From nobody Tue Nov 26 22:25:59 2024 Received: from mail-pl1-f174.google.com (mail-pl1-f174.google.com [209.85.214.174]) (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 2B6021C4A14 for ; Wed, 16 Oct 2024 06:52:39 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.174 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729061562; cv=none; b=TeWXiyZbsRO0+VvoPxfavfTenhqHyaFwG40mZGv0uvKJhhLwdudzHhv9Kcw9BVabJLjY4AePi53Vgw2dMCnGMMP/v212qLxV3It+y+noOjuVkEg9X4oZVWfIXRt1gtpvGqRsELJ5jXtQ3HPqqZHqmK+MMOOs6MoSnE/bJ01MK6U= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729061562; c=relaxed/simple; bh=1zL/Kuu3X33rbK/O6HcBTigetZJjvb1q9lmL+GbrvQs=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version:Content-Type; b=lNW+kVpO9u5AldHby58OffmMbxijFD5BLuXvt/++QtP+jp5Hv5NGvmACV7Qh0zG7+Kv6tgKolqbE75Ke+7lonfacfMqdRVTUo7kEJfPPm/Dh7W2FvoXyqYeLzTVBku/8eESxDkfVoiItQi68uOSkoabB8bVtmoQ/JaIdKP9ii30= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=rivosinc.com; spf=pass smtp.mailfrom=rivosinc.com; dkim=pass (2048-bit key) header.d=rivosinc-com.20230601.gappssmtp.com header.i=@rivosinc-com.20230601.gappssmtp.com header.b=KP+vjt2J; arc=none smtp.client-ip=209.85.214.174 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=rivosinc.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=rivosinc.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=rivosinc-com.20230601.gappssmtp.com header.i=@rivosinc-com.20230601.gappssmtp.com header.b="KP+vjt2J" Received: by mail-pl1-f174.google.com with SMTP id d9443c01a7336-20c6f492d2dso69250225ad.0 for ; Tue, 15 Oct 2024 23:52:39 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rivosinc-com.20230601.gappssmtp.com; s=20230601; t=1729061559; x=1729666359; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=1pOqmREebTbMBHRQyF403IO1qF240RYWkssV4UoxrnE=; b=KP+vjt2JWYDhebQcOW3yAa3KXa2hHbK25qD55ICmYotqREn0QPbAuHBh37aaUgnjLR vZpGICnye0xo976MizDmVAme+9v6geR9ECxCo2o/pnsPmL9ZNJ7l7u2DAP40oCx+r6yV Qjd/24ZhE3cVQK3SC9UHwHftl1p5RG46LYuakyqQGBtCoNsYKUytQ0Ce7GWZTzbxW0fA 50XOilmHkreBzN6mmq4FxYrFUkHeKN1SBMMOITW7jr76Og20Le+ZoiIm84ETvL9LD9/g eZG4uGyIQJimbSe8iIyeH/Dv2LHqr8PgdrfWZyPiMZknumPZ2575Slu0X9YdrUbRM7Tt Ig5w== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1729061559; x=1729666359; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=1pOqmREebTbMBHRQyF403IO1qF240RYWkssV4UoxrnE=; b=XoQXjEPJkIQ11nTPjriwU7UJg4MV5TEjVu/nqhbz/D3B3/kTd56+XuRZH5VZxdAsmB Wtk35rVcQhXCsutFy1CVJRjCVBfiiCakKFm3w2LIdQPsN/5CZvXZVM75lP2wcmRbp44r LFjDZnkq4moMtha3wtZ6RfsPVgCjFyAl6CVEFe39ssh5KDTiufgHrGwfpKNkT/x0Wx15 5G7af9r+OoCG4AuDdVmCGZzVweFQYZrm38PFhRX6mLVY/BDulj4pqqn57lmx9IO8HFGp IfAAQNRufVxOqVOavZ8tP63KB8qgpZNW7NoIJl16MwAAnHqTyy+h3m4NB8nvfwNEFT14 b9QA== X-Forwarded-Encrypted: i=1; AJvYcCVLuRk6hRIeBIq0veOdL7bO3wNalXj+5CVA1MimXO9vPCcxafvBe9d3GwSqIKA14ix21k28tQxcJmVA/fs=@vger.kernel.org X-Gm-Message-State: AOJu0YwHvRymj8a4obeM3SEOn8cJ55xa4lB/0n1ho3ZFZlwydecJN+hd RdkGbc7OnPLkkBy7nO0L2PvPkIMdPIRxgWjkTak+9UrKnm/YBxJiQwhFvTSRPMw= X-Google-Smtp-Source: AGHT+IHVqGHPQPmHj33ldnVJAxdMliYuGfl0It0cH6NRhHGrH9kRbfocuGqC0VyenGM0nnDBMTKUtA== X-Received: by 2002:a05:6a21:3998:b0:1d5:2a8d:1f24 with SMTP id adf61e73a8af0-1d8c95d56d1mr21357086637.25.1729061559330; Tue, 15 Oct 2024 23:52:39 -0700 (PDT) Received: from tjeznach.ba.rivosinc.com ([64.71.180.162]) by smtp.gmail.com with ESMTPSA id