diff options
Diffstat (limited to 'drivers/iommu')
-rw-r--r-- | drivers/iommu/Makefile | 1 | ||||
-rw-r--r-- | drivers/iommu/amd_iommu.c | 196 | ||||
-rw-r--r-- | drivers/iommu/amd_iommu_init.c | 34 | ||||
-rw-r--r-- | drivers/iommu/amd_iommu_types.h | 1 | ||||
-rw-r--r-- | drivers/iommu/intel-iommu.c | 50 | ||||
-rw-r--r-- | drivers/iommu/omap-iommu-debug.c | 8 | ||||
-rw-r--r-- | drivers/iommu/omap-iommu.c | 113 | ||||
-rw-r--r-- | drivers/iommu/omap-iommu.h | 225 | ||||
-rw-r--r-- | drivers/iommu/omap-iommu2.c | 334 | ||||
-rw-r--r-- | drivers/iommu/omap-iopgtable.h | 98 | ||||
-rw-r--r-- | drivers/iommu/omap-iovmm.c | 50 | ||||
-rw-r--r-- | drivers/iommu/tegra-gart.c | 6 | ||||
-rw-r--r-- | drivers/iommu/tegra-smmu.c | 15 |
13 files changed, 1017 insertions, 114 deletions
diff --git a/drivers/iommu/Makefile b/drivers/iommu/Makefile index 14a4d5f..f66b816 100644 --- a/drivers/iommu/Makefile +++ b/drivers/iommu/Makefile @@ -7,6 +7,7 @@ obj-$(CONFIG_DMAR_TABLE) += dmar.o obj-$(CONFIG_INTEL_IOMMU) += iova.o intel-iommu.o obj-$(CONFIG_IRQ_REMAP) += intel_irq_remapping.o irq_remapping.o obj-$(CONFIG_OMAP_IOMMU) += omap-iommu.o +obj-$(CONFIG_OMAP_IOMMU) += omap-iommu2.o obj-$(CONFIG_OMAP_IOVMM) += omap-iovmm.o obj-$(CONFIG_OMAP_IOMMU_DEBUG) += omap-iommu-debug.o obj-$(CONFIG_TEGRA_IOMMU_GART) += tegra-gart.o diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c index 55074cb..c1c74e0 100644 --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -57,17 +57,9 @@ * physically contiguous memory regions it is mapping into page sizes * that we support. * - * Traditionally the IOMMU core just handed us the mappings directly, - * after making sure the size is an order of a 4KiB page and that the - * mapping has natural alignment. - * - * To retain this behavior, we currently advertise that we support - * all page sizes that are an order of 4KiB. - * - * If at some point we'd like to utilize the IOMMU core's new behavior, - * we could change this to advertise the real page sizes we support. + * 512GB Pages are not supported due to a hardware bug */ -#define AMD_IOMMU_PGSIZES (~0xFFFUL) +#define AMD_IOMMU_PGSIZES ((~0xFFFUL) & ~(2ULL << 38)) static DEFINE_RWLOCK(amd_iommu_devtable_lock); @@ -140,6 +132,9 @@ static void free_dev_data(struct iommu_dev_data *dev_data) list_del(&dev_data->dev_data_list); spin_unlock_irqrestore(&dev_data_list_lock, flags); + if (dev_data->group) + iommu_group_put(dev_data->group); + kfree(dev_data); } @@ -274,41 +269,23 @@ static void swap_pci_ref(struct pci_dev **from, struct pci_dev *to) *from = to; } -#define REQ_ACS_FLAGS (PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF) - -static int iommu_init_device(struct device *dev) +static struct pci_bus *find_hosted_bus(struct pci_bus *bus) { - struct pci_dev *dma_pdev = NULL, *pdev = to_pci_dev(dev); - struct iommu_dev_data *dev_data; - struct iommu_group *group; - u16 alias; - int ret; - - if (dev->archdata.iommu) - return 0; - - dev_data = find_dev_data(get_device_id(dev)); - if (!dev_data) - return -ENOMEM; - - alias = amd_iommu_alias_table[dev_data->devid]; - if (alias != dev_data->devid) { - struct iommu_dev_data *alias_data; + while (!bus->self) { + if (!pci_is_root_bus(bus)) + bus = bus->parent; + else + return ERR_PTR(-ENODEV); + } - alias_data = find_dev_data(alias); - if (alias_data == NULL) { - pr_err("AMD-Vi: Warning: Unhandled device %s\n", - dev_name(dev)); - free_dev_data(dev_data); - return -ENOTSUPP; - } - dev_data->alias_data = alias_data; + return bus; +} - dma_pdev = pci_get_bus_and_slot(alias >> 8, alias & 0xff); - } +#define REQ_ACS_FLAGS (PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF) - if (dma_pdev == NULL) - dma_pdev = pci_dev_get(pdev); +static struct pci_dev *get_isolation_root(struct pci_dev *pdev) +{ + struct pci_dev *dma_pdev = pdev; /* Account for quirked devices */ swap_pci_ref(&dma_pdev, pci_get_dma_source(dma_pdev)); @@ -330,14 +307,9 @@ static int iommu_init_device(struct device *dev) * Finding the next device may require skipping virtual buses. */ while (!pci_is_root_bus(dma_pdev->bus)) { - struct pci_bus *bus = dma_pdev->bus; - - while (!bus->self) { - if (!pci_is_root_bus(bus)) - bus = bus->parent; - else - goto root_bus; - } + struct pci_bus *bus = find_hosted_bus(dma_pdev->bus); + if (IS_ERR(bus)) + break; if (pci_acs_path_enabled(bus->self, NULL, REQ_ACS_FLAGS)) break; @@ -345,19 +317,137 @@ static int iommu_init_device(struct device *dev) swap_pci_ref(&dma_pdev, pci_dev_get(bus->self)); } -root_bus: - group = iommu_group_get(&dma_pdev->dev); - pci_dev_put(dma_pdev); + return dma_pdev; +} + +static int use_pdev_iommu_group(struct pci_dev *pdev, struct device *dev) +{ + struct iommu_group *group = iommu_group_get(&pdev->dev); + int ret; + if (!group) { group = iommu_group_alloc(); if (IS_ERR(group)) return PTR_ERR(group); + + WARN_ON(&pdev->dev != dev); } ret = iommu_group_add_device(group, dev); - iommu_group_put(group); + return ret; +} + +static int use_dev_data_iommu_group(struct iommu_dev_data *dev_data, + struct device *dev) +{ + if (!dev_data->group) { + struct iommu_group *group = iommu_group_alloc(); + if (IS_ERR(group)) + return PTR_ERR(group); + + dev_data->group = group; + } + + return iommu_group_add_device(dev_data->group, dev); +} + +static int init_iommu_group(struct device *dev) +{ + struct iommu_dev_data *dev_data; + struct iommu_group *group; + struct pci_dev *dma_pdev; + int ret; + + group = iommu_group_get(dev); + if (group) { + iommu_group_put(group); + return 0; + } + + dev_data = find_dev_data(get_device_id(dev)); + if (!dev_data) + return -ENOMEM; + + if (dev_data->alias_data) { + u16 alias; + struct pci_bus *bus; + + if (dev_data->alias_data->group) + goto use_group; + + /* + * If the alias device exists, it's effectively just a first + * level quirk for finding the DMA source. + */ + alias = amd_iommu_alias_table[dev_data->devid]; + dma_pdev = pci_get_bus_and_slot(alias >> 8, alias & 0xff); + if (dma_pdev) { + dma_pdev = get_isolation_root(dma_pdev); + goto use_pdev; + } + + /* + * If the alias is virtual, try to find a parent device + * and test whether the IOMMU group is actualy rooted above + * the alias. Be careful to also test the parent device if + * we think the alias is the root of the group. + */ + bus = pci_find_bus(0, alias >> 8); + if (!bus) + goto use_group; + + bus = find_hosted_bus(bus); + if (IS_ERR(bus) || !bus->self) + goto use_group; + + dma_pdev = get_isolation_root(pci_dev_get(bus->self)); + if (dma_pdev != bus->self || (dma_pdev->multifunction && + !pci_acs_enabled(dma_pdev, REQ_ACS_FLAGS))) + goto use_pdev; + + pci_dev_put(dma_pdev); + goto use_group; + } + + dma_pdev = get_isolation_root(pci_dev_get(to_pci_dev(dev))); +use_pdev: + ret = use_pdev_iommu_group(dma_pdev, dev); + pci_dev_put(dma_pdev); + return ret; +use_group: + return use_dev_data_iommu_group(dev_data->alias_data, dev); +} + +static int iommu_init_device(struct device *dev) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct iommu_dev_data *dev_data; + u16 alias; + int ret; + + if (dev->archdata.iommu) + return 0; + + dev_data = find_dev_data(get_device_id(dev)); + if (!dev_data) + return -ENOMEM; + + alias = amd_iommu_alias_table[dev_data->devid]; + if (alias != dev_data->devid) { + struct iommu_dev_data *alias_data; + + alias_data = find_dev_data(alias); + if (alias_data == NULL) { + pr_err("AMD-Vi: Warning: Unhandled device %s\n", + dev_name(dev)); + free_dev_data(dev_data); + return -ENOTSUPP; + } + dev_data->alias_data = alias_data; + } + ret = init_iommu_group(dev); if (ret) return ret; diff --git a/drivers/iommu/amd_iommu_init.c b/drivers/iommu/amd_iommu_init.c index 81837b0..faf10ba 100644 --- a/drivers/iommu/amd_iommu_init.c +++ b/drivers/iommu/amd_iommu_init.c @@ -975,6 +975,38 @@ static void __init free_iommu_all(void) } /* + * Family15h Model 10h-1fh erratum 746 (IOMMU Logging May Stall Translations) + * Workaround: + * BIOS should disable L2B micellaneous clock gating by setting + * L2_L2B_CK_GATE_CONTROL[CKGateL2BMiscDisable](D0F2xF4_x90[2]) = 1b + */ +static void __init amd_iommu_erratum_746_workaround(struct amd_iommu *iommu) +{ + u32 value; + + if ((boot_cpu_data.x86 != 0x15) || + (boot_cpu_data.x86_model < 0x10) || + (boot_cpu_data.x86_model > 0x1f)) + return; + + pci_write_config_dword(iommu->dev, 0xf0, 0x90); + pci_read_config_dword(iommu->dev, 0xf4, &value); + + if (value & BIT(2)) + return; + + /* Select NB indirect register 0x90 and enable writing */ + pci_write_config_dword(iommu->dev, 0xf0, 0x90 | (1 << 8)); + + pci_write_config_dword(iommu->dev, 0xf4, value | 0x4); + pr_info("AMD-Vi: Applying erratum 746 workaround for IOMMU at %s\n", + dev_name(&iommu->dev->dev)); + + /* Clear the enable writing bit */ + pci_write_config_dword(iommu->dev, 0xf0, 0x90); +} + +/* * This function clues the initialization function for one IOMMU * together and also allocates the command buffer and programs the * hardware. It does NOT enable the IOMMU. This is done afterwards. @@ -1172,6 +1204,8 @@ static int iommu_init_pci(struct amd_iommu *iommu) iommu->stored_l2[i] = iommu_read_l2(iommu, i); } + amd_iommu_erratum_746_workaround(iommu); + return pci_enable_device(iommu->dev); } diff --git a/drivers/iommu/amd_iommu_types.h b/drivers/iommu/amd_iommu_types.h index c9aa3d0..e38ab43 100644 --- a/drivers/iommu/amd_iommu_types.h +++ b/drivers/iommu/amd_iommu_types.h @@ -426,6 +426,7 @@ struct iommu_dev_data { struct iommu_dev_data *alias_data;/* The alias dev_data */ struct protection_domain *domain; /* Domain the device is bound to */ atomic_t bind; /* Domain attach reference count */ + struct iommu_group *group; /* IOMMU group for virtual aliases */ u16 devid; /* PCI Device ID */ bool iommu_v2; /* Device can make use of IOMMUv2 */ bool passthrough; /* Default for device is pt_domain */ diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index d4a4cd4..b9d0911 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -1827,10 +1827,17 @@ static int __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn, if (!pte) return -ENOMEM; /* It is large page*/ - if (largepage_lvl > 1) + if (largepage_lvl > 1) { pteval |= DMA_PTE_LARGE_PAGE; - else + /* Ensure that old small page tables are removed to make room + for superpage, if they exist. */ + dma_pte_clear_range(domain, iov_pfn, + iov_pfn + lvl_to_nr_pages(largepage_lvl) - 1); + dma_pte_free_pagetable(domain, iov_pfn, + iov_pfn + lvl_to_nr_pages(largepage_lvl) - 1); + } else { pteval &= ~(uint64_t)DMA_PTE_LARGE_PAGE; + } } /* We don't need lock here, nobody else @@ -2320,8 +2327,39 @@ static int domain_add_dev_info(struct dmar_domain *domain, return 0; } +static