diff options
Diffstat (limited to 'drivers/pci')
-rw-r--r-- | drivers/pci/Kconfig | 2 | ||||
-rw-r--r-- | drivers/pci/access.c | 41 | ||||
-rw-r--r-- | drivers/pci/dmar.c | 82 | ||||
-rw-r--r-- | drivers/pci/hotplug/acpiphp_ibm.c | 5 | ||||
-rw-r--r-- | drivers/pci/hotplug/cpqphp_core.c | 5 | ||||
-rw-r--r-- | drivers/pci/hotplug/pciehp_pci.c | 17 | ||||
-rw-r--r-- | drivers/pci/intel-iommu.c | 129 | ||||
-rw-r--r-- | drivers/pci/intr_remapping.c | 6 | ||||
-rw-r--r-- | drivers/pci/pci-sysfs.c | 89 | ||||
-rw-r--r-- | drivers/pci/pci.c | 6 | ||||
-rw-r--r-- | drivers/pci/pci.h | 2 | ||||
-rw-r--r-- | drivers/pci/pcie/aer/aer_inject.c | 2 | ||||
-rw-r--r-- | drivers/pci/pcie/aer/aerdrv.c | 179 | ||||
-rw-r--r-- | drivers/pci/pcie/aer/aerdrv.h | 6 | ||||
-rw-r--r-- | drivers/pci/pcie/aer/aerdrv_core.c | 558 | ||||
-rw-r--r-- | drivers/pci/quirks.c | 21 | ||||
-rw-r--r-- | drivers/pci/slot.c | 48 |
17 files changed, 645 insertions, 553 deletions
diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig index 7858a117e80..34ef70d562b 100644 --- a/drivers/pci/Kconfig +++ b/drivers/pci/Kconfig @@ -19,7 +19,7 @@ config PCI_MSI by using the 'pci=nomsi' option. This disables MSI for the entire system. - If you don't know what to do here, say N. + If you don't know what to do here, say Y. config PCI_DEBUG bool "PCI Debugging" diff --git a/drivers/pci/access.c b/drivers/pci/access.c index 2f646fe1260..531bc697d80 100644 --- a/drivers/pci/access.c +++ b/drivers/pci/access.c @@ -13,7 +13,7 @@ * configuration space. */ -static DEFINE_SPINLOCK(pci_lock); +static DEFINE_RAW_SPINLOCK(pci_lock); /* * Wrappers for all PCI configuration access functions. They just check @@ -33,10 +33,10 @@ int pci_bus_read_config_##size \ unsigned long flags; \ u32 data = 0; \ if (PCI_##size##_BAD) return PCIBIOS_BAD_REGISTER_NUMBER; \ - spin_lock_irqsave(&pci_lock, flags); \ + raw_spin_lock_irqsave(&pci_lock, flags); \ res = bus->ops->read(bus, devfn, pos, len, &data); \ *value = (type)data; \ - spin_unlock_irqrestore(&pci_lock, flags); \ + raw_spin_unlock_irqrestore(&pci_lock, flags); \ return res; \ } @@ -47,9 +47,9 @@ int pci_bus_write_config_##size \ int res; \ unsigned long flags; \ if (PCI_##size##_BAD) return PCIBIOS_BAD_REGISTER_NUMBER; \ - spin_lock_irqsave(&pci_lock, flags); \ + raw_spin_lock_irqsave(&pci_lock, flags); \ res = bus->ops->write(bus, devfn, pos, len, value); \ - spin_unlock_irqrestore(&pci_lock, flags); \ + raw_spin_unlock_irqrestore(&pci_lock, flags); \ return res; \ } @@ -79,10 +79,10 @@ struct pci_ops *pci_bus_set_ops(struct pci_bus *bus, struct pci_ops *ops) struct pci_ops *old_ops; unsigned long flags; - spin_lock_irqsave(&pci_lock, flags); + raw_spin_lock_irqsave(&pci_lock, flags); old_ops = bus->ops; bus->ops = ops; - spin_unlock_irqrestore(&pci_lock, flags); + raw_spin_unlock_irqrestore(&pci_lock, flags); return old_ops; } EXPORT_SYMBOL(pci_bus_set_ops); @@ -136,9 +136,9 @@ static noinline void pci_wait_ucfg(struct pci_dev *dev) __add_wait_queue(&pci_ucfg_wait, &wait); do { set_current_state(TASK_UNINTERRUPTIBLE); - spin_unlock_irq(&pci_lock); + raw_spin_unlock_irq(&pci_lock); schedule(); - spin_lock_irq(&pci_lock); + raw_spin_lock_irq(&pci_lock); } while (dev->block_ucfg_access); __remove_wait_queue(&pci_ucfg_wait, &wait); } @@ -150,11 +150,11 @@ int pci_user_read_config_##size \ int ret = 0; \ u32 data = -1; \ if (PCI_##size##_BAD) return PCIBIOS_BAD_REGISTER_NUMBER; \ - spin_lock_irq(&pci_lock); \ + raw_spin_lock_irq(&pci_lock); \ if (unlikely(dev->block_ucfg_access)) pci_wait_ucfg(dev); \ ret = dev->bus->ops->read(dev->bus, dev->devfn, \ pos, sizeof(type), &data); \ - spin_unlock_irq(&pci_lock); \ + raw_spin_unlock_irq(&pci_lock); \ *val = (type)data; \ return ret; \ } @@ -165,11 +165,11 @@ int pci_user_write_config_##size \ { \ int ret = -EIO; \ if (PCI_##size##_BAD) return PCIBIOS_BAD_REGISTER_NUMBER; \ - spin_lock_irq(&pci_lock); \ + raw_spin_lock_irq(&pci_lock); \ if (unlikely(dev->block_ucfg_access)) pci_wait_ucfg(dev); \ ret = dev->bus->ops->write(dev->bus, dev->devfn, \ pos, sizeof(type), val); \ - spin_unlock_irq(&pci_lock); \ + raw_spin_unlock_irq(&pci_lock); \ return ret; \ } @@ -220,8 +220,13 @@ static int pci_vpd_pci22_wait(struct pci_dev *dev) return 0; } - if (time_after(jiffies, timeout)) + if (time_after(jiffies, timeout)) { + dev_printk(KERN_DEBUG, &dev->dev, + "vpd r/w failed. This is likely a firmware " + "bug on this device. Contact the card " + "vendor for a firmware update."); return -ETIMEDOUT; + } if (fatal_signal_pending(current)) return -EINTR; if (!cond_resched()) @@ -396,10 +401,10 @@ void pci_block_user_cfg_access(struct pci_dev *dev) unsigned long flags; int was_blocked; - spin_lock_irqsave(&pci_lock, flags); + raw_spin_lock_irqsave(&pci_lock, flags); was_blocked = dev->block_ucfg_access; dev->block_ucfg_access = 1; - spin_unlock_irqrestore(&pci_lock, flags); + raw_spin_unlock_irqrestore(&pci_lock, flags); /* If we BUG() inside the pci_lock, we're guaranteed to hose * the machine */ @@ -417,7 +422,7 @@ void pci_unblock_user_cfg_access(struct pci_dev *dev) { unsigned long flags; - spin_lock_irqsave(&pci_lock, flags); + raw_spin_lock_irqsave(&pci_lock, flags); /* This indicates a problem in the caller, but we don't need * to kill them, unlike a double-block above. */ @@ -425,6 +430,6 @@ void pci_unblock_user_cfg_access(struct pci_dev *dev) dev->block_ucfg_access = 0; wake_up_all(&pci_ucfg_wait); - spin_unlock_irqrestore(&pci_lock, flags); + raw_spin_unlock_irqrestore(&pci_lock, flags); } EXPORT_SYMBOL_GPL(pci_unblock_user_cfg_access); diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c index 33ead97f0c4..0a19708074c 100644 --- a/drivers/pci/dmar.c +++ b/drivers/pci/dmar.c @@ -131,9 +131,10 @@ static int __init dmar_parse_dev_scope(void *start, void *end, int *cnt, if (scope->entry_type == ACPI_DMAR_SCOPE_TYPE_ENDPOINT || scope->entry_type == ACPI_DMAR_SCOPE_TYPE_BRIDGE) (*cnt)++; - else + else if (scope->entry_type != ACPI_DMAR_SCOPE_TYPE_IOAPIC) { printk(KERN_WARNING PREFIX - "Unsupported device scope\n"); + "Unsupported device scope\n"); + } start += scope->length; } if (*cnt == 0) @@ -309,6 +310,8 @@ int dmar_find_matched_atsr_unit(struct pci_dev *dev) struct acpi_dmar_atsr *atsr; struct dmar_atsr_unit *atsru; + dev = pci_physfn(dev); + list_for_each_entry(atsru, &dmar_atsr_units, list) { atsr = container_of(atsru->hdr, struct acpi_dmar_atsr, header); if (atsr->segment == pci_domain_nr(dev->bus)) @@ -358,12 +361,14 @@ dmar_parse_one_rhsa(struct acpi_dmar_header *header) return 0; } } - WARN(1, "Your BIOS is broken; RHSA refers to non-existent DMAR unit at %llx\n" - "BIOS vendor: %s; Ver: %s; Product Version: %s\n", - drhd->reg_base_addr, - dmi_get_system_info(DMI_BIOS_VENDOR), - dmi_get_system_info(DMI_BIOS_VERSION), - dmi_get_system_info(DMI_PRODUCT_VERSION)); + WARN_TAINT( + 1, TAINT_FIRMWARE_WORKAROUND, + "Your BIOS is broken; RHSA refers to non-existent DMAR unit at %llx\n" + "BIOS vendor: %s; Ver: %s; Product Version: %s\n", + drhd->reg_base_addr, + dmi_get_system_info(DMI_BIOS_VENDOR), + dmi_get_system_info(DMI_BIOS_VERSION), + dmi_get_system_info(DMI_PRODUCT_VERSION)); return 0; } @@ -507,7 +512,7 @@ parse_dmar_table(void) return ret; } -int dmar_pci_device_match(struct pci_dev *devices[], int cnt, +static int dmar_pci_device_match(struct pci_dev *devices[], int cnt, struct pci_dev *dev) { int index; @@ -530,6 +535,8 @@ dmar_find_matched_drhd_unit(struct pci_dev *dev) struct dmar_drhd_unit *dmaru = NULL; struct acpi_dmar_hardware_unit *drhd; + dev = pci_physfn(dev); + list_for_each_entry(dmaru, &dmar_drhd_units, list) { drhd = container_of(dmaru->hdr, struct acpi_dmar_hardware_unit, @@ -614,7 +621,17 @@ int __init dmar_table_init(void) return 0; } -static int bios_warned; +static void warn_invalid_dmar(u64 addr, const char *message) +{ + WARN_TAINT_ONCE( + 1, TAINT_FIRMWARE_WORKAROUND, + "Your BIOS is broken; DMAR reported at address %llx%s!