d2e1a72fcca58-71e77370a83sm2410762b3a.5.2024.10.15.23.52.37 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 15 Oct 2024 23:52:38 -0700 (PDT) From: Tomasz Jeznach To: Joerg Roedel , Will Deacon , Robin Murphy , Paul Walmsley Cc: Palmer Dabbelt , Albert Ou , Anup Patel , Sunil V L , Nick Kossifidis , Sebastien Boeuf , Rob Herring , Krzysztof Kozlowski , Conor Dooley , devicetree@vger.kernel.org, iommu@lists.linux.dev, linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org, linux@rivosinc.com, Tomasz Jeznach , Lu Baolu , Zong Li Subject: [PATCH v10 7/7] iommu/riscv: Paging domain support Date: Tue, 15 Oct 2024 23:52:19 -0700 Message-Id: <1109202d389f51c7121cb1460eb2f21429b9bd5d.1729059707.git.tjeznach@rivosinc.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: References: Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Introduce first-stage address translation support. Page table configured by the IOMMU driver will use the highest mode implemented by the hardware, unless not known at the domain allocation time falling back to the CPU=E2=80=99s MMU page mode. This change introduces IOTINVAL.VMA command, required to invalidate any cached IOATC entries after mapping is updated and/or removed from the paging domain. Invalidations for the non-leaf page entries use IOTINVAL for all addresses assigned to the protection domain for hardware not supporting more granular non-leaf page table cache invalidations. Reviewed-by: Lu Baolu Reviewed-by: Zong Li Signed-off-by: Tomasz Jeznach --- drivers/iommu/riscv/iommu.c | 617 +++++++++++++++++++++++++++++++++++- 1 file changed, 609 insertions(+), 8 deletions(-) diff --git a/drivers/iommu/riscv/iommu.c b/drivers/iommu/riscv/iommu.c index 0c6263101815..8a05def774bd 100644 --- a/drivers/iommu/riscv/iommu.c +++ b/drivers/iommu/riscv/iommu.c @@ -41,6 +41,10 @@ #define dev_to_iommu(dev) \ iommu_get_iommu_dev(dev, struct riscv_iommu_device, iommu) =20 +/* IOMMU PSCID allocation namespace. */ +static DEFINE_IDA(riscv_iommu_pscids); +#define RISCV_IOMMU_MAX_PSCID (BIT(20) - 1) + /* Device resource-managed allocations */ struct riscv_iommu_devres { void *addr; @@ -791,6 +795,197 @@ static int riscv_iommu_iodir_set_mode(struct riscv_io= mmu_device *iommu, return 0; } =20 +/* This struct contains protection domain specific IOMMU driver data. */ +struct riscv_iommu_domain { + struct iommu_domain domain; + struct list_head bonds; + spinlock_t lock; /* protect bonds list updates. */ + int pscid; + bool amo_enabled; + int numa_node; + unsigned int pgd_mode; + unsigned long *pgd_root; +}; + +#define iommu_domain_to_riscv(iommu_domain) \ + container_of(iommu_domain, struct riscv_iommu_domain, domain) + +/* Private IOMMU data for managed devices, dev_iommu_priv_* */ +struct riscv_iommu_info { + struct riscv_iommu_domain *domain; +}; + +/* + * Linkage between an iommu_domain and attached devices. + * + * Protection domain requiring IOATC and DevATC translation cache invalida= tions, + * should be linked to attached devices using a riscv_iommu_bond structure. + * Devices should be linked to the domain before first use and unlinked af= ter + * the translations from the referenced protection domain can no longer be= used. + * Blocking and identity domains are not tracked here, as the IOMMU hardwa= re + * does not cache negative and/or