bool device_has_rmrr(struct pci_dev *dev) +{ + struct dmar_rmrr_unit *rmrr; + int i; + + for_each_rmrr_units(rmrr) { + for (i = 0; i < rmrr->devices_cnt; i++) { + /* + * Return TRUE if this RMRR contains the device that + * is passed in. + */ + if (rmrr->devices[i] == dev) + return true; + } + } + return false; +} + static int iommu_should_identity_map(struct pci_dev *pdev, int startup) { + + /* + * We want to prevent any device associated with an RMRR from + * getting placed into the SI Domain. This is done because + * problems exist when devices are moved in and out of domains + * and their respective RMRR info is lost. We exempt USB devices + * from this process due to their usage of RMRRs that are known + * to not be needed after BIOS hand-off to OS. + */ + if (device_has_rmrr(pdev) && + (pdev->class >> 8) != PCI_CLASS_SERIAL_USB) + return 0; + if ((iommu_identity_mapping & IDENTMAP_AZALIA) && IS_AZALIA(pdev)) return 1; @@ -4108,7 +4146,7 @@ static void swap_pci_ref(struct pci_dev **from, struct pci_dev *to) static int intel_iommu_add_device(struct device *dev) { struct pci_dev *pdev = to_pci_dev(dev); - struct pci_dev *bridge, *dma_pdev; + struct pci_dev *bridge, *dma_pdev = NULL; struct iommu_group *group; int ret; @@ -4122,7 +4160,7 @@ static int intel_iommu_add_device(struct device *dev) dma_pdev = pci_get_domain_bus_and_slot( pci_domain_nr(pdev->bus), bridge->subordinate->number, 0); - else + if (!dma_pdev) dma_pdev = pci_dev_get(bridge); } else dma_pdev = pci_dev_get(pdev); @@ -4196,7 +4234,7 @@ static struct iommu_ops intel_iommu_ops = { .pgsize_bitmap = INTEL_IOMMU_PGSIZES, }; -static void __devinit quirk_iommu_rwbf(struct pci_dev *dev) +static void quirk_iommu_rwbf(struct pci_dev *dev) { /* * Mobile 4 Series Chipset neglects to set RWBF capability, @@ -4224,7 +4262,7 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2a40, quirk_iommu_rwbf); #define GGC_MEMORY_SIZE_3M_VT (0xa << 8) #define GGC_MEMORY_SIZE_4M_VT (0xb << 8) -static void __devinit quirk_calpella_no_shadow_gtt(struct pci_dev *dev) +static void quirk_calpella_no_shadow_gtt(struct pci_dev *dev) { unsigned short ggc; diff --git a/drivers/iommu/omap-iommu-debug.c b/drivers/iommu/omap-iommu-debug.c index f55fc5d..d97fbe4 100644 --- a/drivers/iommu/omap-iommu-debug.c +++ b/drivers/iommu/omap-iommu-debug.c @@ -18,11 +18,11 @@ #include <linux/uaccess.h> #include <linux/platform_device.h> #include <linux/debugfs.h> +#include <linux/omap-iommu.h> +#include <linux/platform_data/iommu-omap.h> -#include <plat/iommu.h> -#include <plat/iovmm.h> - -#include <plat/iopgtable.h> +#include "omap-iopgtable.h" +#include "omap-iommu.h" #define MAXCOLUMN 100 /* for short messages */ diff --git a/drivers/iommu/omap-iommu.c b/drivers/iommu/omap-iommu.c index d0b1234..d33c980 100644 --- a/drivers/iommu/omap-iommu.c +++ b/drivers/iommu/omap-iommu.c @@ -16,17 +16,20 @@ #include <linux/slab.h> #include <linux/interrupt.h> #include <linux/ioport.h> -#include <linux/clk.h> #include <linux/platform_device.h> #include <linux/iommu.h> +#include <linux/omap-iommu.h> #include <linux/mutex.h> #include <linux/spinlock.h> +#include <linux/io.h> +#include <linux/pm_runtime.h> #include <asm/cacheflush.h> -#include <plat/iommu.h> +#include <linux/platform_data/iommu-omap.h> -#include <plat/iopgtable.h> +#include "omap-iopgtable.h" +#include "omap-iommu.h" #define for_each_iotlb_cr(obj, n, __i, cr) \ for (__i = 0; \ @@ -51,6 +54,21 @@ struct omap_iommu_domain { spinlock_t lock; }; +#define MMU_LOCK_BASE_SHIFT 10 +#define MMU_LOCK_BASE_MASK (0x1f << MMU_LOCK_BASE_SHIFT) +#define MMU_LOCK_BASE(x) \ + ((x & MMU_LOCK_BASE_MASK) >> MMU_LOCK_BASE_SHIFT) + +#define MMU_LOCK_VICT_SHIFT 4 +#define MMU_LOCK_VICT_MASK (0x1f << MMU_LOCK_VICT_SHIFT) +#define MMU_LOCK_VICT(x) \ + ((x & MMU_LOCK_VICT_MASK) >> MMU_LOCK_VICT_SHIFT) + +struct iotlb_lock { + short base; + short vict; +}; + /* accommodate the difference between omap1 and omap2/3 */ static const struct iommu_functions *arch_iommu; @@ -125,31 +143,44 @@ EXPORT_SYMBOL_GPL(omap_iommu_arch_version); static int iommu_enable(struct omap_iommu *obj) { int err; + struct platform_device *pdev = to_platform_device(obj->dev); + struct iommu_platform_data *pdata = pdev->dev.platform_data; - if (!obj) + if (!obj || !pdata) return -EINVAL; if (!arch_iommu) return -ENODEV; - clk_enable(obj->clk); + if (pdata->deassert_reset) { + err = pdata->deassert_reset(pdev, pdata->reset_name); + if (err) { + dev_err(obj->dev, "deassert_reset failed: %d\n", err); + return err; + } + } + + pm_runtime_get_sync(obj->dev); err = arch_iommu->enable(obj); - clk_disable(obj->clk); return err; } static void iommu_disable(struct omap_iommu *obj) { - if (!obj) - return; + struct platform_device *pdev = to_platform_device(obj->dev); + struct iommu_platform_data *pdata = pdev->dev.platform_data; - clk_enable(obj->clk); + if (!obj || !pdata) + return; arch_iommu->disable(obj); - clk_disable(obj->clk); + pm_runtime_put_sync(obj->dev); + + if (pdata->assert_reset) + pdata->assert_reset(pdev, pdata->reset_name); } /* @@ -272,7 +303,7 @@ static int load_iotlb_entry(struct omap_iommu *obj, struct iotlb_entry *e) if (!obj || !obj->nr_tlb_entries || !e) return -EINVAL; - clk_enable(obj->clk); + pm_runtime_get_sync(obj->dev); iotlb_lock_get(obj, &l); if (l.base == obj->nr_tlb_entries) { @@ -302,7 +333,7 @@ static int load_iotlb_entry(struct omap_iommu *obj, struct iotlb_entry *e) cr = iotlb_alloc_cr(obj, e); if (IS_ERR(cr)) { - clk_disable(obj->clk); + pm_runtime_put_sync(obj->dev); return PTR_ERR(cr); } @@ -316,7 +347,7 @@ static int load_iotlb_entry(struct omap_iommu *obj, struct iotlb_entry *e) l.vict = l.base; iotlb_lock_set(obj, &l); out: - clk_disable(obj->clk); + pm_runtime_put_sync(obj->dev); return err; } @@ -346,7 +377,7 @@ static void flush_iotlb_page(struct omap_iommu *obj, u32 da) int i; struct cr_regs cr; - clk_enable(obj->clk); + pm_runtime_get_sync(obj->dev); for_each_iotlb_cr(obj, obj->nr_tlb_entries, i, cr) { u32 start; @@ -365,7 +396,7 @@ static void flush_iotlb_page(struct omap_iommu *obj, u32 da) iommu_write_reg(obj, 1, MMU_FLUSH_ENTRY); } } - clk_disable(obj->clk); + pm_runtime_put_sync(obj->dev); if (i == obj->nr_tlb_entries) dev_dbg(obj->dev, "%s: no page for %08x\n", __func__, da); @@ -379,7 +410,7 @@ static void flush_iotlb_all(struct omap_iommu *obj) { struct iotlb_lock l; - clk_enable(obj->clk); + pm_runtime_get_sync(obj->dev); l.base = 0; l.vict = 0; @@ -387,7 +418,7 @@ static void flush_iotlb_all(struct omap_iommu *obj) iommu_write_reg(obj, 1, MMU_GFLUSH); - clk_disable(obj->clk); + pm_runtime_put_sync(obj->dev); } #if defined(CONFIG_OMAP_IOMMU_DEBUG) || defined(CONFIG_OMAP_IOMMU_DEBUG_MODULE) @@ -397,11 +428,11 @@ ssize_t omap_iommu_dump_ctx(struct omap_iommu *obj, char *buf, ssize_t bytes) if (!obj || !buf) return -EINVAL; - clk_enable(obj->clk); + pm_runtime_get_sync(obj->dev); bytes = arch_iommu->dump_ctx(obj, buf, bytes); - clk_disable(obj->clk); + pm_runtime_put_sync(obj->dev); return bytes; } @@ -415,7 +446,7 @@ __dump_tlb_entries(struct omap_iommu *obj, struct cr_regs *crs, int num) struct cr_regs tmp; struct cr_regs *p = crs; - clk_enable(obj->clk); + pm_runtime_get_sync(obj->dev); iotlb_lock_get(obj, &saved); for_each_iotlb_cr(obj, num, i, tmp) { @@ -425,7 +456,7 @@ __dump_tlb_entries(struct omap_iommu *obj, struct cr_regs *crs, int num) } iotlb_lock_set(obj, &saved); - clk_disable(obj->clk); + pm_runtime_put_sync(obj->dev); return p - crs; } @@ -789,9 +820,7 @@ static irqreturn_t iommu_fault_handler(int irq, void *data) if (!obj->refcount) return IRQ_NONE; - clk_enable(obj->clk); errs = iommu_report_fault(obj, &da); - clk_disable(obj->clk); if (errs == 0) return IRQ_HANDLED; @@ -905,7 +934,7 @@ static void omap_iommu_detach(struct omap_iommu *obj) /* * OMAP Device MMU(IOMMU) detection */ -static int __devinit omap_iommu_probe(struct platform_device *pdev) +static int omap_iommu_probe(struct platform_device *pdev) { int err = -ENODEV; int irq; @@ -913,17 +942,10 @@ static int __devinit omap_iommu_probe(struct platform_device *pdev) struct resource *res; struct iommu_platform_data *pdata = pdev->dev.platform_data; - if (pdev->num_resources != 2) - return -EINVAL; - obj = kzalloc(sizeof(*obj) + MMU_REG_SIZE, GFP_KERNEL); if (!obj) return -ENOMEM; - obj->clk = clk_get(&pdev->dev, pdata->clk_name); - if (IS_ERR(obj->clk)) - goto err_clk; - obj->nr_tlb_entries = pdata->nr_tlb_entries; obj->name = pdata->name; obj->dev = &pdev->dev; @@ -966,6 +988,9 @@ static int __devinit omap_iommu_probe(struct platform_device *pdev) goto err_irq; platform_set_drvdata(pdev, obj); + pm_runtime_irq_safe(obj->dev); + pm_runtime_enable(obj->dev); + dev_info(&pdev->dev, "%s registered\n", obj->name); return 0; @@ -974,13 +999,11 @@ err_irq: err_ioremap: release_mem_region(res->start, resource_size(res)); err_mem: - clk_put(obj->clk); -err_clk: kfree(obj); return err; } -static int __devexit omap_iommu_remove(struct platform_device *pdev) +static int omap_iommu_remove(struct platform_device *pdev) { int irq; struct resource *res; @@ -996,7 +1019,8 @@ static int __devexit omap_iommu_remove(struct platform_device *pdev) release_mem_region(res->start, resource_size(res)); iounmap(obj->regbase); - clk_put(obj->clk); + pm_runtime_disable(obj->dev); + dev_info(&pdev->dev, "%s removed\n", obj->name); kfree(obj); return 0; @@ -1004,7 +1028,7 @@ static int __devexit omap_iommu_remove(struct platform_device *pdev) static struct platform_driver omap_iommu_driver = { .probe = omap_iommu_probe, - .remove = __devexit_p(omap_iommu_remove), + .remove = omap_iommu_remove, .driver = { .name = "omap-iommu", }, @@ -1015,6 +1039,23 @@ static void iopte_cachep_ctor(void *iopte) clean_dcache_area(iopte, IOPTE_TABLE_SIZE); } +static u32 iotlb_init_entry(struct iotlb_entry *e, u32 da, u32 pa, + u32 flags) +{ + memset(e, 0, sizeof(*e)); + + e->da = da; + e->pa = pa; + e->valid = 1; + /* FIXME: add OMAP1 support */ + e->pgsz = flags & MMU_CAM_PGSZ_MASK; + e->endian = flags & MMU_RAM_ENDIAN_MASK; + e->elsz = flags & MMU_RAM_ELSZ_MASK; + e->mixed = flags & MMU_RAM_MIXED_MASK; + + return iopgsz_to_bytes(e->pgsz); +} + static int omap_iommu_map(struct iommu_domain *domain, unsigned long da, phys_addr_t pa, size_t bytes, int prot) { diff --git a/drivers/iommu/omap-iommu.h b/drivers/iommu/omap-iommu.h new file mode 100644 index 0000000..1200842 --- /dev/null +++ b/drivers/iommu/omap-iommu.h @@ -0,0 +1,225 @@ +/* + * omap iommu: main structures + * + * Copyright (C) 2008-2009 Nokia Corporation + * + * Written by Hiroshi DOYU <Hiroshi.DOYU@nokia.