diff options
Diffstat (limited to 'drivers/iommu')
| -rw-r--r-- | drivers/iommu/amd/amd_iommu.h | 2 | ||||
| -rw-r--r-- | drivers/iommu/amd/amd_iommu_types.h | 1 | ||||
| -rw-r--r-- | drivers/iommu/amd/init.c | 13 | ||||
| -rw-r--r-- | drivers/iommu/amd/iommu.c | 59 | ||||
| -rw-r--r-- | drivers/iommu/apple-dart.c | 3 | ||||
| -rw-r--r-- | drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c | 6 | ||||
| -rw-r--r-- | drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 25 | ||||
| -rw-r--r-- | drivers/iommu/arm/arm-smmu/arm-smmu-qcom-debug.c | 9 | ||||
| -rw-r--r-- | drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c | 44 | ||||
| -rw-r--r-- | drivers/iommu/arm/arm-smmu/arm-smmu.c | 6 | ||||
| -rw-r--r-- | drivers/iommu/fsl_pamu_domain.c | 2 | ||||
| -rw-r--r-- | drivers/iommu/intel/dmar.c | 4 | ||||
| -rw-r--r-- | drivers/iommu/intel/iommu.c | 159 | ||||
| -rw-r--r-- | drivers/iommu/intel/iommu.h | 22 | ||||
| -rw-r--r-- | drivers/iommu/intel/nested.c | 4 | ||||
| -rw-r--r-- | drivers/iommu/io-pgtable-arm.c | 27 | ||||
| -rw-r--r-- | drivers/iommu/iommu.c | 43 | ||||
| -rw-r--r-- | drivers/iommu/ipmmu-vmsa.c | 3 | ||||
| -rw-r--r-- | drivers/iommu/mtk_iommu.c | 37 | ||||
| -rw-r--r-- | drivers/iommu/s390-iommu.c | 345 |
20 files changed, 599 insertions, 215 deletions
diff --git a/drivers/iommu/amd/amd_iommu.h b/drivers/iommu/amd/amd_iommu.h index 220c598b7e14..29a8864381c3 100644 --- a/drivers/iommu/amd/amd_iommu.h +++ b/drivers/iommu/amd/amd_iommu.h @@ -147,6 +147,8 @@ static inline int get_pci_sbdf_id(struct pci_dev *pdev) return PCI_SEG_DEVID_TO_SBDF(seg, devid); } +bool amd_iommu_ht_range_ignore(void); + /* * This must be called after device probe completes. During probe * use rlookup_amd_iommu() get the iommu. diff --git a/drivers/iommu/amd/amd_iommu_types.h b/drivers/iommu/amd/amd_iommu_types.h index bc9f0bbf3c08..ccbab3a4811a 100644 --- a/drivers/iommu/amd/amd_iommu_types.h +++ b/drivers/iommu/amd/amd_iommu_types.h @@ -109,6 +109,7 @@ #define FEATURE_SNPAVICSUP GENMASK_ULL(7, 5) #define FEATURE_SNPAVICSUP_GAM(x) \ (FIELD_GET(FEATURE_SNPAVICSUP, x) == 0x1) +#define FEATURE_HT_RANGE_IGNORE BIT_ULL(11) #define FEATURE_NUM_INT_REMAP_SUP GENMASK_ULL(9, 8) #define FEATURE_NUM_INT_REMAP_SUP_2K(x) \ diff --git a/drivers/iommu/amd/init.c b/drivers/iommu/amd/init.c index df1094893eb0..c06b62f87b9b 100644 --- a/drivers/iommu/amd/init.c +++ b/drivers/iommu/amd/init.c @@ -248,6 +248,11 @@ int amd_iommu_get_num_iommus(void) return amd_iommus_present; } +bool amd_iommu_ht_range_ignore(void) +{ + return check_feature2(FEATURE_HT_RANGE_IGNORE); +} + /* * Iterate through all the IOMMUs to get common EFR * masks among all IOMMUs and warn if found inconsistency. @@ -3651,6 +3656,14 @@ found: while (*uid == '0' && *(uid + 1)) uid++; + if (strlen(hid) >= ACPIHID_HID_LEN) { + pr_err("Invalid command line: hid is too long\n"); + return 1; + } else if (strlen(uid) >= ACPIHID_UID_LEN) { + pr_err("Invalid command line: uid is too long\n"); + return 1; + } + i = early_acpihid_map_size++; memcpy(early_acpihid_map[i].hid, hid, strlen(hid)); memcpy(early_acpihid_map[i].uid, uid, strlen(uid)); diff --git a/drivers/iommu/amd/iommu.c b/drivers/iommu/amd/iommu.c index a9d539fce982..3117d99cf83d 100644 --- a/drivers/iommu/amd/iommu.c +++ b/drivers/iommu/amd/iommu.c @@ -241,7 +241,9 @@ static inline int get_acpihid_device_id(struct device *dev, struct acpihid_map_entry **entry) { struct acpi_device *adev = ACPI_COMPANION(dev); - struct acpihid_map_entry *p; + struct acpihid_map_entry *p, *p1 = NULL; + int hid_count = 0; + bool fw_bug; if (!adev) return -ENODEV; @@ -249,12 +251,33 @@ static inline int get_acpihid_device_id(struct device *dev, list_for_each_entry(p, &acpihid_map, list) { if (acpi_dev_hid_uid_match(adev, p->hid, p->uid[0] ? p->uid : NULL)) { - if (entry) - *entry = p; - return p->devid; + p1 = p; + fw_bug = false; + hid_count = 1; + break; + } + + /* + * Count HID matches w/o UID, raise FW_BUG but allow exactly one match + */ + if (acpi_dev_hid_match(adev, p->hid)) { + p1 = p; + hid_count++; + fw_bug = true; } } - return -EINVAL; + + if (!p1) + return -EINVAL; + if (fw_bug) + dev_err_once(dev, FW_BUG "No ACPI device matched UID, but %d device%s matched HID.\n", + hid_count, hid_count > 1 ? "s" : ""); + if (hid_count > 1) + return -EINVAL; + if (entry) + *entry = p1; + + return p1->devid; } static inline int get_device_sbdf_id(struct device *dev) @@ -982,6 +1005,14 @@ int amd_iommu_register_ga_log_notifier(int (*notifier)(u32)) { iommu_ga_log_notifier = notifier; + /* + * Ensure all in-flight IRQ handlers run to completion before returning + * to the caller, e.g. to ensure module code isn't unloaded while it's + * being executed in the IRQ handler. + */ + if (!notifier) + synchronize_rcu(); + return 0; } EXPORT_SYMBOL(amd_iommu_register_ga_log_notifier); @@ -2908,6 +2939,9 @@ static void amd_iommu_get_resv_regions(struct device *dev, return; list_add_tail(®ion->list, head); + if (amd_iommu_ht_range_ignore()) + return; + region = iommu_alloc_resv_region(HT_RANGE_START, HT_RANGE_END - HT_RANGE_START + 1, 0, IOMMU_RESV_RESERVED, GFP_KERNEL); @@ -3835,6 +3869,9 @@ static int amd_ir_set_vcpu_affinity(struct irq_data *data, void *vcpu_info) struct irq_2_irte *irte_info = &ir_data->irq_2_irte; struct iommu_dev_data *dev_data; + if (WARN_ON_ONCE(!AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir))) + return -EINVAL; + if (ir_data->iommu == NULL) return -EINVAL; @@ -3845,21 +3882,11 @@ static int amd_ir_set_vcpu_affinity(struct irq_data *data, void *vcpu_info) * we should not modify the IRTE */ if (!dev_data || !dev_data->use_vapic) - return 0; + return -EINVAL; ir_data->cfg = irqd_cfg(data); pi_data->ir_data = ir_data; - /* Note: - * SVM tries to set up for VAPIC mode, but we are in - * legacy mode. So, we force legacy mode instead. - */ - if (!AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir)) { - pr_debug("%s: Fall back to using intr legacy remap\n", - __func__); - pi_data->is_guest_mode = false; - } - pi_data->prev_ga_tag = ir_data->cached_ga_tag; if (pi_data->is_guest_mode) { ir_data->ga_root_ptr = (pi_data->base >> 12); diff --git a/drivers/iommu/apple-dart.c b/drivers/iommu/apple-dart.c index e13501541fdd..757d24f67ad4 100644 --- a/drivers/iommu/apple-dart.c +++ b/drivers/iommu/apple-dart.c @@ -776,8 +776,7 @@ static void apple_dart_domain_free(struct iommu_domain *domain) { struct apple_dart_domain *dart_domain = to_dart_domain(domain); - if (dart_domain->pgtbl_ops) - free_io_pgtable_ops(dart_domain->pgtbl_ops); + free_io_pgtable_ops(dart_domain->pgtbl_ops); kfree(dart_domain); } diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c