\n" + "BIOS vendor: %s; Ver: %s; Product Version: %s\n", + addr, message, + dmi_get_system_info(DMI_BIOS_VENDOR), + dmi_get_system_info(DMI_BIOS_VERSION), + dmi_get_system_info(DMI_PRODUCT_VERSION)); +} int __init check_zero_address(void) { @@ -640,13 +657,7 @@ int __init check_zero_address(void) drhd = (void *)entry_header; if (!drhd->address) { - /* Promote an attitude of violence to a BIOS engineer today */ - WARN(1, "Your BIOS is broken; DMAR reported at address zero!\n" - "BIOS vendor: %s; Ver: %s; Product Version: %s\n", - dmi_get_system_info(DMI_BIOS_VENDOR), - dmi_get_system_info(DMI_BIOS_VERSION), - dmi_get_system_info(DMI_PRODUCT_VERSION)); - bios_warned = 1; + warn_invalid_dmar(0, ""); goto failed; } @@ -659,14 +670,8 @@ int __init check_zero_address(void) ecap = dmar_readq(addr + DMAR_ECAP_REG); early_iounmap(addr, VTD_PAGE_SIZE); if (cap == (uint64_t)-1 && ecap == (uint64_t)-1) { - /* Promote an attitude of violence to a BIOS engineer today */ - WARN(1, "Your BIOS is broken; DMAR reported at address %llx returns all ones!\n" - "BIOS vendor: %s; Ver: %s; Product Version: %s\n", - drhd->address, - dmi_get_system_info(DMI_BIOS_VENDOR), - dmi_get_system_info(DMI_BIOS_VERSION), - dmi_get_system_info(DMI_PRODUCT_VERSION)); - bios_warned = 1; + warn_invalid_dmar(drhd->address, + " returns all ones"); goto failed; } } @@ -731,14 +736,7 @@ int alloc_iommu(struct dmar_drhd_unit *drhd) int msagaw = 0; if (!drhd->reg_base_addr) { - if (!bios_warned) { - WARN(1, "Your BIOS is broken; DMAR reported at address zero!\n" - "BIOS vendor: %s; Ver: %s; Product Version: %s\n", - dmi_get_system_info(DMI_BIOS_VENDOR), - dmi_get_system_info(DMI_BIOS_VERSION), - dmi_get_system_info(DMI_PRODUCT_VERSION)); - bios_warned = 1; - } + warn_invalid_dmar(0, ""); return -EINVAL; } @@ -758,16 +756,7 @@ int alloc_iommu(struct dmar_drhd_unit *drhd) iommu->ecap = dmar_readq(iommu->reg + DMAR_ECAP_REG); if (iommu->cap == (uint64_t)-1 && iommu->ecap == (uint64_t)-1) { - if (!bios_warned) { - /* Promote an attitude of violence to a BIOS engineer today */ - WARN(1, "Your BIOS is broken; DMAR reported at address %llx returns all ones!\n" - "BIOS vendor: %s; Ver: %s; Product Version: %s\n", - drhd->reg_base_addr, - dmi_get_system_info(DMI_BIOS_VENDOR), - dmi_get_system_info(DMI_BIOS_VERSION), - dmi_get_system_info(DMI_PRODUCT_VERSION)); - bios_warned = 1; - } + warn_invalid_dmar(drhd->reg_base_addr, " returns all ones"); goto err_unmap; } @@ -806,7 +795,8 @@ int alloc_iommu(struct dmar_drhd_unit *drhd) } ver = readl(iommu->reg + DMAR_VER_REG); - pr_info("IOMMU %llx: ver %d:%d cap %llx ecap %llx\n", + pr_info("IOMMU %d: reg_base_addr %llx ver %d:%d cap %llx ecap %llx\n", + iommu->seq_id, (unsigned long long)drhd->reg_base_addr, DMAR_VER_MAJOR(ver), DMAR_VER_MINOR(ver), (unsigned long long)iommu->cap, @@ -1457,9 +1447,11 @@ int dmar_reenable_qi(struct intel_iommu *iommu) /* * Check interrupt remapping support in DMAR table description. */ -int dmar_ir_support(void) +int __init dmar_ir_support(void) { struct acpi_table_dmar *dmar; dmar = (struct acpi_table_dmar *)dmar_tbl; + if (!dmar) + return 0; return dmar->flags & 0x1; } diff --git a/drivers/pci/hotplug/acpiphp_ibm.c b/drivers/pci/hotplug/acpiphp_ibm.c index 6ecbfb27db9..e525263210e 100644 --- a/drivers/pci/hotplug/acpiphp_ibm.c +++ b/drivers/pci/hotplug/acpiphp_ibm.c @@ -108,7 +108,7 @@ static int ibm_set_attention_status(struct hotplug_slot *slot, u8 status); static int ibm_get_attention_status(struct hotplug_slot *slot, u8 *status); static