diff options
author | Dmitry Torokhov <dmitry.torokhov@gmail.com> | 2009-09-13 21:16:56 -0700 |
---|---|---|
committer | Dmitry Torokhov <dmitry.torokhov@gmail.com> | 2009-09-13 21:16:56 -0700 |
commit | fc8e1ead9314cf0e0f1922e661428b93d3a50d88 (patch) | |
tree | f3cb97c4769b74f6627a59769f1ed5c92a13c58a /drivers/pci | |
parent | 2bcaa6a4238094c5695d5b1943078388d82d3004 (diff) | |
parent | 9de48cc300fb10f7d9faa978670becf5e352462a (diff) |
Merge branch 'next' into for-linus
Diffstat (limited to 'drivers/pci')
54 files changed, 4501 insertions, 2878 deletions
diff --git a/drivers/pci/Makefile b/drivers/pci/Makefile index ba6af162fd3..1ebd6b4c743 100644 --- a/drivers/pci/Makefile +++ b/drivers/pci/Makefile @@ -2,10 +2,11 @@ # Makefile for the PCI bus specific drivers. # -obj-y += access.o bus.o probe.o remove.o pci.o quirks.o slot.o \ +obj-y += access.o bus.o probe.o remove.o pci.o quirks.o \ pci-driver.o search.o pci-sysfs.o rom.o setup-res.o \ irq.o obj-$(CONFIG_PROC_FS) += proc.o +obj-$(CONFIG_SYSFS) += slot.o # Build PCI Express stuff if needed obj-$(CONFIG_PCIEPORTBUS) += pcie/ @@ -39,7 +40,6 @@ obj-$(CONFIG_ALPHA) += setup-bus.o setup-irq.o obj-$(CONFIG_ARM) += setup-bus.o setup-irq.o obj-$(CONFIG_PARISC) += setup-bus.o obj-$(CONFIG_SUPERH) += setup-bus.o setup-irq.o -obj-$(CONFIG_PPC32) += setup-irq.o obj-$(CONFIG_PPC) += setup-bus.o obj-$(CONFIG_MIPS) += setup-bus.o setup-irq.o obj-$(CONFIG_X86_VISWS) += setup-irq.o diff --git a/drivers/pci/access.c b/drivers/pci/access.c index 0f370651268..db23200c487 100644 --- a/drivers/pci/access.c +++ b/drivers/pci/access.c @@ -66,6 +66,25 @@ EXPORT_SYMBOL(pci_bus_write_config_byte); EXPORT_SYMBOL(pci_bus_write_config_word); EXPORT_SYMBOL(pci_bus_write_config_dword); +/** + * pci_bus_set_ops - Set raw operations of pci bus + * @bus: pci bus struct + * @ops: new raw operations + * + * Return previous raw operations + */ +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); + old_ops = bus->ops; + bus->ops = ops; + spin_unlock_irqrestore(&pci_lock, flags); + return old_ops; +} +EXPORT_SYMBOL(pci_bus_set_ops); /** * pci_read_vpd - Read one entry from Vital Product Data diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c index 97a8194063b..cef28a79103 100644 --- a/drivers/pci/bus.c +++ b/drivers/pci/bus.c @@ -41,9 +41,14 @@ pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res, void *alignf_data) { int i, ret = -ENOMEM; + resource_size_t max = -1; type_mask |= IORESOURCE_IO | IORESOURCE_MEM; + /* don't allocate too high if the pref mem doesn't support 64bit*/ + if (!(res->flags & IORESOURCE_MEM_64)) + max = PCIBIOS_MAX_MEM_32; + for (i = 0; i < PCI_BUS_NUM_RESOURCES; i++) { struct resource *r = bus->resource[i]; if (!r) @@ -62,7 +67,7 @@ pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res, /* Ok, try it out.. */ ret = allocate_resource(r, res, size, r->start ? : min, - -1, align, + max, align, alignf, alignf_data); if (ret == 0) break; @@ -201,13 +206,18 @@ void pci_enable_bridges(struct pci_bus *bus) * Walk the given bus, including any bridged devices * on buses under this bus. Call the provided callback * on each device found. + * + * We check the return of @cb each time. If it returns anything + * other than 0, we break out. + * */ -void pci_walk_bus(struct pci_bus *top, void (*cb)(struct pci_dev *, void *), +void pci_walk_bus(struct pci_bus *top, int (*cb)(struct pci_dev *, void *), void *userdata) { struct pci_dev *dev; struct pci_bus *bus; struct list_head *next; + int retval; bus = top; down_read(&pci_bus_sem); @@ -231,8 +241,10 @@ void pci_walk_bus(struct pci_bus *top, void (*cb)(struct pci_dev *, void *), /* Run device routines with the device locked */ down(&dev->dev.sem); - cb(dev, userdata); + retval = cb(dev, userdata); up(&dev->dev.sem); + if (retval) + break; } up_read(&pci_bus_sem); } diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c index fa3a11365ec..7b287cb38b7 100644 --- a/drivers/pci/dmar.c +++ b/drivers/pci/dmar.c @@ -267,6 +267,84 @@ rmrr_parse_dev(struct dmar_rmrr_unit *rmrru) } return ret; } + +static LIST_HEAD(dmar_atsr_units); + +static int __init dmar_parse_one_atsr(struct acpi_dmar_header *hdr) +{ + struct acpi_dmar_atsr *atsr; + struct dmar_atsr_unit *atsru; + + atsr = container_of(hdr, struct acpi_dmar_atsr, header); + atsru = kzalloc(sizeof(*atsru), GFP_KERNEL); + if (!atsru) + return -ENOMEM; + + atsru->hdr = hdr; + atsru->include_all = atsr->flags & 0x1; + + list_add(&atsru->list, &dmar_atsr_units); + + return 0; +} + +static int __init atsr_parse_dev(struct dmar_atsr_unit *atsru) +{ + int rc; + struct acpi_dmar_atsr *atsr; + + if (atsru->include_all) + return 0; + + atsr = container_of(atsru->hdr, struct acpi_dmar_atsr, header); + rc = dmar_parse_dev_scope((void *)(atsr + 1), + (void *)atsr + atsr->header.length, + &atsru->devices_cnt, &atsru->devices, + atsr->segment); + if (rc || !atsru->devices_cnt) { + list_del(&atsru->list); + kfree(atsru); + } + + return rc; +} + +int dmar_find_matched_atsr_unit(struct pci_dev *dev) +{ + int i; + struct pci_bus *bus; + struct acpi_dmar_atsr *atsr; + struct dmar_atsr_unit *atsru; + + 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)) + goto found; + } + + return 0; + +found: + for (bus = dev->bus; bus; bus = bus->parent) { + struct pci_dev *bridge = bus->self; + + if (!bridge || !bridge->is_pcie || + bridge->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE) + return 0; + + if (bridge->pcie_type == PCI_EXP_TYPE_ROOT_PORT) { + for (i = 0; i < atsru->devices_cnt; i++) + if (atsru->devices[i] == bridge) + return 1; + break; + } + } + + if (atsru->include_all) + return 1; + + return 0; +} #endif static void __init @@ -274,22 +352,28 @@ dmar_table_print_dmar_entry(struct acpi_dmar_header *header) { struct acpi_dmar_hardware_unit *drhd; struct acpi_dmar_reserved_memory *rmrr; + struct acpi_dmar_atsr *atsr; switch (header->type) { case ACPI_DMAR_TYPE_HARDWARE_UNIT: - drhd = (struct acpi_dmar_hardware_unit *)header; + drhd = container_of(header, struct acpi_dmar_hardware_unit, + header); printk (KERN_INFO PREFIX - "DRHD (flags: 0x%08x)base: 0x%016Lx\n", - drhd->flags, (unsigned long long)drhd->address); + "DRHD base: %#016Lx flags: %#x\n", + (unsigned long long)drhd->address, drhd->flags); break; case ACPI_DMAR_TYPE_RESERVED_MEMORY: - rmrr = (struct acpi_dmar_reserved_memory *)header; - + rmrr = container_of(header, struct acpi_dmar_reserved_memory, + header); printk (KERN_INFO PREFIX - "RMRR base: 0x%016Lx end: 0x%016Lx\n", + "RMRR base: %#016Lx end: %#016Lx\n", (unsigned long long)rmrr->base_address, (unsigned long long)rmrr->end_address); break; + case ACPI_DMAR_TYPE_ATSR: + atsr = container_of(header, struct acpi_dmar_atsr, header); + printk(KERN_INFO PREFIX "ATSR flags: %#x\n", atsr->flags); + break; } } @@ -363,6 +447,11 @@ parse_dmar_table(void) ret = dmar_parse_one_rmrr(entry_header); #endif break; + case ACPI_DMAR_TYPE_ATSR: +#ifdef CONFIG_DMAR + ret = dmar_parse_one_atsr(entry_header); +#endif + break; default: printk(KERN_WARNING PREFIX "Unknown DMAR structure type\n"); @@ -431,11 +520,19 @@ int __init dmar_dev_scope_init(void) #ifdef CONFIG_DMAR { struct dmar_rmrr_unit *rmrr, *rmrr_n; + struct dmar_atsr_unit *atsr, *atsr_n; + list_for_each_entry_safe(rmrr, rmrr_n, &dmar_rmrr_units, list) { ret = rmrr_parse_dev(rmrr); if (ret) return ret; } + + list_for_each_entry_safe(atsr, atsr_n, &dmar_atsr_units, list) { + ret = atsr_parse_dev(atsr); + if (ret) + return ret; + } } #endif @@ -468,6 +565,9 @@ int __init dmar_table_init(void) #ifdef CONFIG_DMAR if (list_empty(&dmar_rmrr_units)) printk(KERN_INFO PREFIX "No RMRR found\n"); + + if (list_empty(&dmar_atsr_units)) + printk(KERN_INFO PREFIX "No ATSR found\n"); #endif #ifdef CONFIG_INTR_REMAP @@ -515,6 +615,7 @@ int alloc_iommu(struct dmar_drhd_unit *drhd) u32 ver; static int iommu_allocated = 0; int agaw = 0; + int msagaw = 0; iommu = kzalloc(sizeof(*iommu), GFP_KERNEL); if (!iommu) @@ -535,12 +636,20 @@ int alloc_iommu(struct dmar_drhd_unit *drhd) agaw = iommu_calculate_agaw(iommu); if (agaw < 0) { printk(KERN_ERR - "Cannot get a valid agaw for iommu (seq_id = %d)\n", + "Cannot get a valid agaw for iommu (seq_id = %d)\n", + iommu->seq_id); + goto error; + } + msagaw = iommu_calculate_max_sagaw(iommu); + if (msagaw < 0) { + printk(KERN_ERR + "Cannot get a valid max agaw for iommu (seq_id = %d)\n", iommu->seq_id); goto error; } #endif iommu->agaw = agaw; + iommu->msagaw = msagaw; /* the registers might be more than one page */ map_size = max_t(int, ecap_max_iotlb_offset(iommu->ecap), @@ -590,7 +699,8 @@ void free_iommu(struct intel_iommu *iommu) */ static inline void reclaim_free_desc(struct q_inval *qi) { - while (qi->desc_status[qi->free_tail] == QI_DONE) { + while (qi->desc_status[qi->free_tail] == QI_DONE || + qi->desc_status[qi->free_tail] == QI_ABORT) { qi->desc_status[qi->free_tail] = QI_FREE; qi->free_tail = (qi->free_tail + 1) % QI_LENGTH; qi->free_cnt++; @@ -600,10 +710,13 @@ static inline void reclaim_free_desc(struct q_inval *qi) static int qi_check_fault(struct intel_iommu *iommu, int index) { u32 fault; - int head; + int head, tail; struct q_inval *qi = iommu->qi; int wait_index = (index + 1) % QI_LENGTH; + if (qi->desc_status[wait_index] == QI_ABORT) + return -EAGAIN; + fault = readl(iommu->reg + DMAR_FSTS_REG); /* @@ -613,7 +726,11 @@ static int qi_check_fault(struct intel_iommu *iommu, int index) */ if (fault & DMA_FSTS_IQE) { head = readl(iommu->reg + DMAR_IQH_REG); - if ((head >> 4) == index) { + if ((head >> DMAR_IQ_SHIFT) == index) { + printk(KERN_ERR "VT-d detected invalid descriptor: " + "low=%llx, high=%llx\n", + (unsigned long long)qi->desc[index].low, + (unsigned long long)qi->desc[index].high); memcpy(&qi->desc[index], &qi->desc[wait_index], sizeof(struct qi_desc)); __iommu_flush_cache(iommu, &qi->desc[index], @@ -623,6 +740,32 @@ static int qi_check_fault(struct intel_iommu *iommu, int index) } } + /* + * If ITE happens, all pending wait_desc commands are aborted. + * No new descriptors are fetched until the ITE is cleared. + */ + if (fault & DMA_FSTS_ITE) { + head = readl(iommu->reg + DMAR_IQH_REG); + head = ((head >> DMAR_IQ_SHIFT) - 1 + QI_LENGTH) % QI_LENGTH; + head |= 1; + tail = readl(iommu->reg + DMAR_IQT_REG); + tail = ((tail >> DMAR_IQ_SHIFT) - 1 + QI_LENGTH) % QI_LENGTH; + + writel(DMA_FSTS_ITE, iommu->reg + DMAR_FSTS_REG); + + do { + if (qi->desc_status[head] == QI_IN_USE) + qi->desc_status[head] = QI_ABORT; + head = (head - 2 + QI_LENGTH) % QI_LENGTH; + } while (head != tail); + + if (qi->desc_status[wait_index] == QI_ABORT) + return -EAGAIN; + } + + if (fault & DMA_FSTS_ICE) + writel(DMA_FSTS_ICE, iommu->reg + DMAR_FSTS_REG); + return 0; } @@ -632,7 +775,7 @@ static int qi_check_fault(struct intel_iommu *iommu, int index) */ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu) { - int rc = 0; + int rc; struct q_inval *qi = iommu->qi; struct qi_desc *hw, wait_desc; int wait_index, index; @@ -643,6 +786,9 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu) hw = qi->desc; +restart: + rc = 0; + spin_lock_irqsave(&qi->q_lock, flags); while (qi->free_cnt < 3) { spin_unlock_irqrestore(&qi->q_lock, flags); @@ -673,7 +819,7 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu) * update the HW tail register indicating the presence of * new descriptors. */ - writel(qi->free_head << 4, iommu->reg + DMAR_IQT_REG); + writel(qi->free_head << DMAR_IQ_SHIFT, iommu->reg + DMAR_IQT_REG); while (qi->desc_status[wait_index] != QI_DONE) { /* @@ -685,18 +831,21 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu) */ rc = qi_check_fault(iommu, index); if (rc) - goto out; + break; spin_unlock(&qi->q_lock); cpu_relax(); spin_lock(&qi->q_lock); } -out: - qi->desc_status[index] = qi->desc_status[wait_index] = QI_DONE; + + qi->desc_status[index] = QI_DONE; reclaim_free_desc(qi); spin_unlock_irqrestore(&qi->q_lock, flags); + if (rc == -EAGAIN) + goto restart; + return rc; } @@ -714,41 +863,26 @@ void qi_global_iec(struct intel_iommu *iommu) qi_submit_sync(&desc, iommu); } -int qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid, u8 fm, - u64 type, int non_present_entry_flush) +void qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid, u8 fm, + u64 type) { struct qi_desc desc; - if (non_present_entry_flush) { - if (!cap_caching_mode(iommu->cap)) - return 1; - else - did = 0; - } - desc.low = QI_CC_FM(fm) | QI_CC_SID(sid) | QI_CC_DID(did) | QI_CC_GRAN(type) | QI_CC_TYPE; desc.high = 0; - return qi_submit_sync(&desc, iommu); + qi_submit_sync(&desc, iommu); } -int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr, - unsigned