGBpJTuLSBde xPjQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1784116755; x=1784721555; h=content-transfer-encoding:content-type: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 :content-type; bh=M3eZs1fgfcJqE0urG21tDoGKs0yo1icKieB+aaaZBxk=; b=Bkinef+13gT+898Pmx/qg4rfU8Hr3eP7HXOoOOPD9bysFvAMNEVARDKu9YhbOCZTyM 0CN+0COwsGHLWNt1ELQsbEONhCn0z2mjp7cpObzBPHjfYauKD0bK4uCUueXCE6zRe1bk N5W6bR9bH2a+872XfgdZ2dTvwH5wVJT0IrZ52OeKX63VVCEGbdpxRrLupLjE51VoR8Q8 xEfKiCSngtaIlNGq4ZEbLYncHN/mq3bESEcU01D9+jkED65/G1r2LwsJ13qgHNGofWCc VG8uuvMHXOtlFVuwuFarNtZ51PuejfTFQef3prItmc6fKN3zVSQHpvSvrdDnsYV+qNIR 3HPw== X-Forwarded-Encrypted: i=1; AHgh+Rr3yXY/2ni01quj3s7Ufon+9C7MHH7pZJCsZpbpKgY98mBGROHa0SqiLJ+R9JStC4L63fJT+g==@lists.linux.dev X-Gm-Message-State: AOJu0YwaUPfoP/1Y53FTIAMgXiTDJIHcvjwiTUq4Z7G0p5UcICmJkMGF 3NBLbKofFiUI2ZYfOwVpATaLUjTR/AbaosXiFXcWyJXxdBYp/yCmh5pYyZzQ6qRyKS59XcfAu3d VTYiiQY1J0FimXA== X-Received: from wmdn11.prod.google.com ([2002:a05:600c:294b:b0:494:415:3ecd]) (user=smostafa job=prod-delivery.src-stubby-dispatcher) by 2002:a05:600c:621b:b0:493:faf3:3ead with SMTP id 5b1f17b1804b1-4952cc3c566mr74288995e9.4.1784116755156; Wed, 15 Jul 2026 04:59:15 -0700 (PDT) Date: Wed, 15 Jul 2026 11:58:43 +0000 In-Reply-To: <20260715115906.2664882-1-smostafa@google.com> Precedence: bulk X-Mailing-List: iommu@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20260715115906.2664882-1-smostafa@google.com> X-Mailer: git-send-email 2.55.0.141.g00534a21ce-goog Message-ID: <20260715115906.2664882-3-smostafa@google.com> Subject: [PATCH v7 02/24] KVM: arm64: Donate MMIO to the hypervisor 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, 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 Add a function to donate MMIO to the hypervisor so IOMMU hypervisor drivers can protect and access the MMIO of IOMMUs. As donating MMIO is very rare, and we don=E2=80=99t need to encode the full state, it=E2=80=99s reasonable to have a separate function to do this. It will init the host s2 page table with an invalid leaf with the owner ID to prevent the host from mapping the page on faults. Also, prevent kvm_pgtable_stage2_unmap() from removing owner ID from stage-2 PTEs, as this can be triggered from recycle logic under memory pressure. There is no code relying on this, as all ownership changes is done via kvm_pgtable_stage2_set_owner() For the error path in IOMMU drivers, add a function to donate MMIO back from hyp to host. However, that leaks the hypervisor virtual address range which should be acceptable as this is quite rare and it matches the behaviour of fix_map/block. Signed-off-by: Mostafa Saleh --- arch/arm64/kvm/hyp/include/nvhe/mem_protect.h | 7 ++ arch/arm64/kvm/hyp/nvhe/mem_protect.c | 91 ++++++++++++++++++- arch/arm64/kvm/hyp/pgtable.c | 11 +-- 3 files changed, 102 insertions(+), 7 deletions(-) diff --git a/arch/arm64/kvm/hyp/include/nvhe/mem_protect.h b/arch/arm64/kvm= /hyp/include/nvhe/mem_protect.h index 29935c7da1de..51b0eb3844a9 100644 --- a/arch/arm64/kvm/hyp/include/nvhe/mem_protect.h +++ b/arch/arm64/kvm/hyp/include/nvhe/mem_protect.h @@ -36,6 +36,13 @@ int __pkvm_guest_share_host(struct pkvm_hyp_vcpu *vcpu, = u64 gfn); int __pkvm_guest_unshare_host(struct pkvm_hyp_vcpu *vcpu, u64 gfn); int __pkvm_host_unshare_hyp(u64 pfn); int __pkvm_host_donate_hyp(u64 pfn, u64 nr_pages); +/* + * Donate MMIO range to the hypervisor, it will be mapped in the hyperviso= r's + * private range and unmapped from the host stage-2. + */ +int __pkvm_host_donate_hyp_mmio(phys_addr_t addr, size_t size, unsigned lo= ng *haddr); +/* Remaps MMIO range in the host, typically used in error path. */ +int __pkvm_hyp_donate_host_mmio(phys_addr_t addr, size_t size); int __pkvm_hyp_donate_host(u64 pfn, u64 nr_pages); int __pkvm_host_share_ffa(u64 pfn, u64 nr_pages); int __pkvm_host_unshare_ffa(u64 pfn, u64 nr_pages); diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvh= e/mem_protect.c index 