void ibm_handle_events(acpi_handle handle, u32 event, void *context); static int ibm_get_table_from_acpi(char **bufp); -static ssize_t ibm_read_apci_table(struct kobject *kobj, +static ssize_t ibm_read_apci_table(struct file *filp, struct kobject *kobj, struct bin_attribute *bin_attr, char *buffer, loff_t pos, size_t size); static acpi_status __init ibm_find_acpi_device(acpi_handle handle, @@ -351,6 +351,7 @@ read_table_done: /** * ibm_read_apci_table - callback for the sysfs apci_table file + * @filp: the open sysfs file * @kobj: the kobject this binary attribute is a part of * @bin_attr: struct bin_attribute for this file * @buffer: the kernel space buffer to fill @@ -364,7 +365,7 @@ read_table_done: * things get really tricky here... * our solution is to only allow reading the table in all at once. */ -static ssize_t ibm_read_apci_table(struct kobject *kobj, +static ssize_t ibm_read_apci_table(struct file *filp, struct kobject *kobj, struct bin_attribute *bin_attr, char *buffer, loff_t pos, size_t size) { diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c index f184d1d2ecb..b3e5580c837 100644 --- a/drivers/pci/hotplug/cpqphp_core.c +++ b/drivers/pci/hotplug/cpqphp_core.c @@ -848,7 +848,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_disable_device; } - /* Check for the proper subsytem ID's + /* Check for the proper subsystem ID's * Intel uses a different SSID programming model than Compaq. * For Intel, each SSID bit identifies a PHP capability. * Also Intel HPC's may have RID=0. @@ -1075,13 +1075,12 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) /* make our own copy of the pci bus structure, * as we like tweaking it a lot */ - ctrl->pci_bus = kmalloc(sizeof(*ctrl->pci_bus), GFP_KERNEL); + ctrl->pci_bus = kmemdup(pdev->bus, sizeof(*ctrl->pci_bus), GFP_KERNEL); if (!ctrl->pci_bus) { err("out of memory\n"); rc = -ENOMEM; goto err_free_ctrl; } - memcpy(ctrl->pci_bus, pdev->bus, sizeof(*ctrl->pci_bus)); ctrl->bus = pdev->bus->number; ctrl->rev = pdev->revision; diff --git a/drivers/pci/hotplug/pciehp_pci.c b/drivers/pci/hotplug/pciehp_pci.c index 0a16444c14c..2fce726758d 100644 --- a/drivers/pci/hotplug/pciehp_pci.c +++ b/drivers/pci/hotplug/pciehp_pci.c @@ -84,12 +84,6 @@ int pciehp_configure_device(struct slot *p_slot) dev = pci_get_slot(parent, PCI_DEVFN(0, fn)); if (!dev) continue; - if ((dev->class >> 16) == PCI_BASE_CLASS_DISPLAY) { - ctrl_err(ctrl, "Cannot hot-add display device %s\n", - pci_name(dev)); - pci_dev_put(dev); - continue; - } if ((dev->hdr_type == PCI_HEADER_TYPE_BRIDGE) || (dev->hdr_type == PCI_HEADER_TYPE_CARDBUS)) { pciehp_add_bridge(dev); @@ -133,15 +127,9 @@ int pciehp_unconfigure_device(struct slot *p_slot) presence = 0; for (j = 0; j < 8; j++) { - struct pci_dev* temp = pci_get_slot(parent, PCI_DEVFN(0, j)); + struct pci_dev *temp = pci_get_slot(parent, PCI_DEVFN(0, j)); if (!temp) continue; - if ((temp->class >> 16) == PCI_BASE_CLASS_DISPLAY) { - ctrl_err(ctrl, "Cannot remove display device %s\n", - pci_name(temp)); - pci_dev_put(temp); - continue; - } if (temp->hdr_type == PCI_HEADER_TYPE_BRIDGE && presence) { pci_read_config_byte(temp, PCI_BRIDGE_CONTROL, &bctl); if (bctl & PCI_BRIDGE_CTL_VGA) { @@ -149,7 +137,8 @@ int pciehp_unconfigure_device(struct slot *p_slot) "Cannot remove display device %s\n", pci_name(temp)); pci_dev_put(temp); - continue; + rc = EINVAL; + break; } } pci_remove_bus_device(temp); diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c index 371dc564e2e..796828fce34 100644 --- a/drivers/pci/intel-iommu.c +++ b/drivers/pci/intel-iommu.c @@ -491,13 +491,11 @@ static void domain_update_iommu_coherency(struct dmar_domain *domain) domain->iommu_coherency = 1; - i = find_first_bit(&domain->iommu_bmp, g_num_of_iommus); - for (; i < g_num_of_iommus; ) { + for_each_set_bit(i, &domain->iommu_bmp, g_num_of_iommus) { if (!ecap_coherent(g_iommus[i]->ecap)) { domain->iommu_coherency = 0; break; } - i = find_next_bit(&domain->iommu_bmp, g_num_of_iommus, i+1); } } @@ -507,13 +505,11 @@ static void domain_update_iommu_snooping(struct dmar_domain *domain) domain->iommu_snooping = 1; - i = find_first_bit(&domain->iommu_bmp, g_num_of_iommus); - for (; i < g_num_of_iommus; ) { + for_each_set_bit(i, &domain->iommu_bmp, g_num_of_iommus) { if (!ecap_sc_support(g_iommus[i]->ecap)) { domain->iommu_snooping = 0; break; } - i = find_next_bit(&domain->iommu_bmp, g_num_of_iommus, i+1); } } @@ -1068,7 +1064,7 @@ static void iommu_flush_dev_iotlb(struct dmar_domain *domain, } static void iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did, - unsigned long pfn, unsigned int pages) + unsigned long pfn, unsigned int pages, int map) { unsigned int mask = ilog2(__roundup_pow_of_two(pages)); uint64_t addr = (uint64_t)pfn << VTD_PAGE_SHIFT; @@ -1089,10 +1085,10 @@ static void iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did, DMA_TLB_PSI_FLUSH); /* - * In caching mode, domain ID 0 is reserved for non-present to present - * mapping flush. Device IOTLB doesn't need to be flushed in this case. + * In caching mode, changes of pages from non-present to present require + * flush. However, device IOTLB doesn't need to be flushed in this case. */ - if (!cap_caching_mode(iommu->cap) || did) + if (!cap_caching_mode(iommu->cap) || !map) iommu_flush_dev_iotlb(iommu->domains[did], addr, mask); } @@ -1154,7 +1150,8 @@ static int iommu_init_domains(struct intel_iommu *iommu) unsigned long nlongs; ndomains = cap_ndoms(iommu->cap); - pr_debug("Number of Domains supportd <%ld>\n", ndomains); + pr_debug("IOMMU %d: Number of Domains supportd <%ld>\n", iommu->seq_id, + ndomains); nlongs = BITS_TO_LONGS(ndomains); spin_lock_init(&iommu->lock); @@ -1194,8 +1191,7 @@ void free_dmar_iommu(struct intel_iommu *iommu) unsigned long flags; if ((iommu->domains) && (iommu->domain_ids)) { - i = find_first_bit(iommu->domain_ids, cap_ndoms(iommu->cap)); - for (; i < cap_ndoms(iommu->cap); ) { + for_each_set_bit(i, iommu->domain_ids, cap_ndoms(iommu->cap)) { domain = iommu->domains[i]; clear_bit(i, iommu->domain_ids); @@ -1207,9 +1203,6 @@ void free_dmar_iommu(struct intel_iommu *iommu) domain_exit(domain); } spin_unlock_irqrestore(&domain->iommu_lock, flags); - - i = find_next_bit(iommu->domain_ids, - cap_ndoms(iommu->cap), i+1); } } @@ -1292,14 +1285,11 @@ static void iommu_detach_domain(struct dmar_domain *domain, spin_lock_irqsave(&iommu->lock, flags); ndomains = cap_ndoms(iommu->cap); - num = find_first_bit(iommu->domain_ids, ndomains); - for (; num < ndomains; ) { + for_each_set_bit(num, iommu->domain_ids, ndomains) { if (iommu->domains[num] == domain) { found = 1; break; } - num = find_next_bit(iommu->domain_ids, - cap_ndoms(iommu->cap), num+1); } if (found) { @@ -1485,15 +1475,12 @@ static int domain_context_mapping_one(struct dmar_domain *domain, int segment, /* find an available domain id for this device in iommu */ ndomains = cap_ndoms(iommu->cap); - num = find_first_bit(iommu->domain_ids, ndomains); - for (; num < ndomains; ) { + for_each_set_bit(num, iommu->domain_ids, ndomains) { if (iommu->domains[num] == domain) { id = num; found = 1; break; } - num = find_next_bit(iommu->domain_ids, - cap_ndoms(iommu->cap), num+1); } if (found == 0) { @@ -1558,7 +1545,7 @@ static int domain_context_mapping_one(struct dmar_domain *domain, int segment, (((u16)bus) << 8) | devfn, DMA_CCMD_MASK_NOBIT, DMA_CCMD_DEVICE_INVL); - iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_DSI_FLUSH); + iommu->flush.flush_iotlb(iommu, domain->id, 0, 0, DMA_TLB_DSI_FLUSH); } else { iommu_flush_write_buffer(iommu); } @@ -2333,14 +2320,16 @@ int __init init_dmars(void) */ iommu->flush.flush_context = __iommu_flush_context; iommu->flush.flush_iotlb = __iommu_flush_iotlb; - printk(KERN_INFO "IOMMU 0x%Lx: using Register based " + printk(KERN_INFO "IOMMU %d 0x%Lx: using Register based " "invalidation\n", + iommu->seq_id, (unsigned long long)drhd->reg_base_addr); } else { iommu->flush.flush_context = qi_flush_context; iommu->flush.flush_iotlb = qi_flush_iotlb; - printk(KERN_INFO "IOMMU 0x%Lx: using Queued " + printk(KERN_INFO "IOMMU %d 0x%Lx: using Queued " "invalidation\n", + iommu->seq_id, (unsigned long long)drhd->reg_base_addr); } } @@ -2621,7 +2610,7 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr, /* it's a non-present to present mapping. Only flush if caching mode */ if (cap_caching_mode(iommu->cap)) - iommu_flush_iotlb_psi(iommu, 0, mm_to_dma_pfn(iova->pfn_lo), size); + iommu_flush_iotlb_psi(iommu, domain->id, mm_to_dma_pfn(iova->pfn_lo), size, 1); else iommu_flush_write_buffer(iommu); @@ -2661,15 +2650,24 @@ static void flush_unmaps(void) if (!deferred_flush[i].next) continue; - iommu->flush.flush_iotlb(iommu, 0, 0, 0, + /* In caching mode, global flushes turn emulation expensive */ + if (!cap_caching_mode(iommu->cap)) + iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_GLOBAL_FLUSH); for (j = 0; j < deferred_flush[i].next; j++) { unsigned long mask; struct iova *iova = deferred_flush[i].iova[j]; - - mask = ilog2(mm_to_dma_pfn(iova->pfn_hi - iova->pfn_lo + 1)); - iommu_flush_dev_iotlb(deferred_flush[i].domain[j], - (uint64_t)iova->pfn_lo << PAGE_SHIFT, mask); + struct dmar_domain *domain = deferred_flush[i].domain[j]; + + /* On real hardware multiple invalidations are expensive */ + if (cap_caching_mode(iommu->cap)) + iommu_flush_iotlb_psi(iommu, domain->id, + iova->pfn_lo, iova->pfn_hi - iova->pfn_lo + 1, 0); + else { + mask = ilog2(mm_to_dma_pfn(iova->pfn_hi - iova->pfn_lo + 1)); + iommu_flush_dev_iotlb(deferred_flush[i].domain[j], + (uint64_t)iova->pfn_lo << PAGE_SHIFT, mask); + } __free_iova(&deferred_flush[i].domain[j]->iovad, iova); } deferred_flush[i].next = 0; @@ -2750,7 +2748,7 @@ static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr, if (intel_iommu_strict) { iommu_flush_iotlb_psi(iommu, domain->id, start_pfn, - last_pfn - start_pfn + 1); + last_pfn - start_pfn + 1, 0); /* free iova */ __free_iova(&domain->iovad, iova); } else { @@ -2840,7 +2838,7 @@ static void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist, if (intel_iommu_strict) { iommu_flush_iotlb_psi(iommu, domain->id, start_pfn, - last_pfn - start_pfn + 1); + last_pfn - start_pfn + 1, 0); /* free iova */ __free_iova(&domain->iovad, iova); } else { @@ -2874,7 +2872,6 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne struct dmar_domain *domain; size_t size = 0; int prot = 0; - size_t offset_pfn = 0; struct iova *iova = NULL; int ret; struct scatterlist *sg; @@ -2928,7 +2925,7 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne /* it's a non-present to present mapping. Only flush if caching mode */ if (cap_caching_mode(iommu->cap)) - iommu_flush_iotlb_psi(iommu, 0, start_vpfn, offset_pfn); + iommu_flush_iotlb_psi(iommu, domain->id, start_vpfn, size, 1); else iommu_flush_write_buffer(iommu); @@ -3436,22 +3433,6 @@ static void vm_domain_remove_all_dev_info(struct dmar_domain *domain) /* domain id for virtual machine, it won't be set in context */ static unsigned long vm_domid; -static int vm_domain_min_agaw(struct dmar_domain *domain) -{ - int i; - int min_agaw = domain->agaw; - - i = find_first_bit(&domain->iommu_bmp, g_num_of_iommus); - for (; i < g_num_of_iommus; ) { - if (min_agaw > g_iommus[i]->agaw) - min_agaw = g_iommus[i]->agaw; - - i = find_next_bit(&domain->iommu_bmp, g_num_of_iommus, i+1); - } - - return min_agaw; -} - static struct dmar_domain *iommu_alloc_vm_domain(void) { struct dmar_domain *domain; @@ -3512,8 +3493,7 @@ static