int size_order, u64 type, - int non_present_entry_flush) +void qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr, + unsigned int size_order, u64 type) { u8 dw = 0, dr = 0; struct qi_desc desc; int ih = 0; - if (non_present_entry_flush) { - if (!cap_caching_mode(iommu->cap)) - return 1; - else - did = 0; - } - if (cap_write_drain(iommu->cap)) dw = 1; @@ -760,7 +894,28 @@ int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr, desc.high = QI_IOTLB_ADDR(addr) | QI_IOTLB_IH(ih) | QI_IOTLB_AM(size_order); - return qi_submit_sync(&desc, iommu); + qi_submit_sync(&desc, iommu); +} + +void qi_flush_dev_iotlb(struct intel_iommu *iommu, u16 sid, u16 qdep, + u64 addr, unsigned mask) +{ + struct qi_desc desc; + + if (mask) { + BUG_ON(addr & ((1 << (VTD_PAGE_SHIFT + mask)) - 1)); + addr |= (1 << (VTD_PAGE_SHIFT + mask - 1)) - 1; + desc.high = QI_DEV_IOTLB_ADDR(addr) | QI_DEV_IOTLB_SIZE; + } else + desc.high = QI_DEV_IOTLB_ADDR(addr); + + if (qdep >= QI_DEV_IOTLB_MAX_INVS) + qdep = 0; + + desc.low = QI_DEV_IOTLB_SID(sid) | QI_DEV_IOTLB_QDEP(qdep) | + QI_DIOTLB_TYPE; + + qi_submit_sync(&desc, iommu); } /* @@ -790,7 +945,6 @@ void dmar_disable_qi(struct intel_iommu *iommu) cpu_relax(); iommu->gcmd &= ~DMA_GCMD_QIE; - writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG); IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, readl, @@ -804,7 +958,7 @@ end: */ static void __dmar_enable_qi(struct intel_iommu *iommu) { - u32 cmd, sts; + u32 sts; unsigned long flags; struct q_inval *qi = iommu->qi; @@ -818,9 +972,8 @@ static void __dmar_enable_qi(struct intel_iommu *iommu) dmar_writeq(iommu->reg + DMAR_IQA_REG, virt_to_phys(qi->desc)); - cmd = iommu->gcmd | DMA_GCMD_QIE; iommu->gcmd |= DMA_GCMD_QIE; - writel(cmd, iommu->reg + DMAR_GCMD_REG); + writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG); /* Make sure hardware complete it */ IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, readl, (sts & DMA_GSTS_QIES), sts); @@ -1096,7 +1249,7 @@ int dmar_set_interrupt(struct intel_iommu *iommu) set_irq_data(irq, NULL); iommu->irq = 0; destroy_irq(irq); - return 0; + return ret; } ret = request_irq(irq, dmar_fault, 0, iommu->name, iommu); diff --git a/drivers/pci/hotplug/Kconfig b/drivers/pci/hotplug/Kconfig index 9aa4fe100a0..66f29bc00be 100644 --- a/drivers/pci/hotplug/Kconfig +++ b/drivers/pci/hotplug/Kconfig @@ -4,7 +4,7 @@ menuconfig HOTPLUG_PCI tristate "Support for PCI Hotplug" - depends on PCI && HOTPLUG + depends on PCI && HOTPLUG && SYSFS ---help--- Say Y here if you have a motherboard with a PCI Hotplug controller. This allows you to add and remove PCI cards while the machine is @@ -41,7 +41,7 @@ config HOTPLUG_PCI_FAKE config HOTPLUG_PCI_COMPAQ tristate "Compaq PCI Hotplug driver" - depends on X86 && PCI_BIOS && PCI_LEGACY + depends on X86 && PCI_BIOS help Say Y here if you have a motherboard with a Compaq PCI Hotplug controller. diff --git a/drivers/pci/hotplug/acpi_pcihp.c b/drivers/pci/hotplug/acpi_pcihp.c index fbc63d5e459..eb159587d0b 100644 --- a/drivers/pci/hotplug/acpi_pcihp.c +++ b/drivers/pci/hotplug/acpi_pcihp.c @@ -354,7 +354,7 @@ acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus, status = acpi_run_hpp(handle, hpp); if (ACPI_SUCCESS(status)) break; - if (acpi_root_bridge(handle)) + if (acpi_is_root_bridge(handle)) break; status = acpi_get_parent(handle, &phandle); if (ACPI_FAILURE(status)) @@ -428,7 +428,7 @@ int acpi_get_hp_hw_control_from_firmware(struct pci_dev *pdev, u32 flags) status = acpi_run_oshp(handle); if (ACPI_SUCCESS(status)) goto got_one; - if (acpi_root_bridge(handle)) + if (acpi_is_root_bridge(handle)) break; chandle = handle; status = acpi_get_parent(chandle, &handle); @@ -449,42 +449,6 @@ got_one: } EXPORT_SYMBOL(acpi_get_hp_hw_control_from_firmware); -/* acpi_root_bridge - check to see if this acpi object is a root bridge - * - * @handle - the acpi object in question. - */ -int acpi_root_bridge(acpi_handle handle) -{ - acpi_status status; - struct acpi_device_info *info; - struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL}; - int i; - - status = acpi_get_object_info(handle, &buffer); - if (ACPI_SUCCESS(status)) { - info = buffer.pointer; - if ((info->valid & ACPI_VALID_HID) && - !strcmp(PCI_ROOT_HID_STRING, - info->hardware_id.value)) { - kfree(buffer.pointer); - return 1; - } - if (info->valid & ACPI_VALID_CID) { - for (i=0; i < info->compatibility_id.count; i++) { - if (!strcmp(PCI_ROOT_HID_STRING, - info->compatibility_id.id[i].value)) { - kfree(buffer.pointer); - return 1; - } - } - } - kfree(buffer.pointer); - } - return 0; -} -EXPORT_SYMBOL_GPL(acpi_root_bridge); - - static int is_ejectable(acpi_handle handle) { acpi_status status; diff --git a/drivers/pci/hotplug/acpiphp_core.c b/drivers/pci/hotplug/acpiphp_core.c index 43c10bd261b..4dd7114964a 100644 --- a/drivers/pci/hotplug/acpiphp_core.c +++ b/drivers/pci/hotplug/acpiphp_core.c @@ -77,7 +77,6 @@ static int get_latch_status (struct hotplug_slot *slot, u8 *value); static int get_adapter_status (struct hotplug_slot *slot, u8 *value); static struct hotplug_slot_ops acpi_hotplug_slot_ops = { - .owner = THIS_MODULE, .enable_slot = enable_slot, .disable_slot = disable_slot, .set_attention_status = set_attention_status, diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index 3a6064bce56..0cb0f830a99 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -678,18 +678,9 @@ static void remove_bridge(acpi_handle handle) static struct pci_dev * get_apic_pci_info(acpi_handle handle) { - struct acpi_pci_id id; - struct pci_bus *bus; struct pci_dev *dev; - if (ACPI_FAILURE(acpi_get_pci_id(handle, &id))) - return NULL; - - bus = pci_find_bus(id.segment, id.bus); - if (!bus) - return NULL; - - dev = pci_get_slot(bus, PCI_DEVFN(id.device, id.function)); + dev = acpi_get_pci_dev(handle); if (!dev) return NULL; @@ -1396,19 +1387,16 @@ static void acpiphp_sanitize_bus(struct pci_bus *bus) /* Program resources in newly inserted bridge */ static int acpiphp_configure_bridge (acpi_handle handle) { - struct acpi_pci_id pci_id; + struct pci_dev *dev; struct pci_bus *bus; - if (ACPI_FAILURE(acpi_get_pci_id(handle, &pci_id))) { + dev = acpi_get_pci_dev(handle); + if (!dev) { err("cannot get PCI domain and bus number for bridge\n"); return -EINVAL; } - bus = pci_find_bus(pci_id.segment, pci_id.bus); - if (!bus) { - err("cannot find bus %d:%d\n", - pci_id.segment, pci_id.bus); - return -EINVAL; - } + + bus = dev->bus; pci_bus_size_bridges(bus); pci_bus_assign_resources(bus); @@ -1416,6 +1404,7 @@ static int acpiphp_configure_bridge (acpi_handle handle) acpiphp_set_hpp_values(handle, bus); pci_enable_bridges(bus); acpiphp_configure_ioapics(handle); + pci_dev_put(dev); return 0; } @@ -1631,7 +1620,7 @@ find_root_bridges(acpi_handle handle, u32 lvl, void *context, void **rv) { int *count = (int *)context; - if (acpi_root_bridge(handle)) { + if (acpi_is_root_bridge(handle)) { acpi_install_notify_handler(handle, ACPI_SYSTEM_NOTIFY, handle_hotplug_event_bridge, NULL); (*count)++; diff --git a/drivers/pci/hotplug/cpci_hotplug_core.c b/drivers/pci/hotplug/cpci_hotplug_core.c index de94f4feef8..d703e73fffa 100644 --- a/drivers/pci/hotplug/cpci_hotplug_core.c +++ b/drivers/pci/hotplug/cpci_hotplug_core.c @@ -32,7 +32,6 @@ #include <linux/pci_hotplug.h> #include <linux/init.h> #include <linux/interrupt.h> -#include <linux/smp_lock.h> #include <asm/atomic.h> #include <linux/delay.h> #include <linux/kthread.h> @@ -72,7 +71,6 @@ static int get_adapter_status(struct hotplug_slot *slot, u8 * value); static int get_latch_status(struct hotplug_slot *slot, u8 * value); static struct hotplug_slot_ops cpci_hotplug_slot_ops = { - .owner = THIS_MODULE, .enable_slot = enable_slot, .disable_slot = disable_slot, .set_attention_status = set_attention_status, diff --git a/drivers/pci/hotplug/cpqphp.h b/drivers/pci/hotplug/cpqphp.h index afaf8f69f73..53836001d51 100644 --- a/drivers/pci/hotplug/cpqphp.h +++ b/drivers/pci/hotplug/cpqphp.h @@ -150,25 +150,25 @@ struct ctrl_reg { /* offset */ /* offsets to the controller registers based on the above structure layout */ enum ctrl_offsets { - SLOT_RST = offsetof(struct ctrl_reg, slot_RST), + SLOT_RST = offsetof(struct ctrl_reg, slot_RST), SLOT_ENABLE = offsetof(struct ctrl_reg, slot_enable), MISC = offsetof(struct ctrl_reg, misc), LED_CONTROL = offsetof(struct ctrl_reg, led_control), INT_INPUT_CLEAR = offsetof(struct ctrl_reg, int_input_clear), - INT_MASK = offsetof(struct ctrl_reg, int_mask), - CTRL_RESERVED0 = offsetof(struct ctrl_reg, reserved0), + INT_MASK = offsetof(struct ctrl_reg, int_mask), + CTRL_RESERVED0 = offsetof(struct ctrl_reg, reserved0), CTRL_RESERVED1 = offsetof(struct ctrl_reg, reserved1), CTRL_RESERVED2 = offsetof(struct ctrl_reg, reserved1), - GEN_OUTPUT_AB = offsetof(struct ctrl_reg, gen_output_AB), - NON_INT_INPUT = offsetof(struct ctrl_reg, non_int_input), + GEN_OUTPUT_AB = offsetof(struct ctrl_reg, gen_output_AB), + NON_INT_INPUT = offsetof(struct ctrl_reg, non_int_input), CTRL_RESERVED3 = offsetof(struct ctrl_reg, reserved3), CTRL_RESERVED4 = offsetof(struct ctrl_reg, reserved4), CTRL_RESERVED5 = offsetof(struct ctrl_reg, reserved5), CTRL_RESERVED6 = offsetof(struct ctrl_reg, reserved6), CTRL_RESERVED7 = offsetof(struct ctrl_reg, reserved7), CTRL_RESERVED8 = offsetof(struct ctrl_reg, reserved8), - SLOT_MASK = offsetof(struct ctrl_reg, slot_mask), - CTRL_RESERVED9 = offsetof(struct ctrl_reg, reserved9), + SLOT_MASK = offsetof(struct ctrl_reg, slot_mask), + CTRL_RESERVED9 = offsetof(struct ctrl_reg, reserved9), CTRL_RESERVED10 = offsetof(struct ctrl_reg, reserved10), CTRL_RESERVED11 = offsetof(struct ctrl_reg, reserved11), SLOT_SERR = offsetof(struct ctrl_reg, slot_SERR), @@ -190,7 +190,9 @@ struct hrt { u32 reserved2; } __attribute__ ((packed)); -/* offsets to the hotplug resource table registers based on the above structure layout */ +/* offsets to the hotplug resource table registers based on the above + * structure layout + */ enum hrt_offsets { SIG0 = offsetof(struct hrt, sig0), SIG1 = offsetof(struct hrt, sig1), @@ -217,18 +219,20 @@ struct slot_rt { u16 pre_mem_length; } __attribute__ ((packed)); -/* offsets to the hotplug slot resource table registers based on the above structure layout */ +/* offsets to the hotplug slot resource table registers based on the above + * structure layout + */ enum slot_rt_offsets { DEV_FUNC = offsetof(struct slot_rt, dev_func), - PRIMARY_BUS = offsetof(struct slot_rt, primary_bus), - SECONDARY_BUS = offsetof(struct slot_rt, secondary_bus), - MAX_BUS = offsetof(struct slot_rt, max_bus), - IO_BASE = offsetof(struct slot_rt, io_base), - IO_LENGTH = offsetof(struct slot_rt, io_length), - MEM_BASE = offsetof(struct slot_rt, mem_base), - MEM_LENGTH = offsetof(struct slot_rt, mem_length), - PRE_MEM_BASE = offsetof(struct slot_rt, pre_mem_base), - PRE_MEM_LENGTH = offsetof(struct slot_rt, pre_mem_length), + PRIMARY_BUS = offsetof(struct slot_rt, primary_bus), + SECONDARY_BUS = offsetof(struct slot_rt, secondary_bus), + MAX_BUS = offsetof(struct slot_rt, max_bus), + IO_BASE = offsetof(struct slot_rt, io_base), + IO_LENGTH = offsetof(struct slot_rt, io_length), + MEM_BASE = offsetof(struct slot_rt, mem_base), + MEM_LENGTH = offsetof(struct slot_rt, mem_length), + PRE_MEM_BASE = offsetof(struct slot_rt, pre_mem_base), + PRE_MEM_LENGTH = offsetof(struct slot_rt, pre_mem_length), }; struct pci_func { @@ -286,8 +290,8 @@ struct event_info { struct controller { struct controller *next; u32 ctrl_int_comp; - struct mutex crit_sect; /* critical section mutex */ - void __iomem *hpc_reg; /* cookie for our pci controller location */ + struct mutex crit_sect; /* critical section mutex */ + void __iomem *hpc_reg; /* cookie for our pci controller location */ struct pci_resource *mem_head; struct pci_resource *p_mem_head; struct pci_resource *io_head; @@ -299,7 +303,7 @@ struct controller { u8 next_event; u8 interrupt; u8 cfgspc_irq; - u8 bus; /* bus number for the pci hotplug controller */ + u8 bus; /* bus number for the pci hotplug controller */ u8 rev; u8 slot_device_offset; u8 first_slot; @@ -401,46 +405,57 @@ struct resource_lists { /* debugfs functions for the hotplug controller info */ -extern void cpqhp_initialize_debugfs (void); -extern void cpqhp_shutdown_debugfs (void); -extern void cpqhp_create_debugfs_files (struct controller *ctrl); -extern void cpqhp_remove_debugfs_files (struct controller *ctrl); +extern void cpqhp_initialize_debugfs(void); +extern void cpqhp_shutdown_debugfs(void); +extern void cpqhp_create_debugfs_files(struct controller *ctrl); +extern void cpqhp_remove_debugfs_files(struct controller *ctrl); /* controller functions */ -extern void cpqhp_pushbutton_thread (unsigned long event_pointer); -extern irqreturn_t cpqhp_ctrl_intr (int IRQ, void *data); -extern int cpqhp_find_available_resources (struct controller *ctrl, void __iomem *rom_start); -extern int cpqhp_event_start_thread (void); -extern void cpqhp_event_stop_thread (void); -extern struct pci_func *cpqhp_slot_create (unsigned char busnumber); -extern struct pci_func *cpqhp_slot_find (unsigned char bus, unsigned char device, unsigned char index); -extern int cpqhp_process_SI (struct controller *ctrl, struct pci_func *func); -extern int cpqhp_process_SS (struct controller *ctrl, struct pci_func *func); -extern int cpqhp_hardware_test (struct controller *ctrl, int test_num); +extern void cpqhp_pushbutton_thread(unsigned long event_pointer); +extern irqreturn_t cpqhp_ctrl_intr(int IRQ, void *data); +extern int cpqhp_find_available_resources(struct controller *ctrl, + void __iomem *rom_start); +extern int cpqhp_event_start_thread(void); +extern void cpqhp_event_stop_thread(void); +extern struct pci_func *cpqhp_slot_create(unsigned char busnumber); +extern struct pci_func *cpqhp_slot_find(unsigned char bus, unsigned char device, + unsigned char index); +extern int cpqhp_process_SI(struct controller *ctrl, struct pci_func *func); +extern int cpqhp_process_SS(struct controller *ctrl, struct pci_func *func); +extern int cpqhp_hardware_test(struct controller *ctrl, int test_num); /* resource functions */ extern int cpqhp_resource_sort_and_combine (struct pci_resource **head); /* pci functions */ -extern int cpqhp_set_irq (u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num); -extern int cpqhp_get_bus_dev (struct controller *ctrl, u8 *bus_num, u8 *dev_num, u8 slot); -extern int cpqhp_save_config (struct controller *ctrl, int busnumber, int is_hot_plug); -extern int cpqhp_save_base_addr_length (struct controller *ctrl, struct pci_func * func); -extern int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func); -extern int cpqhp_configure_board (struct controller *ctrl, struct pci_func * func); -extern int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot); -extern int cpqhp_valid_replace (struct controller *ctrl, struct pci_func * func); -extern void cpqhp_destroy_board_resources (struct pci_func * func); -extern int cpqhp_return_board_resources (struct pci_func * func, struct resource_lists * resources); -extern void cpqhp_destroy_resource_list (struct resource_lists * resources); -extern int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func); -extern int cpqhp_unconfigure_device (struct pci_func* func); +extern int cpqhp_set_irq(u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num); +extern int cpqhp_get_bus_dev(struct controller *ctrl, u8 *bus_num, u8 *dev_num, + u8 slot); +extern int cpqhp_save_config(struct controller *ctrl, int busnumber, + int is_hot_plug); +extern int cpqhp_save_base_addr_length(struct controller *ctrl, + struct pci_func *func); +extern int cpqhp_save_used_resources(struct controller *ctrl, + struct pci_func *func); +extern int cpqhp_configure_board(struct controller *ctrl, + struct pci_func *func); +extern int cpqhp_save_slot_config(struct controller *ctrl, + struct pci_func *new_slot); +extern int cpqhp_valid_replace(struct controller *ctrl, struct pci_func *func); +extern void cpqhp_destroy_board_resources(struct pci_func *func); +extern int cpqhp_return_board_resources (struct pci_func *func, + struct resource_lists *resources); +extern void cpqhp_destroy_resource_list(struct resource_lists *resources); +extern int cpqhp_configure_device(struct controller *ctrl, + struct pci_func *func); +extern int cpqhp_unconfigure_device(struct pci_func *func); /* Global variables */ extern int cpqhp_debug; extern int cpqhp_legacy_mode; extern struct controller *cpqhp_ctrl_list; extern struct pci_func *cpqhp_slot_list[256]; +extern struct irq_routing_table *cpqhp_routing_table; /* these can be gotten rid of, but for debugging they are purty */ extern u8 cpqhp_nic_irq; @@ -449,7 +464,7 @@ extern u8 cpqhp_disk_irq; /* inline functions */ -static inline char *slot_name(struct slot *slot) +static inline const char *slot_name(struct slot *slot) { return hotplug_slot_name(slot->hotplug_slot); } @@ -458,9 +473,9 @@ static inline char *slot_name(struct slot *slot) * return_resource * * Puts node back in the resource list pointed to by head - * */ -static inline void return_resource(struct pci_resource **head, struct pci_resource *node) +static inline void return_resource(struct pci_resource **head, + struct pci_resource *node) { if (!node || !head) return; @@ -471,7 +486,7 @@ static inline void return_resource(struct pci_resource **head, struct pci_resour static inline void set_SOGO(struct controller *ctrl) { u16 misc; - + misc = readw(ctrl->hpc_reg + MISC); misc = (misc | 0x0001) & 0xFFFB; writew(misc, ctrl->hpc_reg + MISC); @@ -481,7 +496,7 @@ static inline void set_SOGO(struct controller *ctrl) static inline void amber_LED_on(struct controller *ctrl, u8 slot) { u32 led_control; - + led_control = readl(ctrl->hpc_reg + LED_CONTROL); led_control |= (0x01010000L << slot); writel(led_control, ctrl->hpc_reg + LED_CONTROL); @@ -491,7 +506,7 @@ static inline void amber_LED_on(struct controller *ctrl, u8 slot) static inline void amber_LED_off(struct controller *ctrl, u8 slot) { u32 led_control; - + led_control = readl(ctrl->hpc_reg + LED_CONTROL); led_control &= ~(0x01010000L << slot); writel(led_control, ctrl->hpc_reg + LED_CONTROL); @@ -504,7 +519,7 @@ static inline int read_amber_LED(struct controller *ctrl, u8 slot) led_control = readl(ctrl->hpc_reg + LED_CONTROL); led_control &= (0x01010000L << slot); - + return led_control ? 