4e329e39a695..d803b3dd4cb4 100644 --- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c +++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c @@ -378,7 +378,11 @@ static int host_stage2_unmap_dev_all(void) u64 addr =3D 0; int i, ret; =20 - /* Unmap all non-memory regions to recycle the pages */ + /* + * Unmap all non-memory regions to recycle the pages. + * That relies on kvm_pgtable_stage2_unmap() not clearing + * counted PTEs which include hypervisor MMIO. + */ for (i =3D 0; i < hyp_memblock_nr; i++, addr =3D reg->base + reg->size) { reg =3D &hyp_memory[i]; ret =3D kvm_pgtable_stage2_unmap(pgt, addr, reg->base - addr); @@ -1119,6 +1123,91 @@ int __pkvm_host_donate_hyp(u64 pfn, u64 nr_pages) return ret; } =20 +int __pkvm_host_donate_hyp_mmio(phys_addr_t addr, size_t size, unsigned lo= ng *haddr) +{ + kvm_pte_t pte; + u64 offset; + int ret; + + /* Only before de-privilege. */ + if (static_branch_unlikely(&kvm_protected_mode_initialized)) + return -EPERM; + + if (!PAGE_ALIGNED(addr | size) || + !pfn_range_is_valid(hyp_phys_to_pfn(addr), size >> PAGE_SHIFT)) + return -EINVAL; + + ret =3D __pkvm_create_private_mapping(addr, size, PAGE_HYP_DEVICE, haddr)= ; + if (ret) + return ret; + + host_lock_component(); + for (offset =3D 0; offset < size; offset +=3D PAGE_SIZE) { + if (addr_is_memory(addr + offset)) { + ret =3D -EINVAL; + goto unlock; + } + ret =3D kvm_pgtable_get_leaf(&host_mmu.pgt, addr + offset, &pte, NULL); + if (ret) + goto unlock; + if (pte && !kvm_pte_valid(pte)) { + ret =3D -EPERM; + goto unlock; + } + } + /* + * We set HYP as the owner of the MMIO pages in the host stage-2, for: + * - host aborts: host_stage2_adjust_range() would fail for invalid non z= ero PTEs. + * - recycle under memory pressure: host_stage2_unmap_dev_all() would cal= l + * kvm_pgtable_stage2_unmap() which will not clear non zero invalid pte= s (counted). + * - other MMIO donation: Would fail as we check that the PTE is valid or= empty. + */ + ret =3D host_stage2_try(kvm_pgtable_stage2_annotate, &host_mmu.pgt, + addr, size, &host_s2_pool, + KVM_HOST_INVALID_PTE_TYPE_DONATION, + FIELD_PREP(KVM_HOST_DONATION_PTE_OWNER_MASK, PKVM_ID_HYP)); +unlock: + host_unlock_component(); + return ret; +} + +int __pkvm_hyp_donate_host_mmio(phys_addr_t addr, size_t size) +{ + kvm_pte_t pte; + u64 offset; + int ret =3D 0; + + if (static_branch_unlikely(&kvm_protected_mode_initialized)) + return -EPERM; + + if (!PAGE_ALIGNED(addr | size) || + !pfn_range_is_valid(hyp_phys_to_pfn(addr), size >> PAGE_SHIFT)) + return -EINVAL; + + host_lock_component(); + for (offset =3D 0; offset < size; offset +=3D PAGE_SIZE) { + if (addr_is_memory(addr + offset)) { + ret =3D -EINVAL; + goto unlock; + } + ret =3D kvm_pgtable_get_leaf(&host_mmu.pgt, addr + offset, &pte, NULL); + if (ret) + goto unlock; + if (!pte || kvm_pte_valid(pte)) { + ret =3D -EINVAL; + goto unlock; + } + if (FIELD_GET(KVM_HOST_DONATION_PTE_OWNER_MASK, pte) !=3D PKVM_ID_HYP) { + ret =3D -EPERM; + goto unlock; + } + } + WARN_ON(host_stage2_idmap_locked(addr, size, PKVM_HOST_MMIO_PROT)); +unlock: + host_unlock_component(); + return ret; +} + int __pkvm_hyp_donate_host(u64 pfn, u64 nr_pages) { u64 phys =3D hyp_pfn_to_phys(pfn); diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index 91a7dfad6686..3073184cf6ad 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -1161,13 +1161,12 @@ static int stage2_unmap_walker(const struct kvm_pgt= able_visit_ctx *ctx, kvm_pte_t *childp =3D NULL; bool need_flush =3D false; =20 - if (!kvm_pte_valid(ctx->old)) { - if (stage2_pte_is_counted(ctx->old)) { - kvm_clear_pte(ctx->ptep); - mm_ops->put_page(ctx->ptep); - } + /* + * That also ignores stage2_pte_is_counted() instead of clearing + * the PTE as the MMIO can be owned by the hypervisor. + */ + if (!kvm_pte_valid(ctx->old)) return 0; - } =20 if (kvm_pte_table(ctx->old, ctx->level)) { childp =3D kvm_pte_follow(ctx->old, mm_ops); --=20 2.55.0.141.g00534a21ce-goog