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#if defined(CONFIG_ARCH_OMAP1) +#error "iommu for this processor not implemented yet" +#endif + +struct iotlb_entry { + u32 da; + u32 pa; + u32 pgsz, prsvd, valid; + union { + u16 ap; + struct { + u32 endian, elsz, mixed; + }; + }; +}; + +struct omap_iommu { + const char *name; + struct module *owner; + void __iomem *regbase; + struct device *dev; + void *isr_priv; + struct iommu_domain *domain; + + unsigned int refcount; + spinlock_t iommu_lock; /* global for this whole object */ + + /* + * We don't change iopgd for a situation like pgd for a task, + * but share it globally for each iommu. + */ + u32 *iopgd; + spinlock_t page_table_lock; /* protect iopgd */ + + int nr_tlb_entries; + + struct list_head mmap; + struct mutex mmap_lock; /* protect mmap */ + + void *ctx; /* iommu context: registres saved area */ + u32 da_start; + u32 da_end; +}; + +struct cr_regs { + union { + struct { + u16 cam_l; + u16 cam_h; + }; + u32 cam; + }; + union { + struct { + u16 ram_l; + u16 ram_h; + }; + u32 ram; + }; +}; + +/* architecture specific functions */ +struct iommu_functions { + unsigned long version; + + int (*enable)(struct omap_iommu *obj); + void (*disable)(struct omap_iommu *obj); + void (*set_twl)(struct omap_iommu *obj, bool on); + u32 (*fault_isr)(struct omap_iommu *obj, u32 *ra); + + void (*tlb_read_cr)(struct omap_iommu *obj, struct cr_regs *cr); + void (*tlb_load_cr)(struct omap_iommu *obj, struct cr_regs *cr); + + struct cr_regs *(*alloc_cr)(struct omap_iommu *obj, + struct iotlb_entry *e); + int (*cr_valid)(struct cr_regs *cr); + u32 (*cr_to_virt)(struct cr_regs *cr); + void (*cr_to_e)(struct cr_regs *cr, struct iotlb_entry *e); + ssize_t (*dump_cr)(struct omap_iommu *obj, struct cr_regs *cr, + char *buf); + + u32 (*get_pte_attr)(struct iotlb_entry *e); + + void (*save_ctx)(struct omap_iommu *obj); + void (*restore_ctx)(struct omap_iommu *obj); + ssize_t (*dump_ctx)(struct omap_iommu *obj, char *buf, ssize_t len); +}; + +#ifdef CONFIG_IOMMU_API +/** + * dev_to_omap_iommu() - retrieves an omap iommu object from a user device + * @dev: iommu client device + */ +static inline struct omap_iommu *dev_to_omap_iommu(struct device *dev) +{ + struct omap_iommu_arch_data *arch_data = dev->archdata.iommu; + + return arch_data->iommu_dev; +} +#endif + +/* + * MMU Register offsets + */ +#define MMU_REVISION 0x00 +#define MMU_IRQSTATUS 0x18 +#define MMU_IRQENABLE 0x1c +#define MMU_WALKING_ST 0x40 +#define MMU_CNTL 0x44 +#define MMU_FAULT_AD 0x48 +#define MMU_TTB 0x4c +#define MMU_LOCK 0x50 +#define MMU_LD_TLB 0x54 +#define MMU_CAM 0x58 +#define MMU_RAM 0x5c +#define MMU_GFLUSH 0x60 +#define MMU_FLUSH_ENTRY 0x64 +#define MMU_READ_CAM 0x68 +#define MMU_READ_RAM 0x6c +#define MMU_EMU_FAULT_AD 0x70 + +#define MMU_REG_SIZE 256 + +/* + * MMU Register bit definitions + */ +#define MMU_CAM_VATAG_SHIFT 12 +#define MMU_CAM_VATAG_MASK \ + ((~0UL >> MMU_CAM_VATAG_SHIFT) << MMU_CAM_VATAG_SHIFT) +#define MMU_CAM_P (1 << 3) +#define MMU_CAM_V (1 << 2) +#define MMU_CAM_PGSZ_MASK 3 +#define MMU_CAM_PGSZ_1M (0 << 0) +#define MMU_CAM_PGSZ_64K (1 << 0) +#define MMU_CAM_PGSZ_4K (2 << 0) +#define MMU_CAM_PGSZ_16M (3 << 0) + +#define MMU_RAM_PADDR_SHIFT 12 +#define MMU_RAM_PADDR_MASK \ + ((~0UL >> MMU_RAM_PADDR_SHIFT) << MMU_RAM_PADDR_SHIFT) + +#define MMU_RAM_ENDIAN_MASK (1 << MMU_RAM_ENDIAN_SHIFT) +#define MMU_RAM_ENDIAN_BIG (1 << MMU_RAM_ENDIAN_SHIFT) + +#define MMU_RAM_ELSZ_MASK (3 << MMU_RAM_ELSZ_SHIFT) +#define MMU_RAM_ELSZ_8 (0 << MMU_RAM_ELSZ_SHIFT) +#define MMU_RAM_ELSZ_16 (1 << MMU_RAM_ELSZ_SHIFT) +#define MMU_RAM_ELSZ_32 (2 << MMU_RAM_ELSZ_SHIFT) +#define MMU_RAM_ELSZ_NONE (3 << MMU_RAM_ELSZ_SHIFT) +#define MMU_RAM_MIXED_SHIFT 6 +#define MMU_RAM_MIXED_MASK (1 << MMU_RAM_MIXED_SHIFT) +#define MMU_RAM_MIXED MMU_RAM_MIXED_MASK + +/* + * utilities for super page(16MB, 1MB, 64KB and 4KB) + */ + +#define iopgsz_max(bytes) \ + (((bytes) >= SZ_16M) ? SZ_16M : \ + ((bytes) >= SZ_1M) ? SZ_1M : \ + ((bytes) >= SZ_64K) ? SZ_64K : \ + ((bytes) >= SZ_4K) ? SZ_4K : 0) + +#define bytes_to_iopgsz(bytes) \ + (((bytes) == SZ_16M) ? MMU_CAM_PGSZ_16M : \ + ((bytes) == SZ_1M) ? MMU_CAM_PGSZ_1M : \ + ((bytes) == SZ_64K) ? MMU_CAM_PGSZ_64K : \ + ((bytes) == SZ_4K) ? MMU_CAM_PGSZ_4K : -1) + +#define iopgsz_to_bytes(iopgsz) \ + (((iopgsz) == MMU_CAM_PGSZ_16M) ? SZ_16M : \ + ((iopgsz) == MMU_CAM_PGSZ_1M) ? SZ_1M : \ + ((iopgsz) == MMU_CAM_PGSZ_64K) ? SZ_64K : \ + ((iopgsz) == MMU_CAM_PGSZ_4K) ? SZ_4K : 0) + +#define iopgsz_ok(bytes) (bytes_to_iopgsz(bytes) >= 0) + +/* + * global functions + */ +extern u32 omap_iommu_arch_version(void); + +extern void omap_iotlb_cr_to_e(struct cr_regs *cr, struct iotlb_entry *e); + +extern int +omap_iopgtable_store_entry(struct omap_iommu *obj, struct iotlb_entry *e); + +extern void omap_iommu_save_ctx(struct device *dev); +extern void omap_iommu_restore_ctx(struct device *dev); + +extern int omap_foreach_iommu_device(void *data, + int (*fn)(struct device *, void *)); + +extern int omap_install_iommu_arch(const struct iommu_functions *ops); +extern void omap_uninstall_iommu_arch(const struct iommu_functions *ops); + +extern ssize_t +omap_iommu_dump_ctx(struct omap_iommu *obj, char *buf, ssize_t len); +extern size_t +omap_dump_tlb_entries(struct omap_iommu *obj, char *buf, ssize_t len); + +/* + * register accessors + */ +static inline u32 iommu_read_reg(struct omap_iommu *obj, size_t offs) +{ + return __raw_readl(obj->regbase + offs); +} + +static inline void iommu_write_reg(struct omap_iommu *obj, u32 val, size_t offs) +{ + __raw_writel(val, obj->regbase + offs); +} diff --git a/drivers/iommu/omap-iommu2.c b/drivers/iommu/omap-iommu2.c new file mode 100644 index 0000000..d745094 --- /dev/null +++ b/drivers/iommu/omap-iommu2.c @@ -0,0 +1,334 @@ +/* + * omap iommu: omap2/3 architecture specific functions + * + * Copyright (C) 2008-2009 Nokia Corporation + * + * Written by Hiroshi DOYU <Hiroshi.DOYU@nokia.com>, + * Paul Mundt and Toshihiro Kobayashi + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/err.h> +#include <linux/device.h> +#include <linux/io.h> +#include <linux/jiffies.h> +#include <linux/module.h> +#include <linux/omap-iommu.h> +#include <linux/slab.h> +#include <linux/stringify.h> +#include <linux/platform_data/iommu-omap.h> + +#include "omap-iommu.h" + +/* + * omap2 architecture specific register bit definitions + */ +#define IOMMU_ARCH_VERSION 0x00000011 + +/* IRQSTATUS & IRQENABLE */ +#define MMU_IRQ_MULTIHITFAULT (1 << 4) +#define MMU_IRQ_TABLEWALKFAULT (1 << 3) +#define MMU_IRQ_EMUMISS (1 << 2) +#define MMU_IRQ_TRANSLATIONFAULT (1 << 1) +#define MMU_IRQ_TLBMISS (1 << 0) + +#define __MMU_IRQ_FAULT \ + (MMU_IRQ_MULTIHITFAULT | MMU_IRQ_EMUMISS | MMU_IRQ_TRANSLATIONFAULT) +#define MMU_IRQ_MASK \ + (__MMU_IRQ_FAULT | MMU_IRQ_TABLEWALKFAULT | MMU_IRQ_TLBMISS) +#define MMU_IRQ_TWL_MASK (__MMU_IRQ_FAULT | MMU_IRQ_TABLEWALKFAULT) +#define MMU_IRQ_TLB_MISS_MASK (__MMU_IRQ_FAULT | MMU_IRQ_TLBMISS) + +/* MMU_CNTL */ +#define MMU_CNTL_SHIFT 1 +#define MMU_CNTL_MASK (7 << MMU_CNTL_SHIFT) +#define MMU_CNTL_EML_TLB (1 << 3) +#define MMU_CNTL_TWL_EN (1 << 2) +#define MMU_CNTL_MMU_EN (1 << 1) + +#define get_cam_va_mask(pgsz) \ + (((pgsz) == MMU_CAM_PGSZ_16M) ? 0xff000000 : \ + ((pgsz) == MMU_CAM_PGSZ_1M) ? 0xfff00000 : \ + ((pgsz) == MMU_CAM_PGSZ_64K) ? 0xffff0000 : \ + ((pgsz) == MMU_CAM_PGSZ_4K) ? 0xfffff000 : 0) + +/* IOMMU errors */ +#define OMAP_IOMMU_ERR_TLB_MISS (1 << 0) +#define OMAP_IOMMU_ERR_TRANS_FAULT (1 << 1) +#define OMAP_IOMMU_ERR_EMU_MISS (1 << 2) +#define OMAP_IOMMU_ERR_TBLWALK_FAULT (1 << 3) +#define OMAP_IOMMU_ERR_MULTIHIT_FAULT (1 << 4) + +static void __iommu_set_twl(struct omap_iommu *obj, bool on) +{ + u32 l = iommu_read_reg(obj, MMU_CNTL); + + if (on) + iommu_write_reg(obj, MMU_IRQ_TWL_MASK, MMU_IRQENABLE); + else + iommu_write_reg(obj, MMU_IRQ_TLB_MISS_MASK, MMU_IRQENABLE); + + l &= ~MMU_CNTL_MASK; + if (on) + l |= (MMU_CNTL_MMU_EN | MMU_CNTL_TWL_EN); + else + l |= (MMU_CNTL_MMU_EN); + + iommu_write_reg(obj, l, MMU_CNTL); +} + + +static int omap2_iommu_enable(struct omap_iommu *obj) +{ + u32 l, pa; + + if (!obj->iopgd || !IS_ALIGNED((u32)obj->iopgd, SZ_16K)) + return -EINVAL; + + pa = virt_to_phys(obj->iopgd); + if (!IS_ALIGNED(pa, SZ_16K)) + return -EINVAL; + + l = iommu_read_reg(obj, MMU_REVISION); + dev_info(obj->dev, "%s: version %d.%d\n", obj->name, + (l >> 4) & 0xf, l & 0xf); + + iommu_write_reg(obj, pa, MMU_TTB); + + __iommu_set_twl(obj, true); + + return 0; +} + +static void omap2_iommu_disable(struct omap_iommu *obj) +{ + u32 l = iommu_read_reg(obj, MMU_CNTL); + + l &= ~MMU_CNTL_MASK; + iommu_write_reg(obj, l, MMU_CNTL); + + dev_dbg(obj->dev, "%s is shutting down\n", obj->name); +} + +static void omap2_iommu_set_twl(struct omap_iommu *obj, bool on) +{ + __iommu_set_twl(obj, false); +} + +static u32 omap2_iommu_fault_isr(struct omap_iommu *obj, u32 *ra) +{ + u32 stat, da; + u32 errs = 0; + + stat = iommu_read_reg(obj, MMU_IRQSTATUS); + stat &= MMU_IRQ_MASK; + if (!stat) { + *ra = 0; + return 0; + } + + da = iommu_read_reg(obj, MMU_FAULT_AD); + *ra = da; + + if (stat & MMU_IRQ_TLBMISS) + errs |= OMAP_IOMMU_ERR_TLB_MISS; + if (stat & MMU_IRQ_TRANSLATIONFAULT) + errs |= OMAP_IOMMU_ERR_TRANS_FAULT; + if (stat & MMU_IRQ_EMUMISS) + errs |= OMAP_IOMMU_ERR_EMU_MISS; + if (stat & MMU_IRQ_TABLEWALKFAULT) + errs |= OMAP_IOMMU_ERR_TBLWALK_FAULT; + if (stat & MMU_IRQ_MULTIHITFAULT) + errs |= OMAP_IOMMU_ERR_MULTIHIT_FAULT; + iommu_write_reg(obj, stat, MMU_IRQSTATUS); + + return errs; +} + +static void omap2_tlb_read_cr(struct omap_iommu *obj, struct cr_regs *cr) +{ + cr->cam = iommu_read_reg(obj, MMU_READ_CAM); + cr->ram = iommu_read_reg(obj, MMU_READ_RAM); +} + +static void omap2_tlb_load_cr(struct omap_iommu *obj, struct cr_regs *cr) +{ + iommu_write_reg(obj, cr->cam | MMU_CAM_V, MMU_CAM); + iommu_write_reg(obj, cr->ram, MMU_RAM); +} + +static u32 omap2_cr_to_virt(struct cr_regs *cr) +{ + u32 page_size = cr->cam & MMU_CAM_PGSZ_MASK; + u32 mask = get_cam_va_mask(cr->cam & page_size); + + return cr->cam & mask; +} + +static struct cr_regs *omap2_alloc_cr(struct omap_iommu *obj, + struct iotlb_entry *e) +{ + struct cr_regs *cr; + + if (e->da & ~(get_cam_va_mask(e->pgsz))) { + dev_err(obj->dev, "%s:\twrong alignment: %08x\n", __func__, + e->da); + return ERR_PTR(-EINVAL); + } + + cr = kmalloc(sizeof(*cr), GFP_KERNEL); + if (!cr) + return ERR_PTR(-ENOMEM); + + cr->cam = (e->da & MMU_CAM_VATAG_MASK) | e->prsvd | e->pgsz | e->valid; + cr->ram = e->pa | e->endian | e->elsz | e->mixed; + + return cr; +} + +static inline int omap2_cr_valid(struct cr_regs *cr) +{ + return cr->cam & MMU_CAM_V; +} + +static u32 omap2_get_pte_attr(struct iotlb_entry *e) +{ + u32 attr; + + attr = e->mixed << 5; + attr |= e->endian; + attr |= e->elsz >> 3; + attr <<= (((e->pgsz == MMU_CAM_PGSZ_4K) || + (e->pgsz == MMU_CAM_PGSZ_64K)) ? 