1 : 0; } @@ -512,7 +527,7 @@ static inline int read_amber_LED(struct controller *ctrl, u8 slot) static inline void green_LED_on(struct controller *ctrl, u8 slot) { u32 led_control; - + led_control = readl(ctrl->hpc_reg + LED_CONTROL); led_control |= 0x0101L << slot; writel(led_control, ctrl->hpc_reg + LED_CONTROL); @@ -521,7 +536,7 @@ static inline void green_LED_on(struct controller *ctrl, u8 slot) static inline void green_LED_off(struct controller *ctrl, u8 slot) { u32 led_control; - + led_control = readl(ctrl->hpc_reg + LED_CONTROL); led_control &= ~(0x0101L << slot); writel(led_control, ctrl->hpc_reg + LED_CONTROL); @@ -531,7 +546,7 @@ static inline void green_LED_off(struct controller *ctrl, u8 slot) static inline void green_LED_blink(struct controller *ctrl, u8 slot) { u32 led_control; - + led_control = readl(ctrl->hpc_reg + LED_CONTROL); led_control &= ~(0x0101L << slot); led_control |= (0x0001L << slot); @@ -575,22 +590,21 @@ static inline u8 read_slot_enable(struct controller *ctrl) } -/* +/** * get_controller_speed - find the current frequency/mode of controller. * * @ctrl: controller to get frequency/mode for. * * Returns controller speed. - * */ static inline u8 get_controller_speed(struct controller *ctrl) { u8 curr_freq; - u16 misc; - + u16 misc; + if (ctrl->pcix_support) { curr_freq = readb(ctrl->hpc_reg + NEXT_CURR_FREQ); - if ((curr_freq & 0xB0) == 0xB0) + if ((curr_freq & 0xB0) == 0xB0) return PCI_SPEED_133MHz_PCIX; if ((curr_freq & 0xA0) == 0xA0) return PCI_SPEED_100MHz_PCIX; @@ -602,19 +616,18 @@ static inline u8 get_controller_speed(struct controller *ctrl) return PCI_SPEED_33MHz; } - misc = readw(ctrl->hpc_reg + MISC); - return (misc & 0x0800) ? PCI_SPEED_66MHz : PCI_SPEED_33MHz; + misc = readw(ctrl->hpc_reg + MISC); + return (misc & 0x0800) ? PCI_SPEED_66MHz : PCI_SPEED_33MHz; } - -/* + +/** * get_adapter_speed - find the max supported frequency/mode of adapter. * * @ctrl: hotplug controller. * @hp_slot: hotplug slot where adapter is installed. * * Returns adapter speed. - * */ static inline u8 get_adapter_speed(struct controller *ctrl, u8 hp_slot) { @@ -672,7 +685,8 @@ static inline int get_slot_enabled(struct controller *ctrl, struct slot *slot) } -static inline int cpq_get_latch_status(struct controller *ctrl, struct slot *slot) +static inline int cpq_get_latch_status(struct controller *ctrl, + struct slot *slot) { u32 status; u8 hp_slot; @@ -687,7 +701,8 @@ static inline int cpq_get_latch_status(struct controller *ctrl, struct slot *slo } -static inline int get_presence_status(struct controller *ctrl, struct slot *slot) +static inline int get_presence_status(struct controller *ctrl, + struct slot *slot) { int presence_save = 0; u8 hp_slot; @@ -696,7 +711,8 @@ static inline int get_presence_status(struct controller *ctrl, struct slot *slot hp_slot = slot->device - ctrl->slot_device_offset; tempdword = readl(ctrl->hpc_reg + INT_INPUT_CLEAR); - presence_save = (int) ((((~tempdword) >> 23) | ((~tempdword) >> 15)) >> hp_slot) & 0x02; + presence_save = (int) ((((~tempdword) >> 23) | ((~tempdword) >> 15)) + >> hp_slot) & 0x02; return presence_save; } @@ -718,5 +734,12 @@ static inline int wait_for_ctrl_irq(struct controller *ctrl) return retval; } -#endif +#include <asm/pci_x86.h> +static inline int cpqhp_routing_table_length(void) +{ + BUG_ON(cpqhp_routing_table == NULL); + return ((cpqhp_routing_table->size - sizeof(struct irq_routing_table)) / + sizeof(struct irq_info)); +} +#endif diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c index c2e1bcbb28a..075b4f4b6e0 100644 --- a/drivers/pci/hotplug/cpqphp_core.c +++ b/drivers/pci/hotplug/cpqphp_core.c @@ -25,8 +25,7 @@ * Send feedback to <greg@kroah.com> * * Jan 12, 2003 - Added 66/100/133MHz PCI-X support, - * Torben Mathiasen <torben.mathiasen@hp.com> - * + * Torben Mathiasen <torben.mathiasen@hp.com> */ #include <linux/module.h> @@ -45,7 +44,6 @@ #include "cpqphp.h" #include "cpqphp_nvram.h" -#include <asm/pci_x86.h> /* Global variables */ @@ -53,6 +51,7 @@ int cpqhp_debug; int cpqhp_legacy_mode; struct controller *cpqhp_ctrl_list; /* = NULL */ struct pci_func *cpqhp_slot_list[256]; +struct irq_routing_table *cpqhp_routing_table; /* local variables */ static void __iomem *smbios_table; @@ -78,33 +77,6 @@ MODULE_PARM_DESC(debug, "Debugging mode enabled or not"); #define CPQHPC_MODULE_MINOR 208 -static int one_time_init (void); -static int set_attention_status (struct hotplug_slot *slot, u8 value); -static int process_SI (struct hotplug_slot *slot); -static int process_SS (struct hotplug_slot *slot); -static int hardware_test (struct hotplug_slot *slot, u32 value); -static int get_power_status (struct hotplug_slot *slot, u8 *value); -static int get_attention_status (struct hotplug_slot *slot, u8 *value); -static int get_latch_status (struct hotplug_slot *slot, u8 *value); -static int get_adapter_status (struct hotplug_slot *slot, u8 *value); -static int get_max_bus_speed (struct hotplug_slot *slot, enum pci_bus_speed *value); -static int get_cur_bus_speed (struct hotplug_slot *slot, enum pci_bus_speed *value); - -static struct hotplug_slot_ops cpqphp_hotplug_slot_ops = { - .owner = THIS_MODULE, - .set_attention_status = set_attention_status, - .enable_slot = process_SI, - .disable_slot = process_SS, - .hardware_test = hardware_test, - .get_power_status = get_power_status, - .get_attention_status = get_attention_status, - .get_latch_status = get_latch_status, - .get_adapter_status = get_adapter_status, - .get_max_bus_speed = get_max_bus_speed, - .get_cur_bus_speed = get_cur_bus_speed, -}; - - static inline int is_slot64bit(struct slot *slot) { return (readb(slot->p_sm_slot + SMBIOS_SLOT_WIDTH) == 0x06) ? 1 : 0; @@ -144,7 +116,7 @@ static void __iomem * detect_SMBIOS_pointer(void __iomem *begin, void __iomem *e break; } } - + if (!status) fp = NULL; @@ -171,7 +143,7 @@ static int init_SERR(struct controller * ctrl) tempdword = ctrl->first_slot; number_of_slots = readb(ctrl->hpc_reg + SLOT_MASK) & 0x0F; - // Loop through slots + /* Loop through slots */ while (number_of_slots) { physical_slot = tempdword; writeb(0, ctrl->hpc_reg + SLOT_SERR); @@ -182,41 +154,42 @@ static int init_SERR(struct controller * ctrl) return 0; } - -/* nice debugging output */ -static int pci_print_IRQ_route (void) +static int init_cpqhp_routing_table(void) { - struct irq_routing_table *routing_table; int len; - int loop; - - u8 tbus, tdevice, tslot; - routing_table = pcibios_get_irq_routing_table(); - if (routing_table == NULL) { - err("No BIOS Routing Table??? Not good\n"); + cpqhp_routing_table = pcibios_get_irq_routing_table(); + if (cpqhp_routing_table == NULL) return -ENOMEM; - } - len = (routing_table->size - sizeof(struct irq_routing_table)) / - sizeof(struct irq_info); - // Make sure I got