identity (BARE mode) translations, and D= evATC + * is disabled for those protection domains. + * + * The device pointer and IOMMU data remain stable in the bond struct after + * _probe_device() where it's attached to the managed IOMMU, up to the + * completion of the _release_device() call. The release of the bond struc= ture + * is synchronized with the device release. + */ +struct riscv_iommu_bond { + struct list_head list; + struct rcu_head rcu; + struct device *dev; +}; + +static int riscv_iommu_bond_link(struct riscv_iommu_domain *domain, + struct device *dev) +{ + struct riscv_iommu_device *iommu =3D dev_to_iommu(dev); + struct riscv_iommu_bond *bond; + struct list_head *bonds; + + bond =3D kzalloc(sizeof(*bond), GFP_KERNEL); + if (!bond) + return -ENOMEM; + bond->dev =3D dev; + + /* + * List of devices attached to the domain is arranged based on + * managed IOMMU device. + */ + + spin_lock(&domain->lock); + list_for_each(bonds, &domain->bonds) + if (dev_to_iommu(list_entry(bonds, struct riscv_iommu_bond, list)->dev) = =3D=3D iommu) + break; + list_add_rcu(&bond->list, bonds); + spin_unlock(&domain->lock); + + /* Synchronize with riscv_iommu_iotlb_inval() sequence. See comment below= . */ + smp_mb(); + + return 0; +} + +static void riscv_iommu_bond_unlink(struct riscv_iommu_domain *domain, + struct device *dev) +{ + struct riscv_iommu_device *iommu =3D dev_to_iommu(dev); + struct riscv_iommu_bond *bond, *found =3D NULL; + struct riscv_iommu_command cmd; + int count =3D 0; + + if (!domain) + return; + + spin_lock(&domain->lock); + list_for_each_entry(bond, &domain->bonds, list) { + if (found && count) + break; + else if (bond->dev =3D=3D dev) + found =3D bond; + else if (dev_to_iommu(bond->dev) =3D=3D iommu) + count++; + } + if (found) + list_del_rcu(&found->list); + spin_unlock(&domain->lock); + kfree_rcu(found, rcu); + + /* + * If this was the last bond between this domain and the IOMMU + * invalidate all cached entries for domain's PSCID. + */ + if (!count) { + riscv_iommu_cmd_inval_vma(&cmd); + riscv_iommu_cmd_inval_set_pscid(&cmd, domain->pscid); + riscv_iommu_cmd_send(iommu, &cmd); + + riscv_iommu_cmd_sync(iommu, RISCV_IOMMU_IOTINVAL_TIMEOUT); + } +} + +/* + * Send IOTLB.INVAL for whole address space for ranges larger than 2MB. + * This limit will be replaced with range invalidations, if supported by + * the hardware, when RISC-V IOMMU architecture specification update for + * range invalidations update will be available. + */ +#define RISCV_IOMMU_IOTLB_INVAL_LIMIT (2 << 20) + +static void riscv_iommu_iotlb_inval(struct riscv_iommu_domain *domain, + unsigned long start, unsigned long end) +{ + struct riscv_iommu_bond *bond; + struct riscv_iommu_device *iommu, *prev; + struct riscv_iommu_command cmd; + unsigned long len =3D end - start + 1; + unsigned long iova; + + /* + * For each IOMMU linked with this protection domain (via bonds->dev), + * an IOTLB invaliation command will be submitted and executed. + * + * Possbile race with domain attach flow is handled by sequencing + * bond creation - riscv_iommu_bond_link(), and device directory + * update - riscv_iommu_iodir_update(). + * + * PTE Update / IOTLB Inval Device attach & directory update + * -------------------------- -------------------------- + * update page table entries add dev to the bond list + * FENCE RW,RW FENCE RW,RW + * For all IOMMUs: (can be empty) Update FSC/PSCID + * FENCE IOW,IOW FENCE IOW,IOW + * IOTLB.INVAL IODIR.INVAL + * IOFENCE.C + * + * If bond list is not updated with new device, directory context will + * be configured with already valid page table content. If an IOMMU is + * linked to the protection domain it will receive invalidation + * requests for updated page table entries. + */ + smp_mb(); + + rcu_read_lock(); + + prev =3D NULL; + list_for_each_entry_rcu(bond, &domain->bonds, list) { + iommu =3D dev_to_iommu(bond->dev); + + /* + * IOTLB invalidation request can be safely omitted if already sent + * to the IOMMU for the same PSCID, and with domain->bonds list + * arranged based on the device's IOMMU, it's sufficient to check + * last device the invalidation was sent to. + */ + if (iommu =3D=3D prev) + continue; + + riscv_iommu_cmd_inval_vma(&cmd); + riscv_iommu_cmd_inval_set_pscid(&cmd, domain->pscid); + if (len && len < RISCV_IOMMU_IOTLB_INVAL_LIMIT) { + for (iova =3D start; iova < end; iova +=3D PAGE_SIZE) { + riscv_iommu_cmd_inval_set_addr(&cmd, iova); + riscv_iommu_cmd_send(iommu, &cmd); + } + } else { + riscv_iommu_cmd_send(iommu, &cmd); + } + prev =3D iommu; + } + + prev =3D NULL; + list_for_each_entry_rcu(bond, &domain->bonds, list) { + iommu =3D dev_to_iommu(bond->dev); + if (iommu =3D=3D prev) + continue; + + riscv_iommu_cmd_sync(iommu, RISCV_IOMMU_IOTINVAL_TIMEOUT); + prev =3D iommu; + } + rcu_read_unlock(); +} + #define RISCV_IOMMU_FSC_BARE 0 =20 /* @@ -810,10 +1005,28 @@ static void riscv_iommu_iodir_update(struct riscv_io= mmu_device *iommu, { struct iommu_fwspec *fwspec =3D dev_iommu_fwspec_get(dev); struct riscv_iommu_dc *dc; + struct riscv_iommu_command cmd; + bool sync_required =3D false; u64 tc; int i; =20 - /* Device context invalidation ignored for now. */ + for (i =3D 0; i < fwspec->num_ids; i++) { + dc =3D riscv_iommu_get_dc(iommu, fwspec->ids[i]); + tc =3D READ_ONCE(dc->tc); + if (!(tc & RISCV_IOMMU_DC_TC_V)) + continue; + + WRITE_ONCE(dc->tc, tc & ~RISCV_IOMMU_DC_TC_V); + + /* Invalidate device context cached values */ + riscv_iommu_cmd_iodir_inval_ddt(&cmd); + riscv_iommu_cmd_iodir_set_did(&cmd, fwspec->ids[i]); + riscv_iommu_cmd_send(iommu, &cmd); + sync_required =3D true; + } + + if (sync_required) + riscv_iommu_cmd_sync(iommu, RISCV_IOMMU_IOTINVAL_TIMEOUT); =20 /* * For device context with DC_TC_PDTV =3D 0, translation attributes valid= bit @@ -829,15 +1042,388 @@ static void riscv_iommu_iodir_update(struct riscv_i= ommu_device *iommu, /* Update device context, write TC.V as the last step. */ dma_wmb(); WRITE_ONCE(dc->tc, tc); + + /* Invalidate device context after update */ + riscv_iommu_cmd_iodir_inval_ddt(&cmd); + riscv_iommu_cmd_iodir_set_did(&cmd, fwspec->ids[i]); + riscv_iommu_cmd_send(iommu, &cmd); } + + riscv_iommu_cmd_sync(iommu, RISCV_IOMMU_IOTINVAL_TIMEOUT); +} + +/* + * IOVA page translation tree management. + */ + +static void riscv_iommu_iotlb_flush_all(struct iommu_domain *iommu_domain) +{ + struct riscv_iommu_domain *domain =3D iommu_domain_to_riscv(iommu_domain); + + riscv_iommu_iotlb_inval(domain, 0, ULONG_MAX); +} + +static void riscv_iommu_iotlb_sync(struct iommu_domain *iommu_domain, + struct iommu_iotlb_gather *gather) +{ + struct riscv_iommu_domain *domain =3D