0 : 6); + return attr; +} + +static ssize_t +omap2_dump_cr(struct omap_iommu *obj, struct cr_regs *cr, char *buf) +{ + char *p = buf; + + /* FIXME: Need more detail analysis of cam/ram */ + p += sprintf(p, "%08x %08x %01x\n", cr->cam, cr->ram, + (cr->cam & MMU_CAM_P) ? 1 : 0); + + return p - buf; +} + +#define pr_reg(name) \ + do { \ + ssize_t bytes; \ + const char *str = "%20s: %08x\n"; \ + const int maxcol = 32; \ + bytes = snprintf(p, maxcol, str, __stringify(name), \ + iommu_read_reg(obj, MMU_##name)); \ + p += bytes; \ + len -= bytes; \ + if (len < maxcol) \ + goto out; \ + } while (0) + +static ssize_t +omap2_iommu_dump_ctx(struct omap_iommu *obj, char *buf, ssize_t len) +{ + char *p = buf; + + pr_reg(REVISION); + pr_reg(IRQSTATUS); + pr_reg(IRQENABLE); + pr_reg(WALKING_ST); + pr_reg(CNTL); + pr_reg(FAULT_AD); + pr_reg(TTB); + pr_reg(LOCK); + pr_reg(LD_TLB); + pr_reg(CAM); + pr_reg(RAM); + pr_reg(GFLUSH); + pr_reg(FLUSH_ENTRY); + pr_reg(READ_CAM); + pr_reg(READ_RAM); + pr_reg(EMU_FAULT_AD); +out: + return p - buf; +} + +static void omap2_iommu_save_ctx(struct omap_iommu *obj) +{ + int i; + u32 *p = obj->ctx; + + for (i = 0; i < (MMU_REG_SIZE / sizeof(u32)); i++) { + p[i] = iommu_read_reg(obj, i * sizeof(u32)); + dev_dbg(obj->dev, "%s\t[%02d] %08x\n", __func__, i, p[i]); + } + + BUG_ON(p[0] != IOMMU_ARCH_VERSION); +} + +static void omap2_iommu_restore_ctx(struct omap_iommu *obj) +{ + int i; + u32 *p = obj->ctx; + + for (i = 0; i < (MMU_REG_SIZE / sizeof(u32)); i++) { + iommu_write_reg(obj, p[i], i * sizeof(u32)); + dev_dbg(obj->dev, "%s\t[%02d] %08x\n", __func__, i, p[i]); + } + + BUG_ON(p[0] != IOMMU_ARCH_VERSION); +} + +static void omap2_cr_to_e(struct cr_regs *cr, struct iotlb_entry *e) +{ + e->da = cr->cam & MMU_CAM_VATAG_MASK; + e->pa = cr->ram & MMU_RAM_PADDR_MASK; + e->valid = cr->cam & MMU_CAM_V; + e->pgsz = cr->cam & MMU_CAM_PGSZ_MASK; + e->endian = cr->ram & MMU_RAM_ENDIAN_MASK; + e->elsz = cr->ram & MMU_RAM_ELSZ_MASK; + e->mixed = cr->ram & MMU_RAM_MIXED; +} + +static const struct iommu_functions omap2_iommu_ops = { + .version = IOMMU_ARCH_VERSION, + + .enable = omap2_iommu_enable, + .disable = omap2_iommu_disable, + .set_twl = omap2_iommu_set_twl, + .fault_isr = omap2_iommu_fault_isr, + + .tlb_read_cr = omap2_tlb_read_cr, + .tlb_load_cr = omap2_tlb_load_cr, + + .cr_to_e = omap2_cr_to_e, + .cr_to_virt = omap2_cr_to_virt, + .alloc_cr = omap2_alloc_cr, + .cr_valid = omap2_cr_valid, + .dump_cr = omap2_dump_cr, + + .get_pte_attr = omap2_get_pte_attr, + + .save_ctx = omap2_iommu_save_ctx, + .restore_ctx = omap2_iommu_restore_ctx, + .dump_ctx = omap2_iommu_dump_ctx, +}; + +static int __init omap2_iommu_init(void) +{ + return omap_install_iommu_arch(&omap2_iommu_ops); +} +module_init(omap2_iommu_init); + +static void __exit omap2_iommu_exit(void) +{ + omap_uninstall_iommu_arch(&omap2_iommu_ops); +} +module_exit(omap2_iommu_exit); + +MODULE_AUTHOR("Hiroshi DOYU, Paul Mundt and Toshihiro Kobayashi"); +MODULE_DESCRIPTION("omap iommu: omap2/3 architecture specific functions"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/iommu/omap-iopgtable.h b/drivers/iommu/omap-iopgtable.h new file mode 100644 index 0000000..cd4ae9e --- /dev/null +++ b/drivers/iommu/omap-iopgtable.h @@ -0,0 +1,98 @@ +/* + * omap iommu: pagetable definitions + * + * Copyright (C) 2008-2010 Nokia Corporation + * + * Written by Hiroshi DOYU <Hiroshi.DOYU@nokia.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +/* + * "L2 table" address mask and size definitions. + */ +#define IOPGD_SHIFT 20 +#define IOPGD_SIZE (1UL << IOPGD_SHIFT) +#define IOPGD_MASK (~(IOPGD_SIZE - 1)) + +/* + * "section" address mask and size definitions. + */ +#define IOSECTION_SHIFT 20 +#define IOSECTION_SIZE (1UL << IOSECTION_SHIFT) +#define IOSECTION_MASK (~(IOSECTION_SIZE - 1)) + +/* + * "supersection" address mask and size definitions. + */ +#define IOSUPER_SHIFT 24 +#define IOSUPER_SIZE (1UL << IOSUPER_SHIFT) +#define IOSUPER_MASK (~(IOSUPER_SIZE - 1)) + +#define PTRS_PER_IOPGD (1UL << (32 - IOPGD_SHIFT)) +#define IOPGD_TABLE_SIZE (PTRS_PER_IOPGD * sizeof(u32)) + +/* + * "small page" address mask and size definitions. + */ +#define IOPTE_SHIFT 12 +#define IOPTE_SIZE (1UL << IOPTE_SHIFT) +#define IOPTE_MASK (~(IOPTE_SIZE - 1)) + +/* + * "large page" address mask and size definitions. + */ +#define IOLARGE_SHIFT 16 +#define IOLARGE_SIZE (1UL << IOLARGE_SHIFT) +#define IOLARGE_MASK (~(IOLARGE_SIZE - 1)) + +#define PTRS_PER_IOPTE (1UL << (IOPGD_SHIFT - IOPTE_SHIFT)) +#define IOPTE_TABLE_SIZE (PTRS_PER_IOPTE * sizeof(u32)) + +#define IOPAGE_MASK IOPTE_MASK + +/** + * omap_iommu_translate() - va to pa translation + * @d: omap iommu descriptor + * @va: virtual address + * @mask: omap iommu descriptor mask + * + * va to pa translation + */ +static inline phys_addr_t omap_iommu_translate(u32 d, u32 va, u32 mask) +{ + return (d & mask) | (va & (~mask)); +} + +/* + * some descriptor attributes. + */ +#define IOPGD_TABLE (1 << 0) +#define IOPGD_SECTION (2 << 0) +#define IOPGD_SUPER (1 << 18 | 2 << 0) + +#define iopgd_is_table(x) (((x) & 3) == IOPGD_TABLE) +#define iopgd_is_section(x) (((x) & (1 << 18 | 3)) == IOPGD_SECTION) +#define iopgd_is_super(x) (((x) & (1 << 18 | 3)) == IOPGD_SUPER) + +#define IOPTE_SMALL (2 << 0) +#define IOPTE_LARGE (1 << 0) + +#define iopte_is_small(x) (((x) & 2) == IOPTE_SMALL) +#define iopte_is_large(x) (((x) & 3) == IOPTE_LARGE) + +/* to find an entry in a page-table-directory */ +#define iopgd_index(da) (((da) >> IOPGD_SHIFT) & (PTRS_PER_IOPGD - 1)) +#define iopgd_offset(obj, da) ((obj)->iopgd + iopgd_index(da)) + +#define iopgd_page_paddr(iopgd) (*iopgd & ~((1 << 10) - 1)) +#define iopgd_page_vaddr(iopgd) ((u32 *)phys_to_virt(iopgd_page_paddr(iopgd))) + +/* to find an entry in the second-level page table. */ +#define iopte_index(da) (((da) >> IOPTE_SHIFT) & (PTRS_PER_IOPTE - 1)) +#define iopte_offset(iopgd, da) (iopgd_page_vaddr(iopgd) + iopte_index(da)) + +#define to_iommu(dev) \ + (struct omap_iommu *)platform_get_drvdata(to_platform_device(dev)) diff --git a/drivers/iommu/omap-iovmm.c b/drivers/iommu/omap-iovmm.c index 2e10c3e..46d87569 100644 --- a/drivers/iommu/omap-iovmm.c +++ b/drivers/iommu/omap-iovmm.c @@ -17,14 +17,58 @@ #include <linux/device.h> #include <linux/scatterlist.h> #include <linux/iommu.h> +#include <linux/omap-iommu.h> +#include <linux/platform_data/iommu-omap.h> #include <asm/cacheflush.h> #include <asm/mach/map.h> -#include <plat/iommu.h> -#include <plat/iovmm.h> +#include "omap-iopgtable.h" +#include "omap-iommu.h" -#include <plat/iopgtable.h> +/* + * IOVMF_FLAGS: attribute for iommu virtual memory area(iovma) + * + * lower 16 bit is used for h/w and upper 16 bit is for s/w. + */ +#define IOVMF_SW_SHIFT 16 + +/* + * iovma: h/w flags derived from cam and ram attribute + */ +#define IOVMF_CAM_MASK (~((1 << 10) - 1)) +#define IOVMF_RAM_MASK (~IOVMF_CAM_MASK) + +#define IOVMF_PGSZ_MASK (3 << 0) +#define IOVMF_PGSZ_1M MMU_CAM_PGSZ_1M +#define IOVMF_PGSZ_64K MMU_CAM_PGSZ_64K +#define IOVMF_PGSZ_4K MMU_CAM_PGSZ_4K +#define IOVMF_PGSZ_16M MMU_CAM_PGSZ_16M + +#define IOVMF_ENDIAN_MASK (1 << 9) +#define IOVMF_ENDIAN_BIG MMU_RAM_ENDIAN_BIG + +#define IOVMF_ELSZ_MASK (3 << 7) +#define IOVMF_ELSZ_16 MMU_RAM_ELSZ_16 +#define IOVMF_ELSZ_32 MMU_RAM_ELSZ_32 +#define IOVMF_ELSZ_NONE MMU_RAM_ELSZ_NONE + +#define IOVMF_MIXED_MASK (1 << 6) +#define IOVMF_MIXED MMU_RAM_MIXED + +/* + * iovma: s/w flags, used for mapping and umapping internally. + */ +#define IOVMF_MMIO (1 << IOVMF_SW_SHIFT) +#define IOVMF_ALLOC (2 << IOVMF_SW_SHIFT) +#define IOVMF_ALLOC_MASK (3 << IOVMF_SW_SHIFT) + +/* "superpages" is supported just with physically linear pages */ +#define IOVMF_DISCONT (1 << (2 + IOVMF_SW_SHIFT)) +#define IOVMF_LINEAR (2 << (2 + IOVMF_SW_SHIFT)) +#define IOVMF_LINEAR_MASK (3 << (2 + IOVMF_SW_SHIFT)) + +#define IOVMF_DA_FIXED (1 << (4 + IOVMF_SW_SHIFT)) static struct kmem_cache *iovm_area_cachep; diff --git a/drivers/iommu/tegra-gart.c b/drivers/iommu/tegra-gart.c index c16e8fc..8219f1d 100644 --- a/drivers/iommu/tegra-gart.c +++ b/drivers/iommu/tegra-gart.c @@ -398,6 +398,7 @@ static int tegra_gart_probe(struct platform_device *pdev) do_gart_setup(gart, NULL); gart_handle = gart; + bus_set_iommu(&platform_bus_type, &gart_iommu_ops); return 0; fail: @@ -430,7 +431,7 @@ const struct dev_pm_ops tegra_gart_pm_ops = { }; #ifdef CONFIG_OF -static struct of_device_id tegra_gart_of_match[] __devinitdata = { +static struct of_device_id tegra_gart_of_match[] = { { .compatible = "nvidia,tegra20-gart", }, { }, }; @@ -448,9 +449,8 @@ static struct platform_driver tegra_gart_driver = { }, }; -static int __devinit tegra_gart_init(void) +static int tegra_gart_init(void) { - bus_set_iommu(&platform_bus_type, &gart_iommu_ops); return platform_driver_register(&tegra_gart_driver); } diff --git a/drivers/iommu/tegra-smmu.c b/drivers/iommu/tegra-smmu.c index a649f14..fc17889 100644 --- a/drivers/iommu/tegra-smmu.c +++ b/drivers/iommu/tegra-smmu.c @@ -34,13 +34,11 @@ #include <linux/of_iommu.h> #include <linux/debugfs.h> #include <linux/seq_file.h> +#include <linux/tegra-ahb.h> #include <asm/page.h> #include <asm/cacheflush.h> -#include <mach/iomap.h> -#include <mach/tegra-ahb.h> - enum smmu_hwgrp { HWGRP_AFI, HWGRP_AVPC, @@ -696,10 +694,8 @@ static void __smmu_iommu_unmap(struct smmu_as *as, dma_addr_t iova) *pte = _PTE_VACANT(iova); FLUSH_CPU_DCACHE(pte, page, sizeof(*pte)); flush_ptc_and_tlb(as->smmu, as, iova, pte, page, 0); - if (!--(*count)) { + if (!--(*count)) free_ptbl(as, iova); - smmu_flush_regs(as->smmu, 0); - } } static void __smmu_iommu_map_pfn(struct smmu_as *as, dma_addr_t iova, @@ -1054,6 +1050,7 @@ static int smmu_debugfs_stats_show(struct seq_file *s, void *v) stats[i], val, offs); } seq_printf(s, "\n"); + dput(dent); return 0; } @@ -1233,6 +1230,7 @@ static int tegra_smmu_probe(struct platform_device *pdev) smmu_debugfs_create(smmu); smmu_handle = smmu; + bus_set_iommu(&platform_bus_type, &smmu_iommu_ops); return 0; } @@ -1257,7 +1255,7 @@ const struct dev_pm_ops tegra_smmu_pm_ops = { }; #ifdef CONFIG_OF -static struct of_device_id tegra_smmu_of_match[] __devinitdata = { +static struct of_device_id tegra_smmu_of_match[] = { { .compatible = "nvidia,tegra30-smmu", }, { }, }; @@ -1275,9 +1273,8 @@ static struct platform_driver tegra_smmu_driver = { }, }; -static int __devinit tegra_smmu_init(void) +static int tegra_smmu_init(void) { - bus_set_iommu(&platform_bus_type, &smmu_iommu_ops); return platform_driver_register(&tegra_smmu_driver); } |