at least one entry + len = cpqhp_routing_table_length(); if (len == 0) { - kfree(routing_table); + kfree(cpqhp_routing_table); + cpqhp_routing_table = NULL; return -1; } - dbg("bus dev func slot\n"); + return 0; +} + +/* nice debugging output */ +static void pci_print_IRQ_route(void) +{ + int len; + int loop; + u8 tbus, tdevice, tslot; + + len = cpqhp_routing_table_length(); + dbg("bus dev func slot\n"); for (loop = 0; loop < len; ++loop) { - tbus = routing_table->slots[loop].bus; - tdevice = routing_table->slots[loop].devfn; - tslot = routing_table->slots[loop].slot; + tbus = cpqhp_routing_table->slots[loop].bus; + tdevice = cpqhp_routing_table->slots[loop].devfn; + tslot = cpqhp_routing_table->slots[loop].slot; dbg("%d %d %d %d\n", tbus, tdevice >> 3, tdevice & 0x7, tslot); } - kfree(routing_table); - return 0; + return; } @@ -242,9 +215,9 @@ static void __iomem *get_subsequent_smbios_entry(void __iomem *smbios_start, void __iomem *p_max; if (!smbios_table || !curr) - return(NULL); + return NULL; - // set p_max to the end of the table + /* set p_max to the end of the table */ p_max = smbios_start + readw(smbios_table + ST_LENGTH); p_temp = curr; @@ -253,20 +226,19 @@ static void __iomem *get_subsequent_smbios_entry(void __iomem *smbios_start, while ((p_temp < p_max) && !bail) { /* Look for the double NULL terminator * The first condition is the previous byte - * and the second is the curr */ - if (!previous_byte && !(readb(p_temp))) { + * and the second is the curr + */ + if (!previous_byte && !(readb(p_temp))) bail = 1; - } previous_byte = readb(p_temp); p_temp++; } - if (p_temp < p_max) { + if (p_temp < p_max) return p_temp; - } else { + else return NULL; - } } @@ -292,21 +264,18 @@ static void __iomem *get_SMBIOS_entry(void __iomem *smbios_start, if (!smbios_table) return NULL; - if (!previous) { + if (!previous) previous = smbios_start; - } else { + else previous = get_subsequent_smbios_entry(smbios_start, smbios_table, previous); - } - while (previous) { - if (readb(previous + SMBIOS_GENERIC_TYPE) != type) { + while (previous) + if (readb(previous + SMBIOS_GENERIC_TYPE) != type) previous = get_subsequent_smbios_entry(smbios_start, smbios_table, previous); - } else { + else break; - } - } return previous; } @@ -322,144 +291,6 @@ static void release_slot(struct hotplug_slot *hotplug_slot) kfree(slot); } -#define SLOT_NAME_SIZE 10 - -static int ctrl_slot_setup(struct controller *ctrl, - void __iomem *smbios_start, - void __iomem *smbios_table) -{ - struct slot *slot; - struct hotplug_slot *hotplug_slot; - struct hotplug_slot_info *hotplug_slot_info; - u8 number_of_slots; - u8 slot_device; - u8 slot_number; - u8 ctrl_slot; - u32 tempdword; - char name[SLOT_NAME_SIZE]; - void __iomem *slot_entry= NULL; - int result = -ENOMEM; - - dbg("%s\n", __func__); - - tempdword = readl(ctrl->hpc_reg + INT_INPUT_CLEAR); - - number_of_slots = readb(ctrl->hpc_reg + SLOT_MASK) & 0x0F; - slot_device = readb(ctrl->hpc_reg + SLOT_MASK) >> 4; - slot_number = ctrl->first_slot; - - while (number_of_slots) { - slot = kzalloc(sizeof(*slot), GFP_KERNEL); - if (!slot) - goto error; - - slot->hotplug_slot = kzalloc(sizeof(*(slot->hotplug_slot)), - GFP_KERNEL); - if (!slot->hotplug_slot) - goto error_slot; - hotplug_slot = slot->hotplug_slot; - - hotplug_slot->info = - kzalloc(sizeof(*(hotplug_slot->info)), - GFP_KERNEL); - if (!hotplug_slot->info) - goto error_hpslot; - hotplug_slot_info = hotplug_slot->info; - - slot->ctrl = ctrl; - slot->bus = ctrl->bus; - slot->device = slot_device; - slot->number = slot_number; - dbg("slot->number = %u\n", slot->number); - - slot_entry = get_SMBIOS_entry(smbios_start, smbios_table, 9, - slot_entry); - - while (slot_entry && (readw(slot_entry + SMBIOS_SLOT_NUMBER) != - slot->number)) { - slot_entry = get_SMBIOS_entry(smbios_start, - smbios_table, 9, slot_entry); - } - - slot->p_sm_slot = slot_entry; - - init_timer(&slot->task_event); - slot->task_event.expires = jiffies + 5 * HZ; - slot->task_event.function = cpqhp_pushbutton_thread; - - //FIXME: these capabilities aren't used but if they are - // they need to be correctly implemented - slot->capabilities |= PCISLOT_REPLACE_SUPPORTED; - slot->capabilities |= PCISLOT_INTERLOCK_SUPPORTED; - - if (is_slot64bit(slot)) - slot->capabilities |= PCISLOT_64_BIT_SUPPORTED; - if (is_slot66mhz(slot)) - slot->capabilities |= PCISLOT_66_MHZ_SUPPORTED; - if (ctrl->speed == PCI_SPEED_66MHz) - slot->capabilities |= PCISLOT_66_MHZ_OPERATION; - - ctrl_slot = - slot_device - (readb(ctrl->hpc_reg + SLOT_MASK) >> 4); - - // Check presence - slot->capabilities |= - ((((~tempdword) >> 23) | - ((~tempdword) >> 15)) >> ctrl_slot) & 0x02; - // Check the switch state - slot->capabilities |= - ((~tempdword & 0xFF) >> ctrl_slot) & 0x01; - // Check the slot enable - slot->capabilities |= - ((read_slot_enable(ctrl) << 2) >> ctrl_slot) & 0x04; - - /* register this slot with the hotplug pci core */ - hotplug_slot->release = &release_slot; - hotplug_slot->private = slot; - snprintf(name, SLOT_NAME_SIZE, "%u", slot->number); - hotplug_slot->ops = &cpqphp_hotplug_slot_ops; - - hotplug_slot_info->power_status = get_slot_enabled(ctrl, slot); - hotplug_slot_info->attention_status = - cpq_get_attention_status(ctrl, slot); - hotplug_slot_info->latch_status = - cpq_get_latch_status(ctrl, slot); - hotplug_slot_info->adapter_status = - get_presence_status(ctrl, slot); - - dbg("registering bus %d, dev %d, number %d, " - "ctrl->slot_device_offset %d, slot %d\n", - slot->bus, slot->device, - slot->number, ctrl->slot_device_offset, - slot_number); - result = pci_hp_register(hotplug_slot, - ctrl->pci_dev->bus, - slot->device, - name); - if (result) { - err("pci_hp_register failed with error %d\n", result); - goto error_info; - } - - slot->next = ctrl->slot; - ctrl->slot = slot; - - number_of_slots--; - slot_device++; - slot_number++; - } - - return 0; -error_info: - kfree(hotplug_slot_info); -error_hpslot: - kfree(hotplug_slot); -error_slot: - kfree(slot); -error: - return result; -} - static int ctrl_slot_cleanup (struct controller * ctrl) { struct slot *old_slot, *next_slot; @@ -476,36 +307,32 @@ static int ctrl_slot_cleanup (struct controller * ctrl) cpqhp_remove_debugfs_files(ctrl); - //Free IRQ associated with hot plug device + /* Free IRQ associated