index 605d1dd0e1cc..0601dece0a0d 100644 --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c @@ -337,6 +337,12 @@ struct iommu_domain *arm_smmu_sva_domain_alloc(struct device *dev, return ERR_CAST(smmu_domain); smmu_domain->domain.type = IOMMU_DOMAIN_SVA; smmu_domain->domain.ops = &arm_smmu_sva_domain_ops; + + /* + * Choose page_size as the leaf page size for invalidation when + * ARM_SMMU_FEAT_RANGE_INV is present + */ + smmu_domain->domain.pgsize_bitmap = PAGE_SIZE; smmu_domain->smmu = smmu; ret = xa_alloc(&arm_smmu_asid_xa, &asid, smmu_domain, diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c index 09d51efd9095..10cc6dc26b7b 100644 --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c @@ -3022,7 +3022,7 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev) smmu = master->smmu; if (smmu_domain->smmu != smmu) - return ret; + return -EINVAL; if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) { cdptr = arm_smmu_alloc_cd_ptr(master, IOMMU_NO_PASID); @@ -3457,6 +3457,7 @@ static int arm_smmu_insert_master(struct arm_smmu_device *smmu, mutex_lock(&smmu->streams_mutex); for (i = 0; i < fwspec->num_ids; i++) { struct arm_smmu_stream *new_stream = &master->streams[i]; + struct rb_node *existing; u32 sid = fwspec->ids[i]; new_stream->id = sid; @@ -3467,11 +3468,21 @@ static int arm_smmu_insert_master(struct arm_smmu_device *smmu, break; /* Insert into SID tree */ - if (rb_find_add(&new_stream->node, &smmu->streams, - arm_smmu_streams_cmp_node)) { - dev_warn(master->dev, "stream %u already in tree\n", - sid); - ret = -EINVAL; + existing = rb_find_add(&new_stream->node, &smmu->streams, + arm_smmu_streams_cmp_node); + if (existing) { + struct arm_smmu_master *existing_master = + rb_entry(existing, struct arm_smmu_stream, node) + ->master; + + /* Bridged PCI devices may end up with duplicated IDs */ + if (existing_master == master) + continue; + + dev_warn(master->dev, + "Aliasing StreamID 0x%x (from %s) unsupported, expect DMA to be broken\n", + sid, dev_name(existing_master->dev)); + ret = -ENODEV; break; } } @@ -4443,6 +4454,8 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu) reg = readl_relaxed(smmu->base + ARM_SMMU_IDR3); if (FIELD_GET(IDR3_RIL, reg)) smmu->features |= ARM_SMMU_FEAT_RANGE_INV; + if (FIELD_GET(IDR3_FWB, reg)) + smmu->features |= ARM_SMMU_FEAT_S2FWB; /* IDR5 */ reg = readl_relaxed(smmu->base + ARM_SMMU_IDR5); diff --git a/drivers/iommu/arm/arm-smmu/arm-smmu-qcom-debug.c b/drivers/iommu/arm/arm-smmu/arm-smmu-qcom-debug.c index d03b2239baad..65e0ef6539fe 100644 --- a/drivers/iommu/arm/arm-smmu/arm-smmu-qcom-debug.c +++ b/drivers/iommu/arm/arm-smmu/arm-smmu-qcom-debug.c @@ -406,6 +406,12 @@ irqreturn_t qcom_smmu_context_fault(int irq, void *dev) arm_smmu_print_context_fault_info(smmu, idx, &cfi); arm_smmu_cb_write(smmu, idx, ARM_SMMU_CB_FSR, cfi.fsr); + + if (cfi.fsr & ARM_SMMU_CB_FSR_SS) { + arm_smmu_cb_write(smmu, idx, ARM_SMMU_CB_RESUME, + ret == -EAGAIN ? 0 : ARM_SMMU_RESUME_TERMINATE); + } + return IRQ_HANDLED; } @@ -416,6 +422,9 @@ irqreturn_t qcom_smmu_context_fault(int irq, void *dev) if (!tmp || tmp == -EBUSY) { ret = IRQ_HANDLED; resume = ARM_SMMU_RESUME_TERMINATE; + } else if (tmp == -EAGAIN) { + ret = IRQ_HANDLED; + resume = 0; } else { phys_addr_t phys_atos = qcom_smmu_verify_fault(smmu_domain, cfi.iova, cfi.fsr); diff --git a/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c b/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c index 59d02687280e..62874b18f645 100644 --- a/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c +++ b/drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c @@ -112,25 +112,39 @@ static void qcom_adreno_smmu_set_stall(const void *cookie, bool enabled) { struct arm_smmu_domain *smmu_domain = (void *)cookie; struct arm_smmu_cfg *cfg = &smmu_domain->cfg; - struct qcom_smmu *qsmmu = to_qcom_smmu(smmu_domain->smmu); + struct arm_smmu_device *smmu = smmu_domain->smmu; + struct qcom_smmu *qsmmu = to_qcom_smmu(smmu); + u32 mask = BIT(cfg->cbndx); + bool stall_changed = !!(qsmmu->stall_enabled & mask) != enabled; + unsigned long flags; if (enabled) - qsmmu->stall_enabled |= BIT(cfg->cbndx); + qsmmu->stall_enabled |= mask; else - qsmmu->stall_enabled &= ~BIT(cfg->cbndx); -} + qsmmu->stall_enabled &= ~mask; -static void qcom_adreno_smmu_resume_translation(const void *cookie, bool terminate) -{ - struct arm_smmu_domain *smmu_domain = (void *)cookie; - struct arm_smmu_cfg *cfg = &smmu_domain->cfg; - struct arm_smmu_device *smmu = smmu_domain->smmu; - u32 reg = 0; + /* + * If the device is on and we changed the setting, update the register. + * The spec pseudocode says that CFCFG is resampled after a fault, and + * we believe that no implementations cache it in the TLB, so it should + * be safe to change it without a TLB invalidation. + */ + if (stall_changed && pm_runtime_get_if_active(smmu->dev) > 0) { + u32 reg; + + spin_lock_irqsave(&smmu_domain->cb_lock, flags); + reg = arm_smmu_cb_read(smmu, cfg->cbndx, ARM_SMMU_CB_SCTLR); + + if (enabled) + reg |= ARM_SMMU_SCTLR_CFCFG; + else + reg &= ~ARM_SMMU_SCTLR_CFCFG; - if (terminate) - reg |= ARM_SMMU_RESUME_TERMINATE; + arm_smmu_cb_write(smmu, cfg->cbndx, ARM_SMMU_CB_SCTLR, reg); + spin_unlock_irqrestore(&smmu_domain->cb_lock, flags); - arm_smmu_cb_write(smmu, cfg->cbndx, ARM_SMMU_CB_RESUME, reg); + pm_runtime_put_autosuspend(smmu->dev); + } } static void qcom_adreno_smmu_set_prr_bit(const void *cookie, bool set) @@ -337,7 +351,6 @@ static int qcom_adreno_smmu_init_context(struct arm_smmu_domain *smmu_domain, priv->set_ttbr0_cfg = qcom_adreno_smmu_set_ttbr0_cfg; priv->get_fault_info = qcom_adreno_smmu_get_fault_info; priv->set_stall = qcom_adreno_smmu_set_stall; - priv->resume_translation = qcom_adreno_smmu_resume_translation; priv->set_prr_bit = NULL; priv->set_prr_addr = NULL; @@ -356,6 +369,7 @@ static const struct of_device_id qcom_smmu_client_of_match[] __maybe_unused = { { .compatible = "qcom,mdp4" }, { .compatible = "qcom,mdss" }, { .compatible = "qcom,qcm2290-mdss" }, + { .compatible = "qcom,sar2130p-mdss" }, { .compatible = "qcom,sc7180-mdss" }, { .compatible = "qcom,sc7180-mss-pil" }, { .compatible = "qcom,sc7280-mdss" }, @@ -585,6 +599,7 @@ static const struct arm_smmu_impl qcom_adreno_smmu_v2_impl = { .alloc_context_bank = qcom_adreno_smmu_alloc_context_bank, .write_sctlr = qcom_adreno_smmu_write_sctlr, .tlb_sync = qcom_smmu_tlb_sync, + .context_fault_needs_threaded_irq = true, }; static const struct arm_smmu_impl qcom_adreno_smmu_500_impl = { @@ -594,6 +609,7 @@ static const struct arm_smmu_impl qcom_adreno_smmu_500_impl = { .alloc_context_bank = qcom_adreno_smmu_alloc_context_bank, .write_sctlr = qcom_adreno_smmu_write_sctlr, .tlb_sync = qcom_smmu_tlb_sync, + .context_fault_needs_threaded_irq = true, }; static struct arm_smmu_device *qcom_smmu_create(struct arm_smmu_device *smmu, diff --git a/drivers/iommu/arm/arm-smmu/arm-smmu.c b/drivers/iommu/arm/arm-smmu/arm-smmu.c index 8f439c265a23..8d95b14c7d5a 100644 --- a/drivers/iommu/arm/arm-smmu/arm-smmu.c +++ b/drivers/iommu/arm/arm-smmu/arm-smmu.c @@ -474,6 +474,12 @@ static irqreturn_t arm_smmu_context_fault(int irq, void *dev) arm_smmu_print_context_fault_info(smmu, idx, &cfi); arm_smmu_cb_write(smmu, idx, ARM_SMMU_CB_FSR, cfi.fsr); + + if (cfi.fsr & ARM_SMMU_CB_FSR_SS) { + arm_smmu_cb_write(smmu, idx, ARM_SMMU_CB_RESUME, + ret == -EAGAIN ? 