iommu_domain_to_riscv(iommu_domain); + + riscv_iommu_iotlb_inval(domain, gather->start, gather->end); +} + +#define PT_SHIFT (PAGE_SHIFT - ilog2(sizeof(pte_t))) + +#define _io_pte_present(pte) ((pte) & (_PAGE_PRESENT | _PAGE_PROT_NONE)) +#define _io_pte_leaf(pte) ((pte) & _PAGE_LEAF) +#define _io_pte_none(pte) ((pte) =3D=3D 0) +#define _io_pte_entry(pn, prot) ((_PAGE_PFN_MASK & ((pn) << _PAGE_PFN_SHIF= T)) | (prot)) + +static void riscv_iommu_pte_free(struct riscv_iommu_domain *domain, + unsigned long pte, struct list_head *freelist) +{ + unsigned long *ptr; + int i; + + if (!_io_pte_present(pte) || _io_pte_leaf(pte)) + return; + + ptr =3D (unsigned long *)pfn_to_virt(__page_val_to_pfn(pte)); + + /* Recursively free all sub page table pages */ + for (i =3D 0; i < PTRS_PER_PTE; i++) { + pte =3D READ_ONCE(ptr[i]); + if (!_io_pte_none(pte) && cmpxchg_relaxed(ptr + i, pte, 0) =3D=3D pte) + riscv_iommu_pte_free(domain, pte, freelist); + } + + if (freelist) + list_add_tail(&virt_to_page(ptr)->lru, freelist); + else + iommu_free_page(ptr); +} + +static unsigned long *riscv_iommu_pte_alloc(struct riscv_iommu_domain *dom= ain, + unsigned long iova, size_t pgsize, + gfp_t gfp) +{ + unsigned long *ptr =3D domain->pgd_root; + unsigned long pte, old; + int level =3D domain->pgd_mode - RISCV_IOMMU_DC_FSC_IOSATP_MODE_SV39 + 2; + void *addr; + + do { + const int shift =3D PAGE_SHIFT + PT_SHIFT * level; + + ptr +=3D ((iova >> shift) & (PTRS_PER_PTE - 1)); + /* + * Note: returned entry might be a non-leaf if there was + * existing mapping with smaller granularity. Up to the caller + * to replace and invalidate. + */ + if (((size_t)1 << shift) =3D=3D pgsize) + return ptr; +pte_retry: + pte =3D READ_ONCE(*ptr); + /* + * This is very likely incorrect as we should not be adding + * new mapping with smaller granularity on top + * of existing 2M/1G mapping. Fail. + */ + if (_io_pte_present(pte) && _io_pte_leaf(pte)) + return NULL; + /* + * Non-leaf entry is missing, allocate and try to add to the + * page table. This might race with other mappings, retry. + */ + if (_io_pte_none(pte)) { + addr =3D iommu_alloc_page_node(domain->numa_node, gfp); + if (!addr) + return NULL; + old =3D pte; + pte =3D _io_pte_entry(virt_to_pfn(addr), _PAGE_TABLE); + if (cmpxchg_relaxed(ptr, old, pte) !=3D old) { + iommu_free_page(addr); + goto pte_retry; + } + } + ptr =3D (unsigned long *)pfn_to_virt(__page_val_to_pfn(pte)); + } while (level-- > 0); + + return NULL; +} + +static unsigned long *riscv_iommu_pte_fetch(struct riscv_iommu_domain *dom= ain, + unsigned long iova, size_t *pte_pgsize) +{ + unsigned long *ptr =3D domain->pgd_root; + unsigned long pte; + int level =3D domain->pgd_mode - RISCV_IOMMU_DC_FSC_IOSATP_MODE_SV39 + 2; + + do { + const int shift =3D PAGE_SHIFT + PT_SHIFT * level; + + ptr +=3D ((iova >> shift) & (PTRS_PER_PTE - 1)); + pte =3D READ_ONCE(*ptr); + if (_io_pte_present(pte) && _io_pte_leaf(pte)) { + *pte_pgsize =3D (size_t)1 << shift; + return ptr; + } + if (_io_pte_none(pte)) + return NULL; + ptr =3D (unsigned long *)pfn_to_virt(__page_val_to_pfn(pte)); + } while (level-- > 0); + + return NULL; +} + +static int riscv_iommu_map_pages(struct iommu_domain *iommu_domain, + unsigned long iova, phys_addr_t phys, + size_t pgsize, size_t pgcount, int prot, + gfp_t gfp, size_t *mapped) +{ + struct riscv_iommu_domain *domain =3D iommu_domain_to_riscv(iommu_domain); + size_t size =3D 0; + unsigned long *ptr; + unsigned long pte, old, pte_prot; + int rc =3D 0; + LIST_HEAD(freelist); + + if (!