void iommu_free_vm_domain(struct dmar_domain *domain) iommu = drhd->iommu; ndomains = cap_ndoms(iommu->cap); - i = find_first_bit(iommu->domain_ids, ndomains); - for (; i < ndomains; ) { + for_each_set_bit(i, iommu->domain_ids, ndomains) { if (iommu->domains[i] == domain) { spin_lock_irqsave(&iommu->lock, flags); clear_bit(i, iommu->domain_ids); @@ -3521,7 +3501,6 @@ static void iommu_free_vm_domain(struct dmar_domain *domain) spin_unlock_irqrestore(&iommu->lock, flags); break; } - i = find_next_bit(iommu->domain_ids, ndomains, i+1); } } } @@ -3582,7 +3561,6 @@ static int intel_iommu_attach_device(struct iommu_domain *domain, struct pci_dev *pdev = to_pci_dev(dev); struct intel_iommu *iommu; int addr_width; - u64 end; /* normally pdev is not mapped */ if (unlikely(domain_context_mapped(pdev))) { @@ -3605,14 +3583,30 @@ static int intel_iommu_attach_device(struct iommu_domain *domain, /* check if this iommu agaw is sufficient for max mapped address */ addr_width = agaw_to_width(iommu->agaw); - end = DOMAIN_MAX_ADDR(addr_width); - end = end & VTD_PAGE_MASK; - if (end < dmar_domain->max_addr) { - printk(KERN_ERR "%s: iommu agaw (%d) is not " + if (addr_width > cap_mgaw(iommu->cap)) + addr_width = cap_mgaw(iommu->cap); + + if (dmar_domain->max_addr > (1LL << addr_width)) { + printk(KERN_ERR "%s: iommu width (%d) is not " "sufficient for the mapped address (%llx)\n", - __func__, iommu->agaw, dmar_domain->max_addr); + __func__, addr_width, dmar_domain->max_addr); return -EFAULT; } + dmar_domain->gaw = addr_width; + + /* + * Knock out extra levels of page tables if necessary + */ + while (iommu->agaw < dmar_domain->agaw) { + struct dma_pte *pte; + + pte = dmar_domain->pgd; + if (dma_pte_present(pte)) { + free_pgtable_page(dmar_domain->pgd); + dmar_domain->pgd = (struct dma_pte *)dma_pte_addr(pte); + } + dmar_domain->agaw--; + } return domain_add_dev_info(dmar_domain, pdev, CONTEXT_TT_MULTI_LEVEL); } @@ -3632,7 +3626,6 @@ static int intel_iommu_map(struct iommu_domain *domain, { struct dmar_domain *dmar_domain = domain->priv; u64 max_addr; - int addr_width; int prot = 0; size_t size; int ret; @@ -3647,18 +3640,14 @@ static int intel_iommu_map(struct iommu_domain *domain, size = PAGE_SIZE << gfp_order; max_addr = iova + size; if (dmar_domain->max_addr < max_addr) { - int min_agaw; u64 end; /* check if minimum agaw is sufficient for mapped address */ - min_agaw = vm_domain_min_agaw(dmar_domain); - addr_width = agaw_to_width(min_agaw); - end = DOMAIN_MAX_ADDR(addr_width); - end = end & VTD_PAGE_MASK; + end = __DOMAIN_MAX_ADDR(dmar_domain->gaw) + 1; if (end < max_addr) { - printk(KERN_ERR "%s: iommu agaw (%d) is not " + printk(KERN_ERR "%s: iommu width (%d) is not " "sufficient for the mapped address (%llx)\n", - __func__, min_agaw, max_addr); + __func__, dmar_domain->gaw, max_addr); return -EFAULT; } dmar_domain->max_addr = max_addr; diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c index 6ee98a56946..1315ac688aa 100644 --- a/drivers/pci/intr_remapping.c +++ b/drivers/pci/intr_remapping.c @@ -832,9 +832,9 @@ static int ir_parse_ioapic_hpet_scope(struct acpi_dmar_header *header, return -1; } - printk(KERN_INFO "IOAPIC id %d under DRHD base" - " 0x%Lx\n", scope->enumeration_id, - drhd->address); + printk(KERN_INFO "IOAPIC id %d under DRHD base " + " 0x%Lx IOMMU %d\n", scope->enumeration_id, + drhd->address, iommu->seq_id); ir_parse_one_ioapic_scope(scope, iommu); } else if (scope->entry_type == ACPI_DMAR_SCOPE_TYPE_HPET) { diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index fad93983bfe..afd2fbf7d79 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -21,6 +21,7 @@ #include <linux/stat.h> #include <linux/topology.h> #include <linux/mm.h> +#include <linux/fs.h> #include <linux/capability.h> #include <linux/pci-aspm.h> #include <linux/slab.h> @@ -357,7 +358,8 @@ boot_vga_show(struct device *dev, struct device_attribute *attr, char *buf) struct device_attribute vga_attr = __ATTR_RO(boot_vga); static ssize_t -pci_read_config(struct kobject *kobj, struct bin_attribute *bin_attr, +pci_read_config(struct file *filp, struct kobject *kobj, + struct bin_attribute *bin_attr, char *buf, loff_t off, size_t count) { struct pci_dev *dev = to_pci_dev(container_of(kobj,struct device,kobj)); @@ -366,7 +368,7 @@ pci_read_config(struct kobject *kobj, struct bin_attribute *bin_attr, u8 *data = (u8*) buf; /* Several chips lock up trying to read undefined config space */ - if (capable(CAP_SYS_ADMIN)) { + if (cap_raised(filp->f_cred->cap_effective, CAP_SYS_ADMIN)) { size = dev->cfg_size; } else if (dev->hdr_type == PCI_HEADER_TYPE_CARDBUS) { size = 128; @@ -430,7 +432,8 @@ pci_read_config(struct kobject *kobj, struct bin_attribute *bin_attr, } static ssize_t -pci_write_config(struct kobject *kobj, struct bin_attribute *bin_attr, +pci_write_config(struct file* filp, struct kobject *kobj, + struct bin_attribute *bin_attr, char *buf, loff_t off, size_t count) { struct pci_dev *dev = to_pci_dev(container_of(kobj,struct device,kobj)); @@ -487,7 +490,8 @@ pci_write_config(struct kobject *kobj, struct bin_attribute *bin_attr, } static ssize_t -read_vpd_attr(struct kobject *kobj, struct bin_attribute *bin_attr, +read_vpd_attr(struct file *filp, struct kobject *kobj, + struct bin_attribute *bin_attr, char *buf, loff_t off, size_t count) { struct pci_dev *dev = @@ -502,7 +506,8 @@ read_vpd_attr(struct kobject *kobj, struct bin_attribute *bin_attr, } static ssize_t -write_vpd_attr(struct kobject *kobj, struct bin_attribute *bin_attr, +write_vpd_attr(struct file *filp, struct kobject *kobj, + struct bin_attribute *bin_attr, char *buf, loff_t off, size_t count) { struct pci_dev *dev = @@ -519,6 +524,7 @@ write_vpd_attr(struct kobject *kobj, struct bin_attribute *bin_attr, #ifdef HAVE_PCI_LEGACY /** * pci_read_legacy_io - read byte(s) from legacy I/O port space + * @filp: open sysfs file * @kobj: kobject corresponding to file to read from * @bin_attr: struct bin_attribute for this file * @buf: buffer to store results @@ -529,7 +535,8 @@ write_vpd_attr(struct kobject *kobj, struct bin_attribute *bin_attr, * callback routine (pci_legacy_read). */ static ssize_t -pci_read_legacy_io(struct kobject *kobj, struct bin_attribute *bin_attr, +pci_read_legacy_io(struct file *filp, struct kobject *kobj, + struct bin_attribute *bin_attr, char *buf, loff_t off, size_t count) { struct pci_bus *bus = to_pci_bus(container_of(kobj, @@ -545,6 +552,7 @@ pci_read_legacy_io(struct kobject *kobj, struct bin_attribute *bin_attr, /** * pci_write_legacy_io - write byte(s) to legacy I/O port space + * @filp: open sysfs file * @kobj: kobject corresponding to file to read from * @bin_attr: struct bin_attribute for this file * @buf: buffer containing value to be written @@ -555,7 +563,8 @@ pci_read_legacy_io(struct kobject *kobj, struct bin_attribute *bin_attr, * callback routine (pci_legacy_write). */ static ssize_t -pci_write_legacy_io(struct kobject *kobj, struct bin_attribute *bin_attr, +pci_write_legacy_io(struct file *filp, struct kobject *kobj, + struct bin_attribute *bin_attr, char *buf, loff_t off, size_t count) { struct pci_bus *bus = to_pci_bus(container_of(kobj, @@ -570,6 +579,7 @@ pci_write_legacy_io(struct kobject *kobj, struct bin_attribute *bin_attr, /** * pci_mmap_legacy_mem - map legacy PCI memory into user memory space + * @filp: open sysfs file * @kobj: kobject corresponding to device to be mapped * @attr: struct bin_attribute for this file * @vma: struct vm_area_struct passed to mmap @@ -579,7 +589,8 @@ pci_write_legacy_io(struct kobject *kobj, struct bin_attribute *bin_attr, * memory space. */ static int -pci_mmap_legacy_mem(struct kob |