0 : ARM_SMMU_RESUME_TERMINATE); + } + return IRQ_HANDLED; } diff --git a/drivers/iommu/fsl_pamu_domain.c b/drivers/iommu/fsl_pamu_domain.c index 30be786bff11..5f08523f97cb 100644 --- a/drivers/iommu/fsl_pamu_domain.c +++ b/drivers/iommu/fsl_pamu_domain.c @@ -64,7 +64,7 @@ static int update_liodn_stash(int liodn, struct fsl_dma_domain *dma_domain, spin_lock_irqsave(&iommu_lock, flags); ret = pamu_update_paace_stash(liodn, val); if (ret) { - pr_debug("Failed to update SPAACE for liodn %d\n ", liodn); + pr_debug("Failed to update SPAACE for liodn %d\n", liodn); spin_unlock_irqrestore(&iommu_lock, flags); return ret; } diff --git a/drivers/iommu/intel/dmar.c b/drivers/iommu/intel/dmar.c index 10374b684d6d..b61d9ea27aa9 100644 --- a/drivers/iommu/intel/dmar.c +++ b/drivers/iommu/intel/dmar.c @@ -1099,6 +1099,9 @@ static int alloc_iommu(struct dmar_drhd_unit *drhd) spin_lock_init(&iommu->device_rbtree_lock); mutex_init(&iommu->iopf_lock); iommu->node = NUMA_NO_NODE; + spin_lock_init(&iommu->lock); + ida_init(&iommu->domain_ida); + mutex_init(&iommu->did_lock); ver = readl(iommu->reg + DMAR_VER_REG); pr_info("%s: reg_base_addr %llx ver %d:%d cap %llx ecap %llx\n", @@ -1195,6 +1198,7 @@ static void free_iommu(struct intel_iommu *iommu) if (iommu->reg) unmap_iommu(iommu); + ida_destroy(&iommu->domain_ida); ida_free(&dmar_seq_ids, iommu->seq_id); kfree(iommu); } diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c index e9ffe0a95944..7aa3932251b2 100644 --- a/drivers/iommu/intel/iommu.c +++ b/drivers/iommu/intel/iommu.c @@ -1289,52 +1289,13 @@ static void iommu_disable_translation(struct intel_iommu *iommu) raw_spin_unlock_irqrestore(&iommu->register_lock, flag); } -static int iommu_init_domains(struct intel_iommu *iommu) -{ - u32 ndomains; - - ndomains = cap_ndoms(iommu->cap); - pr_debug("%s: Number of Domains supported <%d>\n", - iommu->name, ndomains); - - spin_lock_init(&iommu->lock); - - iommu->domain_ids = bitmap_zalloc(ndomains, GFP_KERNEL); - if (!iommu->domain_ids) - return -ENOMEM; - - /* - * If Caching mode is set, then invalid translations are tagged - * with domain-id 0, hence we need to pre-allocate it. We also - * use domain-id 0 as a marker for non-allocated domain-id, so - * make sure it is not used for a real domain. - */ - set_bit(0, iommu->domain_ids); - - /* - * Vt-d spec rev3.0 (section 6.2.3.1) requires that each pasid - * entry for first-level or pass-through translation modes should - * be programmed with a domain id different from those used for - * second-level or nested translation. We reserve a domain id for - * this purpose. This domain id is also used for identity domain - * in legacy mode. - */ - set_bit(FLPT_DEFAULT_DID, iommu->domain_ids); - - return 0; -} - static void disable_dmar_iommu(struct intel_iommu *iommu) { - if (!iommu->domain_ids) - return; - /* * All iommu domains must have been detached from the devices, * hence there should be no domain IDs in use. */ - if (WARN_ON(bitmap_weight(iommu->domain_ids, cap_ndoms(iommu->cap)) - > NUM_RESERVED_DID)) + if (WARN_ON(!ida_is_empty(&iommu->domain_ida))) return; if (iommu->gcmd & DMA_GCMD_TE) @@ -1343,11 +1304,6 @@ static void disable_dmar_iommu(struct intel_iommu *iommu) static void free_dmar_iommu(struct intel_iommu *iommu) { - if (iommu->domain_ids) { - bitmap_free(iommu->domain_ids); - iommu->domain_ids = NULL; - } - if (iommu->copied_tables) { bitmap_free(iommu->copied_tables); iommu->copied_tables = NULL; @@ -1380,7 +1336,6 @@ static bool first_level_by_default(struct intel_iommu *iommu) int domain_attach_iommu(struct dmar_domain *domain, struct intel_iommu *iommu) { struct iommu_domain_info *info, *curr; - unsigned long ndomains; int num, ret = -ENOSPC; if (domain->domain.type == IOMMU_DOMAIN_SVA) @@ -1390,40 +1345,36 @@ int domain_attach_iommu(struct dmar_domain *domain, struct intel_iommu *iommu) if (!info) return -ENOMEM; - spin_lock(&iommu->lock); + guard(mutex)(&iommu->did_lock); curr = xa_load(&domain->iommu_array, iommu->seq_id); if (curr) { curr->refcnt++; - spin_unlock(&iommu->lock); kfree(info); return 0; } - ndomains = cap_ndoms(iommu->cap); - num = find_first_zero_bit(iommu->domain_ids, ndomains); - if (num >= ndomains) { + num = ida_alloc_range(&iommu->domain_ida, IDA_START_DID, + cap_ndoms(iommu->cap) - 1, GFP_KERNEL); + if (num < 0) { pr_err("%s: No free domain ids\n", iommu->name); goto err_unlock; } - set_bit(num, iommu->domain_ids); info->refcnt = 1; info->did = num; info->iommu = iommu; curr = xa_cmpxchg(&domain->iommu_array, iommu->seq_id, - NULL, info, GFP_ATOMIC); + NULL, info, GFP_KERNEL); if (curr) { ret = xa_err(curr) ? : -EBUSY; goto err_clear; } - spin_unlock(&iommu->lock); return 0; err_clear: - clear_bit(info->did, iommu->domain_ids); + ida_free(&iommu->domain_ida, info->did); err_unlock: - spin_unlock(&iommu->lock); kfree(info); return ret; } @@ -1435,15 +1386,14 @@ void domain_detach_iommu(struct dmar_domain *domain, struct intel_iommu *iommu) if (domain->domain.type == IOMMU_DOMAIN_SVA) return; - spin_lock(&iommu->lock); + guard(mutex)(&iommu->did_lock); info = xa_load(&domain->iommu_array, iommu->seq_id); if (--info->refcnt == 0) { - clear_bit(info->did, iommu->domain_ids); + ida_free(&iommu->domain_ida, info->did); xa_erase(&domain->iommu_array, iommu->seq_id); domain->nid = NUMA_NO_NODE; kfree(info); } - spin_unlock(&iommu->lock); } static void domain_exit(struct dmar_domain *domain) @@ -1682,9 +1632,8 @@ __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn, } attr = prot & (DMA_PTE_READ | DMA_PTE_WRITE | DMA_PTE_SNP); - attr |= DMA_FL_PTE_PRESENT; if (domain->use_first_level) { - attr |= DMA_FL_PTE_US | DMA_FL_PTE_ACCESS; + attr |= DMA_FL_PTE_PRESENT | DMA_FL_PTE_US | DMA_FL_PTE_ACCESS; if (prot & DMA_PTE_WRITE) attr |= DMA_FL_PTE_DIRTY; } @@ -1860,6 +1809,7 @@ static int dmar_domain_attach_device(struct dmar_domain *domain, return ret; info->domain = domain; + info->domain_attached = true; spin_lock_irqsave(&domain->lock, flags); list_add(&info->link, &domain->devices); spin_unlock_irqrestore(&domain->lock, flags); @@ -2044,7 +1994,7 @@ static int copy_context_table(struct intel_iommu *iommu, did = context_domain_id(&ce); if (did >= 0 && did < cap_ndoms(iommu->cap)) - set_bit(did, iommu->domain_ids); + ida_alloc_range(&iommu->domain_ida, did, did, GFP_KERNEL); set_context_copied(iommu, bus, devfn); new_ce[idx] = ce; @@ -2171,11 +2121,6 @@ static int __init init_dmars(void) } intel_iommu_init_qi(iommu); - - ret = iommu_init_domains(iommu); - if (ret) - goto free_iommu; - init_translation_status(iommu); if (translation_pre_enabled(iommu) && !is_kdump_kernel()) { @@ -2653,9 +2598,7 @@ static int intel_iommu_add(struct dmar_drhd_unit *dmaru) if (iommu->gcmd & DMA_GCMD_TE) iommu_disable_translation(iommu); - ret = iommu_init_domains(iommu); - if (ret == 0) - ret = iommu_alloc_root_entry(iommu); + ret = iommu_alloc_root_entry(iommu); if (ret) goto out; @@ -2746,7 +2689,6 @@ static struct dmar_satc_unit *dmar_find_matched_satc_unit(struct pci_dev *dev) struct device *tmp; int i; - dev = pci_physfn(dev); rcu_read_lock(); list_for_each_entry_rcu(satcu, &dmar_satc_units, list) { @@ -2763,15 +2705,16 @@ out: return satcu; } -static int dmar_ats_supported(struct pci_dev *dev, struct intel_iommu *iommu) +static bool dmar_ats_supported(struct pci_dev *dev, struct intel_iommu *iommu) { - int i, ret = 1; - struct pci_bus *bus; struct pci_dev *bridge = NULL; - struct device *tmp; - struct acpi_dmar_atsr *atsr; struct dmar_atsr_unit *atsru; struct dmar_satc_unit *satcu; + struct acpi_dmar_atsr *atsr; + bool supported = true; + struct pci_bus *bus; + struct device *tmp; + int i; dev = pci_physfn(dev); satcu = dmar_find_matched_satc_unit(dev); @@ -2789,11 +2732,11 @@ static int dmar_ats_supported(struct pci_dev *dev, struct intel_iommu *iommu) bridge = bus->self; /* If it's an integrated device, allow ATS */ if (!bridge) - return 1; + return true; /* Connected via non-PCIe: no ATS */ if (!pci_is_pcie(bridge) || pci_pcie_type(bridge) == PCI_EXP_TYPE_PCI_BRIDGE) - return 0; + return false; /* If we found the root port, look it up in the ATSR */ if (pci_pcie_type(bridge) == PCI_EXP_TYPE_ROOT_PORT) break; @@ -2812,11 +2755,11 @@ static int dmar_ats_supported(struct pci_dev *dev, struct intel_iommu *iommu) if (atsru->include_all) goto out; } - ret = 0; + supported = false; out: rcu_read_unlock(); - return ret; + return supported; } int dmar_iommu_notify_scope_dev(struct dmar_pci_notify_info *info) @@ -2974,9 +2917,14 @@ static ssize_t domains_used_show(struct device *dev, struct device_attribute *attr, char *buf) { struct intel_iommu *iommu = dev_to_intel_iommu(dev); - return sysfs_emit(buf, "%d\n", - bitmap_weight(iommu->domain_ids, - cap_ndoms(iommu->cap))); + unsigned int count = 0; + int id; + + for (id = 0; id < cap_ndoms(iommu->cap); id++) + if (ida_exists(&iommu->domain_ida, id)) + count++; + + return sysfs_emit(buf, "%d\n", count); } static DEVICE_ATTR_RO(domains_used); @@ -3259,6 +3207,10 @@ void device_block_translation(struct device *dev) struct intel_iommu *iommu = info->iommu; unsigned long flags; + /* Device in DMA blocking state. Noting to do. */ + if (!info->domain_attached) + return; + if (info->domain) cache_tag_unassign_domain(info->domain, dev, IOMMU_NO_PASID); @@ -3270,6 +3222,9 @@ void device_block_translation(struct device *dev) domain_context_clear(info); } + /* Device now in DMA blocking state. */ + info->domain_attached = false; + if (!info->domain) return; @@ -3799,6 +3754,22 @@ static struct iommu_device *intel_iommu_probe_device(struct device *dev) intel_iommu_debugfs_create_dev(info); + return &iommu->iommu; +free_table: + intel_pasid_free_table(dev); +clear_rbtree: + device_rbtree_remove(info); +free: + kfree(info); + + return ERR_PTR(ret); +} + +static void intel_iommu_probe_finalize(struct device *dev) +{ + struct device_domain_info *info = dev_iommu_priv_get(dev); + struct intel_iommu *iommu = info->iommu; + /* * The PCIe spec, in its wisdom, declares that the behaviour of the * device is undefined if you enable PASID support after ATS support. @@ -3806,22 +3777,12 @@ static struct iommu_device *intel_iommu_probe_device(struct device *dev) * we can't yet know if we're ever going to use it. */ if (info->pasid_supported && - !pci_enable_pasid(pdev, info->pasid_supported & ~1)) + !pci_enable_pasid(to_pci_dev(dev), info->pasid_supported & ~1)) info->pasid_enabled = 1; - if (sm_supported(iommu)) + if (sm_supported(iommu) && !dev_is_real_dma_subdevice(dev)) iommu_enable_pci_ats(info); iommu_enable_pci_pri(info); - - return &iommu->iommu; -free_table: - intel_pasid_free_table(dev); -clear_rbtree: - device_rbtree_remove(info); -free: - kfree(info); - - return ERR_PTR(ret); } static void intel_iommu_release_device(struct device *dev) @@ -4349,6 +4310,9 @@ static int identity_domain_attach_dev(struct iommu_domain *domain, struct device else ret = device_setup_pass_through(dev); + if (!ret) + info->domain_attached = true; + return ret; } @@ -4395,6 +4359,7 @@ const struct iommu_ops intel_iommu_ops = { .domain_alloc_sva = intel_svm_domain_alloc, .domain_alloc_nested = intel_iommu_domain_alloc_nested, .probe_device = intel_iommu_probe_device, + .probe_finalize = intel_iommu_probe_finalize, .release_device = intel_iommu_release_device, .get_resv_regions = intel_iommu_get_resv_regions, .device_group = intel_iommu_device_group, @@ -4434,6 +4399,9 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e30, quirk_iommu_igfx); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e40, quirk_iommu_igfx); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e90, quirk_iommu_igfx); +/* QM57/QS57 integrated gfx malfunctions with dmar */ +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0044, quirk_iommu_igfx); + /* Broadwell igfx malfunctions with dmar */ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x1606, quirk_iommu_igfx); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x160B, quirk_iommu_igfx); @@ -4511,7 +4479,6 @@ static void quirk_calpella_no_shadow_gtt(struct pci_dev *dev) } } DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0040, quirk_calpella_no_shadow_gtt); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0044, quirk_calpella_no_shadow_gtt); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x0062, quirk_calpella_no_shadow_gtt); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x006a, quirk_calpella_no_shadow_gtt); diff --git a/drivers/iommu/intel/iommu.h b/drivers/iommu/intel/iommu.h index f9dcf0ab93d1..3ddbcc603de2 100644 --- a/drivers/iommu/intel/iommu.h +++ b/drivers/iommu/intel/iommu.h @@ -721,7 +721,9 @@ struct intel_iommu { unsigned char name[16]; /* Device Name */ #ifdef CONFIG_INTEL_IOMMU - unsigned long *domain_ids; /* bitmap of domains */ + /* mutex to protect domain_ida */ + struct mutex did_lock; + struct ida domain_ida; /* domain id allocator */ unsigned long *copied_tables; /* bitmap of copied tables */ spinlock_t lock; /* protect context, domain ids */ struct root_entry *root_entry; /* virtual address */ @@ -772,6 +774,7 @@ struct device_domain_info { u8 ats_supported:1; u8 ats_enabled:1; u8 dtlb_extra_inval:1; /* Quirk for devices need extra flush */ + u8 domain_attached:1; /* Device has domain attached */ u8 ats_qdep; unsigned int iopf_refcount; struct device *dev; /* it's NULL for PCIe-to-PCI bridge */ @@ -808,11 +811,22 @@ static inline struct dmar_domain *to_dmar_domain(struct iommu_domain *dom) } /* - * Domain ID reserved for pasid entries programmed for first-level - * only and pass-through transfer modes. + * Domain ID 0 and 1 are reserved: + * + * If Caching mode is set, then invalid translations are tagged + * with domain-id 0, hence we need to pre-allocate it. We also + * use domain-id 0 as a marker for non-allocated domain-id, so + * make sure it is not used for a real domain. + * + * Vt-d spec rev3.0 (section 6.2.3.1) requires that each pasid + * entry for first-level or pass-through translation modes should + * be programmed with a domain id different from those used for + * second-level or nested translation. We reserve a domain id for + * this purpose. This domain id is also used for identity domain + * in legacy mode. */ #define FLPT_DEFAULT_DID 1 -#define NUM_RESERVED_DID 2 +#define IDA_START_DID 2 /* Retrieve the domain ID which has allocated to the domain */ static inline u16 diff --git a/drivers/iommu/intel/nested.c b/drivers/iommu/intel/nested.c index d2a94025d0a0..fc312f649f9e 100644 --- a/drivers/iommu/intel/nested.c +++ b/drivers/iommu/intel/nested.c @@ -27,8 +27,7 @@ static int intel_nested_attach_dev(struct iommu_domain *domain, unsigned long flags; int ret = 0; - if (info->domain) - device_block_translation(dev); + device_block_translation(dev); if (iommu->agaw < dmar_domain->s2_domain->agaw) { dev_err_ratelimited(dev, "Adjusted guest address width not compatible\n"); @@ -66,6 +65,7 @@ static int intel_nested_attach_dev(struct iommu_domain *domain, goto disable_iopf; info->domain = dmar_domain; + info->domain_attached = true; spin_lock_irqsave(&dmar_domain->lock, flags); list_add(&info->link, &dmar_domain->devices); spin_unlock_irqrestore(&dmar_domain->lock, flags); diff --git a/drivers/iommu/io-pgtable-arm.c b/drivers/iommu/io-pgtable-arm.c index bbd42323c029..96425e92f313 100644 --- a/drivers/iommu/io-pgtable-arm.c +++ b/drivers/iommu/io-pgtable-arm.c @@ -252,8 +252,6 @@ static inline bool arm_lpae_concat_mandatory(struct io_pgtable_cfg *cfg, (data->start_level == 1) && (oas == 40); } -static bool selftest_running = false; - static dma_addr_t __arm_lpae_dma_addr(void *pages) { return (dma_addr_t)virt_to_phys(pages); @@ -376,7 +374,7 @@ static int arm_lpae_init_pte(struct arm_lpae_io_pgtable *data, for (i = 0; i < num_entries; i++) if (iopte_leaf(ptep[i], lvl, data->iop.fmt)) { /* We require an unmap first */ - WARN_ON(!selftest_running); + WARN_ON(!(data->iop.cfg.quirks & IO_PGTABLE_QUIRK_NO_WARN)); return -EEXIST; } else if (iopte_type(ptep[i]) == ARM_LPAE_PTE_TYPE_TABLE) { /* @@ -478,7 +476,7 @@ static int __arm_lpae_map(struct arm_lpae_io_pgtable *data, unsigned long iova, cptep = iopte_deref(pte, data); } else if (pte) { /* We require an unmap first */ - WARN_ON(!selftest_running); + WARN_ON(!(cfg->quirks & IO_PGTABLE_QUIRK_NO_WARN)); return -EEXIST; } @@ -646,8 +644,10 @@ static size_t __arm_lpae_unmap(struct arm_lpae_io_pgtable *data, unmap_idx_start = ARM_LPAE_LVL_IDX(iova, lvl, data); ptep += unmap_idx_start; pte = READ_ONCE(*ptep); - if (WARN_ON(!pte)) - return 0; + if (!pte) { + WARN_ON(!(data->iop.cfg.quirks & IO_PGTABLE_QUIRK_NO_WARN)); + return -ENOENT; + } /* If the size matches this level, we're in the right place */ if (size == ARM_LPAE_BLOCK_SIZE(lvl, data)) { @@ -657,8 +657,10 @@ static size_t __arm_lpae_unmap(struct arm_lpae_io_pgtable *data, /* Find and handle non-leaf entries */ for (i = 0; i < num_entries; i++) { pte = READ_ONCE(ptep[i]); - if (WARN_ON(!pte)) + if (!pte) { + WARN_ON(!(data->iop.cfg.quirks & IO_PGTABLE_QUIRK_NO_WARN)); break; + } if (!iopte_leaf(pte, lvl, iop->fmt)) { __arm_lpae_clear_pte(&ptep[i], &iop->cfg, 1); @@ -973,7 +975,8 @@ arm_64_lpae_alloc_pgtable_s1(struct io_pgtable_cfg *cfg, void *cookie) if (cfg->quirks & ~(IO_PGTABLE_QUIRK_ARM_NS | IO_PGTABLE_QUIRK_ARM_TTBR1 | IO_PGTABLE_QUIRK_ARM_OUTER_WBWA | - IO_PGTABLE_QUIRK_ARM_HD)) + IO_PGTABLE_QUIRK_ARM_HD | + IO_PGTABLE_QUIRK_NO_WARN)) return NULL; data = arm_lpae_alloc_pgtable(cfg); @@ -1074,7 +1077,8 @@ arm_64_lpae_alloc_pgtable_s2(struct io_pgtable_cfg *cfg, void *cookie) struct arm_lpae_io_pgtable *data; typeof(&cfg->arm_lpae_s2_cfg.vtcr) vtcr = &cfg->arm_lpae_s2_cfg.vtcr; - if (cfg->quirks & ~(IO_PGTABLE_QUIRK_ARM_S2FWB)) + if (cfg->quirks & ~(IO_PGTABLE_QUIRK_ARM_S2FWB | + IO_PGTABLE_QUIRK_NO_WARN)) return NULL; data = arm_lpae_alloc_pgtable(cfg); @@ -1315,7 +1319,6 @@ static void __init arm_lpae_dump_ops(struct io_pgtable_ops *ops) #define __FAIL(ops, i) ({ \ WARN(1, "selftest: test failed for fmt idx %d\n", (i)); \ arm_lpae_dump_ops(ops); \ - selftest_running = false; \ -EFAULT; \ }) @@ -1331,8 +1334,6 @@ static int __init arm_lpae_run_tests(struct io_pgtable_cfg *cfg) size_t size, mapped; struct io_pgtable_ops *ops; - selftest_running = true; - for (i = 0; i < ARRAY_SIZE(fmts); ++i) { cfg_cookie = cfg; ops = alloc_io_pgtable_ops(fmts[i], cfg, cfg); @@ -1421,7 +1422,6 @@ static int __init arm_lpae_run_tests(struct io_pgtable_cfg *cfg) free_io_pgtable_ops(ops); } - selftest_running = false; return 0; } @@ -1442,6 +1442,7 @@ static int __init arm_lpae_do_selftests(void) struct io_pgtable_cfg cfg = { .tlb = &dummy_tlb_ops, .coherent_walk = true, + .quirks = IO_PGTABLE_QUIRK_NO_WARN, }; dev = faux_device_create("io-pgtable-test", NULL, 0); diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 900fb782c3a6..60a86579baba 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -3362,10 +3362,12 @@ static int __iommu_set_group_pasid(struct iommu_domain *domain, int ret; for_each_group_device(group, device) { - ret = domain->ops->set_dev_pasid(domain, device->dev, - pasid, old); - if (ret) - goto err_revert; + if (device->dev->iommu->max_pasids > 0) { + ret = domain->ops->set_dev_pasid(domain, device->dev, + pasid, old); + if (ret) + goto err_revert; + } } return 0; @@ -3375,15 +3377,18 @@ err_revert: for_each_group_device(group, device) { if (device == last_gdev) break; - /* - * If no old domain, undo the succeeded devices/pasid. - * Otherwise, rollback the succeeded devices/pasid to the old - * domain. And it is a driver bug to fail attaching with a - * previously good domain. - */ - if (!old || WARN_ON(old->ops->set_dev_pasid(old, device->dev, + if (device->dev->iommu->max_pasids > 0) { + /* + * If no old domain, undo the succeeded devices/pasid. + * Otherwise, rollback the succeeded devices/pasid to + * the old domain. And it is a driver bug to fail + * attaching with a previously good domain. + */ + if (!old || + WARN_ON(old->ops->set_dev_pasid(old, device->dev, pasid, domain))) - iommu_remove_dev_pasid(device->dev, pasid, domain); + iommu_remove_dev_pasid(device->dev, pasid, domain); + } } return ret; } @@ -3394,8 +3399,10 @@ static void __iommu_remove_group_pasid(struct iommu_group *group, { struct group_device *device; - for_each_group_device(group, device) - iommu_remove_dev_pasid(device->dev, pasid, domain); + for_each_group_device(group, device) { + if (device->dev->iommu->max_pasids > 0) + iommu_remove_dev_pasid(device->dev, pasid, domain); + } } /* @@ -3437,7 +3444,13 @@ int iommu_attach_device_pasid(struct iommu_domain *domain, mutex_lock(&group->mutex); for_each_group_device(group, device) { - if (pasid >= device->dev->iommu->max_pasids) { + /* + * Skip PASID validation for devices without PASID support + * (max_pasids = 0). These devices cannot issue transactions + * with PASID, so they don't affect group's PASID usage. + */ + if ((device->dev->iommu->max_pasids > 0) && + (pasid >= device->dev->iommu->max_pasids)) { ret = -EINVAL; goto out_unlock; } diff --git a/drivers/iommu/ipmmu-vmsa.c b/drivers/iommu/ipmmu-vmsa.c index e424b279a8cd..90341b24a811 100644 --- a/drivers/iommu/ipmmu-vmsa.c +++ b/drivers/iommu/ipmmu-vmsa.c @@ -1090,7 +1090,8 @@ static int ipmmu_probe(struct platform_device *pdev) if (mmu->features->has_cache_leaf_nodes && ipmmu_is_root(mmu)) return 0; - ret = iommu_device_sysfs_add(&mmu->iommu, &pdev->dev, NULL, dev_name(&pdev->dev)); + ret = iommu_device_sysfs_add(&mmu->iommu, &pdev->dev, NULL, "%s", + dev_name(&pdev->dev)); if (ret) return ret; diff --git a/drivers/iommu/mtk_iommu.c b/drivers/iommu/mtk_iommu.c index df98d0c65f54..cb95fecf6016 100644 --- a/drivers/iommu/mtk_iommu.c +++ b/drivers/iommu/mtk_iommu.c @@ -1550,6 +1550,31 @@ static const struct mtk_iommu_plat_data mt6795_data = { .larbid_remap = {{0}, {1}, {2}, {3}, {4}}, /* Linear mapping. */ }; +static const unsigned int mt8192_larb_region_msk[MT8192_MULTI_REGION_NR_MAX][MTK_LARB_NR_MAX] = { + [0] = {~0, ~0}, /* Region0: larb0/1 */ + [1] = {0, 0, 0, 0, ~0, ~0, 0, ~0}, /* Region1: larb4/5/7 */ + [2] = {0, 0, ~0, 0, 0, 0, 0, 0, /* Region2: larb2/9/11/13/14/16/17/18/19/20 */ + 0, ~0, 0, ~0, 0, ~(u32)(BIT(9) | BIT(10)), ~(u32)(BIT(4) | BIT(5)), 0, + ~0, ~0, ~0, ~0, ~0}, + [3] = {0}, + [4] = {[13] = BIT(9) | BIT(10)}, /* larb13 port9/10 */ + [5] = {[14] = BIT(4) | BIT(5)}, /* larb14 port4/5 */ +}; + +static const struct mtk_iommu_plat_data mt6893_data = { + .m4u_plat = M4U_MT8192, + .flags = HAS_BCLK | OUT_ORDER_WR_EN | HAS_SUB_COMM_2BITS | + WR_THROT_EN | IOVA_34_EN | SHARE_PGTABLE | MTK_IOMMU_TYPE_MM, + .inv_sel_reg = REG_MMU_INV_SEL_GEN2, + .banks_num = 1, + .banks_enable = {true}, + .iova_region = mt8192_multi_dom, + .iova_region_nr = ARRAY_SIZE(mt8192_multi_dom), + .iova_region_larb_msk = mt8192_larb_region_msk, + .larbid_remap = {{0}, {1}, {4, 5}, {7}, {2}, {9, 11, 19, 20}, + {0, 14, 16}, {0, 13, 18, 17}}, +}; + static const struct mtk_iommu_plat_data mt8167_data = { .m4u_plat = M4U_MT8167, .flags = RESET_AXI | HAS_LEGACY_IVRP_PADDR | MTK_IOMMU_TYPE_MM, @@ -1673,17 +1698,6 @@ static const struct mtk_iommu_plat_data mt8188_data_vpp = { 27, 28 /* ccu0 */, MTK_INVALID_LARBID}, {4, 6}}, }; -static const unsigned int mt8192_larb_region_msk[MT8192_MULTI_REGION_NR_MAX][MTK_LARB_NR_MAX] = { - [0] = {~0, ~0}, /* Region0: larb0/1 */ - [1] = {0, 0, 0, 0, ~0, ~0, 0, ~0}, /* Region1: larb4/5/7 */ - [2] = {0, 0, ~0, 0, 0, 0, 0, 0, /* Region2: larb2/9/11/13/14/16/17/18/19/20 */ - 0, ~0, 0, ~0, 0, ~(u32)(BIT(9) | BIT(10)), ~(u32)(BIT(4) | BIT(5)), 0, - ~0, ~0, ~0, ~0, ~0}, - [3] = {0}, - [4] = {[13] = BIT(9) | BIT(10)}, /* larb13 port9/10 */ - [5] = {[14] = BIT(4) | BIT(5)}, /* larb14 port4/5 */ -}; - static const struct mtk_iommu_plat_data mt8192_data = { .m4u_plat = M4U_MT8192, .flags = HAS_BCLK | HAS_SUB_COMM_2BITS | OUT_ORDER_WR_EN | @@ -1777,6 +1791,7 @@ static const struct of_device_id mtk_iommu_of_ids[] = { { .compatible = "mediatek,mt2712-m4u", .data = &mt2712_data}, { .compatible = "mediatek,mt6779-m4u", .data = &mt6779_data}, { .compatible = "mediatek,mt6795-m4u", .data = &mt6795_data}, + { .compatible = "mediatek,mt6893-iommu-mm", .data = &mt6893_data}, { .compatible = "mediatek,mt8167-m4u", .data = &mt8167_data}, { .compatible = "mediatek,mt8173-m4u", .data = &mt8173_data}, { .compatible = "mediatek,mt8183-m4u", .data = &mt8183_data}, diff --git a/drivers/iommu/s390-iommu.c b/drivers/iommu/s390-iommu.c index e1c76e0f9c2b..433b59f43530 100644 --- a/drivers/iommu/s390-iommu.c +++ b/drivers/iommu/s390-iommu.c @@ -31,10 +31,21 @@ struct s390_domain { unsigned long *dma_table; spinlock_t list_lock; struct rcu_head rcu; + u8 origin_type; }; static struct iommu_domain blocking_domain; +static inline unsigned int calc_rfx(dma_addr_t ptr) +{ + return ((unsigned long)ptr >> ZPCI_RF_SHIFT) & ZPCI_INDEX_MASK; +} + +static inline unsigned int calc_rsx(dma_addr_t ptr) +{ + return ((unsigned long)ptr >> ZPCI_RS_SHIFT) & ZPCI_INDEX_MASK; +} + static inline unsigned int calc_rtx(dma_addr_t ptr) { return ((unsigned long)ptr >> ZPCI_RT_SHIFT) & ZPCI_INDEX_MASK; @@ -56,6 +67,20 @@ static inline void set_pt_pfaa(unsigned long *entry, phys_addr_t pfaa) *entry |= (pfaa & ZPCI_PTE_ADDR_MASK); } +static inline void set_rf_rso(unsigned long *entry, phys_addr_t rso) +{ + *entry &= ZPCI_RTE_FLAG_MASK; + *entry |= (rso & ZPCI_RTE_ADDR_MASK); + *entry |= ZPCI_TABLE_TYPE_RFX; +} + +static inline void set_rs_rto(unsigned long *entry, phys_addr_t rto) +{ + *entry &= ZPCI_RTE_FLAG_MASK; + *entry |= (rto & ZPCI_RTE_ADDR_MASK); + *entry |= ZPCI_TABLE_TYPE_RSX; +} + static inline void set_rt_sto(unsigned long *entry, phys_addr_t sto) { *entry &= ZPCI_RTE_FLAG_MASK; @@ -70,6 +95,22 @@ static inline void set_st_pto(unsigned long *entry, phys_addr_t pto) *entry |= ZPCI_TABLE_TYPE_SX; } +static inline void validate_rf_entry(unsigned long *entry) +{ + *entry &= ~ZPCI_TABLE_VALID_MASK; + *entry &= ~ZPCI_TABLE_OFFSET_MASK; + *entry |= ZPCI_TABLE_VALID; + *entry |= ZPCI_TABLE_LEN_RFX; +} + +static inline void validate_rs_entry(unsigned long *entry) +{ + *entry &= ~ZPCI_TABLE_VALID_MASK; + *entry &= ~ZPCI_TABLE_OFFSET_MASK; + *entry |= ZPCI_TABLE_VALID; + *entry |= ZPCI_TABLE_LEN_RSX; +} + static inline void validate_rt_entry(unsigned long *entry) { *entry &= ~ZPCI_TABLE_VALID_MASK; @@ -120,6 +161,22 @@ static inline int pt_entry_isvalid(unsigned long entry) return (entry & ZPCI_PTE_VALID_MASK) == ZPCI_PTE_VALID; } +static inline unsigned long *get_rf_rso(unsigned long entry) +{ + if ((entry & ZPCI_TABLE_TYPE_MASK) == ZPCI_TABLE_TYPE_RFX) + return phys_to_virt(entry & ZPCI_RTE_ADDR_MASK); + else + return NULL; +} + +static inline unsigned long *get_rs_rto(unsigned long entry) +{ + if ((entry & ZPCI_TABLE_TYPE_MASK) == ZPCI_TABLE_TYPE_RSX) + return phys_to_virt(entry & ZPCI_RTE_ADDR_MASK); + else + return NULL; +} + static inline unsigned long *get_rt_sto(unsigned long entry) { if ((entry & ZPCI_TABLE_TYPE_MASK) == ZPCI_TABLE_TYPE_RTX) @@ -191,18 +248,59 @@ static void dma_free_seg_table(unsigned long entry) dma_free_cpu_table(sto); } -static void dma_cleanup_tables(unsigned long *table) +static void dma_free_rt_table(unsigned long entry) { + unsigned long *rto = get_rs_rto(entry); int rtx; - if (!table) + for (rtx = 0; rtx < ZPCI_TABLE_ENTRIES; rtx++) + if (reg_entry_isvalid(rto[rtx])) + dma_free_seg_table(rto[rtx]); + + dma_free_cpu_table(rto); +} + +static void dma_free_rs_table(unsigned long entry) +{ + unsigned long *rso = get_rf_rso(entry); + int rsx; + + for (rsx = 0; rsx < ZPCI_TABLE_ENTRIES; rsx++) + if (reg_entry_isvalid(rso[rsx])) + dma_free_rt_table(rso[rsx]); + + dma_free_cpu_table(rso); +} + +static void dma_cleanup_tables(struct s390_domain *domain) +{ + int rtx, rsx, rfx; + + if (!domain->dma_table) return; - for (rtx = 0; rtx < ZPCI_TABLE_ENTRIES; rtx++) - if (reg_entry_isvalid(table[rtx])) - dma_free_seg_table(table[rtx]); + switch (domain->origin_type) { + case ZPCI_TABLE_TYPE_RFX: + for (rfx = 0; rfx < ZPCI_TABLE_ENTRIES; rfx++) + if (reg_entry_isvalid(domain->dma_table[rfx])) + dma_free_rs_table(domain->dma_table[rfx]); + break; + case ZPCI_TABLE_TYPE_RSX: + for (rsx = 0; rsx < ZPCI_TABLE_ENTRIES; rsx++) + if (reg_entry_isvalid(domain->dma_table[rsx])) + dma_free_rt_table(domain->dma_table[rsx]); + break; + case ZPCI_TABLE_TYPE_RTX: + for (rtx = 0; rtx < ZPCI_TABLE_ENTRIES; rtx++) + if (reg_entry_isvalid(domain->dma_table[rtx])) + dma_free_seg_table(domain->dma_table[rtx]); + break; + default: + WARN_ONCE(1, "Invalid IOMMU table (%x)\n", domain->origin_type); + return; + } - dma_free_cpu_table(table); + dma_free_cpu_table(domain->dma_table); } static unsigned long *dma_alloc_page_table(gfp_t gfp) @@ -218,6 +316,70 @@ static unsigned long *dma_alloc_page_table(gfp_t gfp) return table; } +static unsigned long *dma_walk_rs_table(unsigned long *rso, + dma_addr_t dma_addr, gfp_t gfp) +{ + unsigned int rsx = calc_rsx(dma_addr); + unsigned long old_rse, rse; + unsigned long *rsep, *rto; + + rsep = &rso[rsx]; + rse = READ_ONCE(*rsep); + if (reg_entry_isvalid(rse)) { + rto = get_rs_rto(rse); + } else { + rto = dma_alloc_cpu_table(gfp); + if (!rto) + return NULL; + + set_rs_rto(&rse, virt_to_phys(rto)); + validate_rs_entry(&rse); + entry_clr_protected(&rse); + + old_rse = cmpxchg(rsep, ZPCI_TABLE_INVALID, rse); + if (old_rse != ZPCI_TABLE_INVALID) { + /* Somone else was faster, use theirs */ + dma_free_cpu_table(rto); + rto = get_rs_rto(old_rse); + } + } + return rto; +} + +static unsigned long *dma_walk_rf_table(unsigned long *rfo, + dma_addr_t dma_addr, gfp_t gfp) +{ + unsigned int rfx = calc_rfx(dma_addr); + unsigned long old_rfe, rfe; + unsigned long *rfep, *rso; + + rfep = &rfo[rfx]; + rfe = READ_ONCE(*rfep); + if (reg_entry_isvalid(rfe)) { + rso = get_rf_rso(rfe); + } else { + rso = dma_alloc_cpu_table(gfp); + if (!rso) + return NULL; + + set_rf_rso(&rfe, virt_to_phys(rso)); + validate_rf_entry(&rfe); + entry_clr_protected(&rfe); + + old_rfe = cmpxchg(rfep, ZPCI_TABLE_INVALID, rfe); + if (old_rfe != ZPCI_TABLE_INVALID) { + /* Somone else was faster, use theirs */ + dma_free_cpu_table(rso); + rso = get_rf_rso(old_rfe); + } + } + + if (!rso) + return NULL; + + return dma_walk_rs_table(rso, dma_addr, gfp); +} + static unsigned long *dma_get_seg_table_origin(unsigned long *rtep, gfp_t gfp) { unsigned long old_rte, rte; @@ -271,11 +433,31 @@ static unsigned long *dma_get_page_table_origin(unsigned long *step, gfp_t gfp) return pto; } -static unsigned long *dma_walk_cpu_trans(unsigned long *rto, dma_addr_t dma_addr, gfp_t gfp) +static unsigned long *dma_walk_region_tables(struct s390_domain *domain, + dma_addr_t dma_addr, gfp_t gfp) +{ + switch (domain->origin_type) { + case ZPCI_TABLE_TYPE_RFX: + return dma_walk_rf_table(domain->dma_table, dma_addr, gfp); + case ZPCI_TABLE_TYPE_RSX: + return dma_walk_rs_table(domain->dma_table, dma_addr, gfp); + case ZPCI_TABLE_TYPE_RTX: + return domain->dma_table; + default: + return NULL; + } +} + +static unsigned long *dma_walk_cpu_trans(struct s390_domain *domain, + dma_addr_t dma_addr, gfp_t gfp) { - unsigned long *sto, *pto; + unsigned long *rto, *sto, *pto; unsigned int rtx, sx, px; + rto = dma_walk_region_tables(domain, dma_addr, gfp); + if (!rto) + return NULL; + rtx = calc_rtx(dma_addr); sto = dma_get_seg_table_origin(&rto[rtx], gfp); if (!sto) @@ -329,9 +511,25 @@ static bool s390_iommu_capable(struct device *dev, enum iommu_cap cap) } } +static inline u64 max_tbl_size(struct s390_domain *domain) +{ + switch (domain->origin_type) { + case ZPCI_TABLE_TYPE_RTX: + return ZPCI_TABLE_SIZE_RT - 1; + case ZPCI_TABLE_TYPE_RSX: + return ZPCI_TABLE_SIZE_RS - 1; + case ZPCI_TABLE_TYPE_RFX: + return U64_MAX; + default: + return 0; + } +} + static struct iommu_domain *s390_domain_alloc_paging(struct device *dev) { + struct zpci_dev *zdev = to_zpci_dev(dev); struct s390_domain *s390_domain; + u64 aperture_size; s390_domain = kzalloc(sizeof(*s390_domain), GFP_KERNEL); if (!s390_domain) @@ -342,9 +540,26 @@ static struct iommu_domain *s390_domain_alloc_paging(struct device *dev) kfree(s390_domain); return NULL; } + + aperture_size = min(s390_iommu_aperture, + zdev->end_dma - zdev->start_dma + 1); + if (aperture_size <= (ZPCI_TABLE_SIZE_RT - zdev->start_dma)) { + s390_domain->origin_type = ZPCI_TABLE_TYPE_RTX; + } else