(prot & IOMMU_WRITE)) + pte_prot =3D _PAGE_BASE | _PAGE_READ; + else if (domain->amo_enabled) + pte_prot =3D _PAGE_BASE | _PAGE_READ | _PAGE_WRITE; + else + pte_prot =3D _PAGE_BASE | _PAGE_READ | _PAGE_WRITE | _PAGE_DIRTY; + + while (pgcount) { + ptr =3D riscv_iommu_pte_alloc(domain, iova, pgsize, gfp); + if (!ptr) { + rc =3D -ENOMEM; + break; + } + + old =3D READ_ONCE(*ptr); + pte =3D _io_pte_entry(phys_to_pfn(phys), pte_prot); + if (cmpxchg_relaxed(ptr, old, pte) !=3D old) + continue; + + riscv_iommu_pte_free(domain, old, &freelist); + + size +=3D pgsize; + iova +=3D pgsize; + phys +=3D pgsize; + --pgcount; + } + + *mapped =3D size; + + if (!list_empty(&freelist)) { + /* + * In 1.0 spec version, the smallest scope we can use to + * invalidate all levels of page table (i.e. leaf and non-leaf) + * is an invalidate-all-PSCID IOTINVAL.VMA with AV=3D0. + * This will be updated with hardware support for + * capability.NL (non-leaf) IOTINVAL command. + */ + riscv_iommu_iotlb_inval(domain, 0, ULONG_MAX); + iommu_put_pages_list(&freelist); + } + + return rc; +} + +static size_t riscv_iommu_unmap_pages(struct iommu_domain *iommu_domain, + unsigned long iova, size_t pgsize, + size_t pgcount, + struct iommu_iotlb_gather *gather) +{ + struct riscv_iommu_domain *domain =3D iommu_domain_to_riscv(iommu_domain); + size_t size =3D pgcount << __ffs(pgsize); + unsigned long *ptr, old; + size_t unmapped =3D 0; + size_t pte_size; + + while (unmapped < size) { + ptr =3D riscv_iommu_pte_fetch(domain, iova, &pte_size); + if (!ptr) + return unmapped; + + /* partial unmap is not allowed, fail. */ + if (iova & (pte_size - 1)) + return unmapped; + + old =3D READ_ONCE(*ptr); + if (cmpxchg_relaxed(ptr, old, 0) !=3D old) + continue; + + iommu_iotlb_gather_add_page(&domain->domain, gather, iova, + pte_size); + + iova +=3D pte_size; + unmapped +=3D pte_size; + } + + return unmapped; +} + +static phys_addr_t riscv_iommu_iova_to_phys(struct iommu_domain *iommu_dom= ain, + dma_addr_t iova) +{ + struct riscv_iommu_domain *domain =3D iommu_domain_to_riscv(iommu_domain); + unsigned long pte_size; + unsigned long *ptr; + + ptr =3D riscv_iommu_pte_fetch(domain, iova, &pte_size); + if (_io_pte_none(*ptr) || !_io_pte_present(*ptr)) + return 0; + + return pfn_to_phys(__page_val_to_pfn(*ptr)) | (iova & (pte_size - 1)); +} + +static void riscv_iommu_free_paging_domain(struct iommu_domain *iommu_doma= in) +{ + struct riscv_iommu_domain *domain =3D iommu_domain_to_riscv(iommu_domain); + const unsigned long pfn =3D virt_to_pfn(domain->pgd_root); + + WARN_ON(!list_empty(&domain->bonds)); + + if ((int)domain->pscid > 0) + ida_free(&riscv_iommu_pscids, domain->pscid); + + riscv_iommu_pte_free(domain, _io_pte_entry(pfn, _PAGE_TABLE), NULL); + kfree(domain); +} + +static bool riscv_iommu_pt_supported(struct riscv_iommu_device *iommu, int= pgd_mode) +{ + switch (pgd_mode) { + case RISCV_IOMMU_DC_FSC_IOSATP_MODE_SV39: + return iommu->caps & RISCV_IOMMU_CAPABILITIES_SV39; + + case RISCV_IOMMU_DC_FSC_IOSATP_MODE_SV48: + return iommu->caps & RISCV_IOMMU_CAPABILITIES_SV48; + + case RISCV_IOMMU_DC_FSC_IOSATP_MODE_SV57: + return iommu->caps & RISCV_IOMMU_CAPABILITIES_SV57; + } + return false; +} + +static int riscv_iommu_attach_paging_domain(struct iommu_domain *iommu_dom= ain, + struct device *dev) +{ + struct riscv_iommu_domain *domain =3D iommu_domain_to_riscv(iommu_domain); + struct riscv_iommu_device *iommu =3D dev_to_iommu(dev); + struct riscv_iommu_info *info =3D dev_iommu_priv_get(dev); + u64 fsc, ta; + + if (!riscv_iommu_pt_supported(iommu, domain->pgd_mode)) + return -ENODEV; + + fsc =3D FIELD_PREP(RISCV_IOMMU_PC_FSC_MODE, domain->pgd_mode) | + FIELD_PREP(RISCV_IOMMU_PC_FSC_PPN, virt_to_pfn(domain->pgd_root)); + ta =3D FIELD_PREP(RISCV_IOMMU_PC_TA_PSCID, domain->pscid) | + RISCV_IOMMU_PC_TA_V; + + if (riscv_iommu_bond_link(domain, dev)) + return -ENOMEM; + + riscv_iommu_iodir_update(iommu, dev, fsc, ta); + riscv_iommu_bond_unlink(info->domain, dev); + info->domain =3D domain; + + return 0; +} + +static const struct iommu_domain_ops riscv_iommu_paging_domain_ops =3D { + .attach_dev =3D riscv_iommu_attach_paging_domain, + .free =3D riscv_iommu_free_paging_domain, + .map_pages =3D riscv_iommu_map_pages, + .unmap_pages =3D riscv_iommu_unmap_pages, + .iova_to_phys =3D riscv_iommu_iova_to_phys, + .iotlb_sync =3D riscv_iommu_iotlb_sync, + .flush_iotlb_all =3D riscv_iommu_iotlb_flush_all, +}; + +static struct iommu_domain *riscv_iommu_alloc_paging_domain(struct device = *dev) +{ + struct riscv_iommu_domain *domain; + struct riscv_iommu_device *iommu; + unsigned int pgd_mode; + dma_addr_t va_mask; + int va_bits; + + iommu =3D dev_to_iommu(dev); + if (iommu->caps & RISCV_IOMMU_CAPABILITIES_SV57) { + pgd_mode =3D RISCV_IOMMU_DC_FSC_IOSATP_MODE_SV57; + va_bits =3D 57; + } else if (iommu->caps & RISCV_IOMMU_CAPABILITIES_SV48) { + pgd_mode =3D RISCV_IOMMU_DC_FSC_IOSATP_MODE_SV48; + va_bits =3D 48; + } else if (iommu->caps & RISCV_IOMMU_CAPABILITIES_SV39) { + pgd_mode =3D RISCV_IOMMU_DC_FSC_IOSATP_MODE_SV39; + va_bits =3D 39; + } else { + dev_err(dev, "cannot find supported page table mode\n"); + return ERR_PTR(-ENODEV); + } + + domain =3D kzalloc(sizeof(*domain), GFP_KERNEL); + if (!domain) + return ERR_PTR(-ENOMEM); + + INIT_LIST_HEAD_RCU(&domain->bonds); + spin_lock_init(&domain->lock); + domain->numa_node =3D dev_to_node(iommu->dev); + domain->amo_enabled =3D !!(iommu->caps & RISCV_IOMMU_CAPABILITIES_AMO_HWA= D); + domain->pgd_mode =3D pgd_mode; + domain->pgd_root =3D iommu_alloc_page_node(domain->numa_node, + GFP_KERNEL_ACCOUNT); + if (!domain->pgd_root) { + kfree(domain); + return ERR_PTR(-ENOMEM); + } + + domain->pscid =3D ida_alloc_range(&riscv_iommu_pscids, 1, + RISCV_IOMMU_MAX_PSCID, GFP_KERNEL); + if (domain->pscid < 0) { + iommu_free_page(domain->pgd_root); + kfree(domain); + return ERR_PTR(-ENOMEM); + } + + /* + * Note: RISC-V Privilege spec mandates that virtual addresses + * need to be sign-extended, so if (VA_BITS - 1) is set, all + * bits >=3D VA_BITS need to also be set or else we'll get a + * page fault. However the code that creates the mappings + * above us (e.g. iommu_dma_alloc_iova()) won't do that for us + * for now, so we'll end up with invalid virtual addresses + * to map. As a workaround until we get this sorted out + * limit the available virtual addresses to VA_BITS - 1. + */ + va_mask =3D DMA_BIT_MASK(va_bits - 1); + + domain->domain.geometry.aperture_start =3D 0; + domain->domain.geometry.aperture_end =3D va_mask; + domain->domain.geometry.force_aperture =3D true; + domain->domain.pgsize_bitmap =3D va_mask & (SZ_4K | SZ_2M | SZ_1G | SZ_51= 2G); + + domain->domain.ops =3D &riscv_iommu_paging_domain_ops; + + return &domain->domain; } =20 static int riscv_iommu_attach_blocking_domain(struct iommu_domain *iommu_d= omain, struct device *dev) { struct riscv_iommu_device *iommu =3D dev_to_iommu(dev); + struct riscv_iommu_info *info =3D dev_iommu_priv_get(dev); =20 + /* Make device context invalid, translation requests will fault w/ #258 */ riscv_iommu_iodir_update(iommu, dev, RISCV_IOMMU_FSC_BARE, 0); + riscv_iommu_bond_unlink(info->domain, dev); + info->domain =3D NULL; =20 return 0; } @@ -853,8 +1439,11 @@ static int riscv_iommu_attach_identity_domain(struct = iommu_domain *iommu_domain, struct device *dev) { struct riscv_iommu_device *iommu =3D dev_to_iommu(dev); + struct riscv_iommu_info *info =3D dev_iommu_priv_get(dev); =20 riscv_iommu_iodir_update(iommu, dev, RISCV_IOMMU_FSC_BARE, RISCV_IOMMU_PC= _TA_V); + riscv_iommu_bond_unlink(info->domain, dev); + info->domain =3D NULL; =20 return 0; } @@ -866,11 +1455,6 @@ static struct iommu_domain riscv_iommu_identity_domai= n =3D { } }; =20 -static int riscv_iommu_device_domain_type(struct device *dev) -{ - return IOMMU_DOMAIN_IDENTITY; -} - static struct iommu_group *riscv_iommu_device_group(struct device *dev) { if (dev_is_pci(dev)) @@ -887,6 +1471,7 @@ static struct iommu_device *riscv_iommu_probe_device(s= truct device *dev) { struct iommu_fwspec *fwspec =3D dev_iommu_fwspec_get(dev); struct riscv_iommu_device *iommu; + struct riscv_iommu_info *info; struct riscv_iommu_dc *dc; u64 tc; int i; @@ -905,6 +1490,9 @@ static struct iommu_device *riscv_iommu_probe_device(s= truct device *dev) if (iommu->ddt_mode <=3D RISCV_IOMMU_DDTP_IOMMU_MODE_BARE) return ERR_PTR(-ENODEV); =20 + info =3D kzalloc(sizeof(*info), GFP_KERNEL); + if (!info) + return ERR_PTR(-ENOMEM); /* * Allocate and pre-configure device context entries in * the device directory. Do not mark the context valid yet. @@ -914,24 +1502,37 @@ static struct iommu_device *riscv_iommu_probe_device= (struct device *dev) tc |=3D RISCV_IOMMU_DC_TC_SADE; for (i =3D 0; i < fwspec->num_ids; i++) { dc =3D riscv_iommu_get_dc(iommu, fwspec->ids[i]); - if (!dc) + if (!dc) { + kfree(info); return ERR_PTR(-ENODEV); + } if (READ_ONCE(dc->tc) & RISCV_IOMMU_DC_TC_V) dev_warn(dev, "already attached to IOMMU device directory\n"); WRITE_ONCE(dc->tc, tc); } =20 + dev_iommu_priv_set(dev, info); + return &iommu->iommu; } =20 +static void riscv_iommu_release_device(struct device *dev) +{ + struct riscv_iommu_info *info =3D dev_iommu_priv_get(dev); + + kfree_rcu_mightsleep(info); +} + static const struct iommu_ops riscv_iommu_ops =3D { + .pgsize_bitmap =3D SZ_4K, .of_xlate =3D riscv_iommu_of_xlate, .identity_domain =3D &riscv_iommu_identity_domain, .blocked_domain =3D &riscv_iommu_blocking_domain, .release_domain =3D &riscv_iommu_blocking_domain, - .def_domain_type =3D riscv_iommu_device_domain_type, + .domain_alloc_paging =3D riscv_iommu_alloc_paging_domain, .device_group =3D riscv_iommu_device_group, .probe_device =3D riscv_iommu_probe_device, + .release_device =3D riscv_iommu_release_device, }; =20 static int riscv_iommu_init_check(struct riscv_iommu_device *iommu) --=20 2.34.1