if (aperture_size <= (ZPCI_TABLE_SIZE_RS - zdev->start_dma) && + (zdev->dtsm & ZPCI_IOTA_DT_RS)) { + s390_domain->origin_type = ZPCI_TABLE_TYPE_RSX; + } else if (zdev->dtsm & ZPCI_IOTA_DT_RF) { + s390_domain->origin_type = ZPCI_TABLE_TYPE_RFX; + } else { + /* Assume RTX available */ + s390_domain->origin_type = ZPCI_TABLE_TYPE_RTX; + aperture_size = ZPCI_TABLE_SIZE_RT - zdev->start_dma; + } + zdev->end_dma = zdev->start_dma + aperture_size - 1; + s390_domain->domain.geometry.force_aperture = true; s390_domain->domain.geometry.aperture_start = 0; - s390_domain->domain.geometry.aperture_end = ZPCI_TABLE_SIZE_RT - 1; + s390_domain->domain.geometry.aperture_end = max_tbl_size(s390_domain); spin_lock_init(&s390_domain->list_lock); INIT_LIST_HEAD_RCU(&s390_domain->devices); @@ -356,7 +571,7 @@ static void s390_iommu_rcu_free_domain(struct rcu_head *head) { struct s390_domain *s390_domain = container_of(head, struct s390_domain, rcu); - dma_cleanup_tables(s390_domain->dma_table); + dma_cleanup_tables(s390_domain); kfree(s390_domain); } @@ -381,6 +596,21 @@ static void zdev_s390_domain_update(struct zpci_dev *zdev, spin_unlock_irqrestore(&zdev->dom_lock, flags); } +static u64 get_iota_region_flag(struct s390_domain *domain) +{ + switch (domain->origin_type) { + case ZPCI_TABLE_TYPE_RTX: + return ZPCI_IOTA_RTTO_FLAG; + case ZPCI_TABLE_TYPE_RSX: + return ZPCI_IOTA_RSTO_FLAG; + case ZPCI_TABLE_TYPE_RFX: + return ZPCI_IOTA_RFTO_FLAG; + default: + WARN_ONCE(1, "Invalid IOMMU table (%x)\n", domain->origin_type); + return 0; + } +} + static int s390_iommu_domain_reg_ioat(struct zpci_dev *zdev, struct iommu_domain *domain, u8 *status) { @@ -399,7 +629,7 @@ static int s390_iommu_domain_reg_ioat(struct zpci_dev *zdev, default: s390_domain = to_s390_domain(domain); iota = virt_to_phys(s390_domain->dma_table) | - ZPCI_IOTA_RTTO_FLAG; + get_iota_region_flag(s390_domain); rc = zpci_register_ioat(zdev, 0, zdev->start_dma, zdev->end_dma, iota, status); } @@ -482,6 +712,8 @@ static void s390_iommu_get_resv_regions(struct device *dev, { struct zpci_dev *zdev = to_zpci_dev(dev); struct iommu_resv_region *region; + u64 max_size, end_resv; + unsigned long flags; if (zdev->start_dma) { region = iommu_alloc_resv_region(0, zdev->start_dma, 0, @@ -491,10 +723,21 @@ static void s390_iommu_get_resv_regions(struct device *dev, list_add_tail(®ion->list, list); } - if (zdev->end_dma < ZPCI_TABLE_SIZE_RT - 1) { - region = iommu_alloc_resv_region(zdev->end_dma + 1, - ZPCI_TABLE_SIZE_RT - zdev->end_dma - 1, - 0, IOMMU_RESV_RESERVED, GFP_KERNEL); + spin_lock_irqsave(&zdev->dom_lock, flags); + if (zdev->s390_domain->type == IOMMU_DOMAIN_BLOCKED || + zdev->s390_domain->type == IOMMU_DOMAIN_IDENTITY) { + spin_unlock_irqrestore(&zdev->dom_lock, flags); + return; + } + + max_size = max_tbl_size(to_s390_domain(zdev->s390_domain)); + spin_unlock_irqrestore(&zdev->dom_lock, flags); + + if (zdev->end_dma < max_size) { + end_resv = max_size - zdev->end_dma; + region = iommu_alloc_resv_region(zdev->end_dma + 1, end_resv, + 0, IOMMU_RESV_RESERVED, + GFP_KERNEL); if (!region) return; list_add_tail(®ion->list, list); @@ -510,13 +753,9 @@ static struct iommu_device *s390_iommu_probe_device(struct device *dev) zdev = to_zpci_dev(dev); - if (zdev->start_dma > zdev->end_dma || - zdev->start_dma > ZPCI_TABLE_SIZE_RT - 1) + if (zdev->start_dma > zdev->end_dma) return ERR_PTR(-EINVAL); - if (zdev->end_dma > ZPCI_TABLE_SIZE_RT - 1) - zdev->end_dma = ZPCI_TABLE_SIZE_RT - 1; - if (zdev->tlb_refresh) dev->iommu->shadow_on_flush = 1; @@ -606,8 +845,7 @@ static int s390_iommu_validate_trans(struct s390_domain *s390_domain, int rc; for (i = 0; i < nr_pages; i++) { - entry = dma_walk_cpu_trans(s390_domain->dma_table, dma_addr, - gfp); + entry = dma_walk_cpu_trans(s390_domain, dma_addr, gfp); if (unlikely(!entry)) { rc = -ENOMEM; goto undo_cpu_trans; @@ -622,8 +860,7 @@ static int s390_iommu_validate_trans(struct s390_domain *s390_domain, undo_cpu_trans: while (i-- > 0) { dma_addr -= PAGE_SIZE; - entry = dma_walk_cpu_trans(s390_domain->dma_table, - dma_addr, gfp); + entry = dma_walk_cpu_trans(s390_domain, dma_addr, gfp); if (!entry) break; dma_update_cpu_trans(entry, 0, ZPCI_PTE_INVALID); @@ -640,8 +877,7 @@ static int s390_iommu_invalidate_trans(struct s390_domain *s390_domain, int rc = 0; for (i = 0; i < nr_pages; i++) { - entry = dma_walk_cpu_trans(s390_domain->dma_table, dma_addr, - GFP_ATOMIC); + entry = dma_walk_cpu_trans(s390_domain, dma_addr, GFP_ATOMIC); if (unlikely(!entry)) { rc = -EINVAL; break; @@ -685,6 +921,51 @@ static int s390_iommu_map_pages(struct iommu_domain *domain, return rc; } +static unsigned long *get_rso_from_iova(struct s390_domain *domain, + dma_addr_t iova) +{ + unsigned long *rfo; + unsigned long rfe; + unsigned int rfx; + + switch (domain->origin_type) { + case ZPCI_TABLE_TYPE_RFX: + rfo = domain->dma_table; + rfx = calc_rfx(iova); + rfe = READ_ONCE(rfo[rfx]); + if (!reg_entry_isvalid(rfe)) + return NULL; + return get_rf_rso(rfe); + case ZPCI_TABLE_TYPE_RSX: + return domain->dma_table; + default: + return NULL; + } +} + +static unsigned long *get_rto_from_iova(struct s390_domain *domain, + dma_addr_t iova) +{ + unsigned long *rso; + unsigned long rse; + unsigned int rsx; + + switch (domain->origin_type) { + case ZPCI_TABLE_TYPE_RFX: + case ZPCI_TABLE_TYPE_RSX: + rso = get_rso_from_iova(domain, iova); + rsx = calc_rsx(iova); + rse = READ_ONCE(rso[rsx]); + if (!reg_entry_isvalid(rse)) + return NULL; + return get_rs_rto(rse); + case ZPCI_TABLE_TYPE_RTX: + return domain->dma_table; + default: + return NULL; + } +} + static phys_addr_t s390_iommu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova) { @@ -698,10 +979,13 @@ static phys_addr_t s390_iommu_iova_to_phys(struct iommu_domain *domain, iova > domain->geometry.aperture_end) return 0; + rto = get_rto_from_iova(s390_domain, iova); + if (!rto) + return 0; + rtx = calc_rtx(iova); sx = calc_sx(iova); px = calc_px(iova); - rto = s390_domain->dma_table; rte = READ_ONCE(rto[rtx]); if (reg_entry_isvalid(rte)) { @@ -756,7 +1040,6 @@ struct zpci_iommu_ctrs *zpci_get_iommu_ctrs(struct zpci_dev *zdev) int zpci_init_iommu(struct zpci_dev *zdev) { - u64 aperture_size; int rc = 0; rc = iommu_device_sysfs_add(&zdev->iommu_dev, NULL, NULL, @@ -774,12 +1057,6 @@ int zpci_init_iommu(struct zpci_dev *zdev) if (rc) goto out_sysfs; - zdev->start_dma = PAGE_ALIGN(zdev->start_dma); - aperture_size = min3(s390_iommu_aperture, - ZPCI_TABLE_SIZE_RT - zdev->start_dma, - zdev->end_dma - zdev->start_dma + 1); - zdev->end_dma = zdev->start_dma + aperture_size - 1; - return 0; out_sysfs: |