with hot plug device */ free_irq(ctrl->interrupt, ctrl); - //Unmap the memory + /* Unmap the memory */ iounmap(ctrl->hpc_reg); - //Finally reclaim PCI mem + /* Finally reclaim PCI mem */ release_mem_region(pci_resource_start(ctrl->pci_dev, 0), pci_resource_len(ctrl->pci_dev, 0)); - return(0); + return 0; } -//============================================================================ -// function: get_slot_mapping -// -// Description: Attempts to determine a logical slot mapping for a PCI -// device. Won't work for more than one PCI-PCI bridge -// in a slot. -// -// Input: u8 bus_num - bus number of PCI device -// u8 dev_num - device number of PCI device -// u8 *slot - Pointer to u8 where slot number will -// be returned -// -// Output: SUCCESS or FAILURE -//============================================================================= +/** + * get_slot_mapping - determine logical slot mapping for PCI device + * + * Won't work for more than one PCI-PCI bridge in a slot. + * + * @bus_num - bus number of PCI device + * @dev_num - device number of PCI device + * @slot - Pointer to u8 where slot number will be returned + * + * Output: SUCCESS or FAILURE + */ static int get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot) { - struct irq_routing_table *PCIIRQRoutingInfoLength; u32 work; long len; long loop; @@ -516,36 +343,25 @@ get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot) bridgeSlot = 0xFF; - PCIIRQRoutingInfoLength = pcibios_get_irq_routing_table(); - if (!PCIIRQRoutingInfoLength) - return -1; - - len = (PCIIRQRoutingInfoLength->size - - sizeof(struct irq_routing_table)) / sizeof(struct irq_info); - // Make sure I got at least one entry - if (len == 0) { - kfree(PCIIRQRoutingInfoLength); - return -1; - } - + len = cpqhp_routing_table_length(); for (loop = 0; loop < len; ++loop) { - tbus = PCIIRQRoutingInfoLength->slots[loop].bus; - tdevice = PCIIRQRoutingInfoLength->slots[loop].devfn >> 3; - tslot = PCIIRQRoutingInfoLength->slots[loop].slot; + tbus = cpqhp_routing_table->slots[loop].bus; + tdevice = cpqhp_routing_table->slots[loop].devfn >> 3; + tslot = cpqhp_routing_table->slots[loop].slot; if ((tbus == bus_num) && (tdevice == dev_num)) { *slot = tslot; - kfree(PCIIRQRoutingInfoLength); return 0; } else { /* Did not get a match on the target PCI device. Check - * if the current IRQ table entry is a PCI-to-PCI bridge - * device. If so, and it's secondary bus matches the - * bus number for the target device, I need to save the - * bridge's slot number. If I can not find an entry for - * the target device, I will have to assume it's on the - * other side of the bridge, and assign it the bridge's - * slot. */ + * if the current IRQ table entry is a PCI-to-PCI + * bridge device. If so, and it's secondary bus + * matches the bus number for the target device, I need + * to save the bridge's slot number. If I can not find + * an entry for the target device, I will have to + * assume it's on the other side of the bridge, and + * assign it the bridge's slot. + */ bus->number = tbus; pci_bus_read_config_dword(bus, PCI_DEVFN(tdevice, 0), PCI_CLASS_REVISION, &work); @@ -555,25 +371,23 @@ get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot) PCI_DEVFN(tdevice, 0), PCI_PRIMARY_BUS, &work); // See if bridge's secondary bus matches target bus. - if (((work >> 8) & 0x000000FF) == (long) bus_num) { + if (((work >> 8) & 0x000000FF) == (long) bus_num) bridgeSlot = tslot; - } } } } - // If we got here, we didn't find an entry in the IRQ mapping table - // for the target PCI device. If we did determine that the target - // device is on the other side of a PCI-to-PCI bridge, return the - // slot number for the bridge. + /* If we got here, we didn't find an entry in the IRQ mapping table for + * the target PCI device. If we did determine that the target device + * is on the other side of a PCI-to-PCI bridge, return the slot number + * for the bridge. + */ if (bridgeSlot != 0xFF) { *slot = bridgeSlot; - kfree(PCIIRQRoutingInfoLength); return 0; } - kfree(PCIIRQRoutingInfoLength); - // Couldn't find an entry in the routing table for this PCI device + /* Couldn't find an entry in the routing table for this PCI device */ return -1; } @@ -591,32 +405,32 @@ cpqhp_set_attention_status(struct controller *ctrl, struct pci_func *func, u8 hp_slot; if (func == NULL) - return(1); + return 1; hp_slot = func->device - ctrl->slot_device_offset; - // Wait for exclusive access to hardware + /* Wait for exclusive access to hardware */ mutex_lock(&ctrl->crit_sect); - if (status == 1) { + if (status == 1) amber_LED_on (ctrl, hp_slot); - } else if (status == 0) { + else if (status == 0) amber_LED_off (ctrl, hp_slot); - } else { - // Done with exclusive hardware access + else { + /* Done with exclusive hardware access */ mutex_unlock(&ctrl->crit_sect); - return(1); + return 1; } set_SOGO(ctrl); - // Wait for SOBS to be unset + /* Wait for SOBS to be unset */ wait_for_ctrl_irq (ctrl); - // Done with exclusive hardware access + /* Done with exclusive hardware access */ mutex_unlock(&ctrl->crit_sect); - return(0); + return 0; } @@ -719,7 +533,7 @@ static int hardware_test(struct hotplug_slot *hotplug_slot, u32 value) dbg("%s - physical_slot = %s\n", __func__, slot_name(slot)); - return cpqhp_hardware_test(ctrl, value); + return cpqhp_hardware_test(ctrl, value); } @@ -738,7 +552,7 @@ static int get_attention_status(struct hotplug_slot *hotplug_slot, u8 *value) { struct slot *slot = hotplug_slot->private; struct controller *ctrl = slot->ctrl; - + dbg("%s - physical_slot = %s\n", __func__, slot_name(slot)); *value = cpq_get_attention_status(ctrl, slot); @@ -793,6 +607,230 @@ static int get_cur_bus_speed (struct hotplug_slot *hotplug_slot, enum pci_bus_sp return 0; } +static struct hotplug_slot_ops cpqphp_hotplug_slot_ops = { + .set_attention_status = set_attention_status, + .enable_slot = process_SI, + .disable_slot = process_SS, + .hardware_test = hardware_test, + .get_power_status = get_power_status, + .get_attention_status = get_attention_status, + .get_latch_status = get_latch_status, + .get_adapter_status = get_adapter_status, + .get_max_bus_speed = get_max_bus_speed, + .get_cur_bus_speed = get_cur_bus_speed, +}; + +#define SLOT_NAME_SIZE 10 + +static int ctrl_slot_setup(struct controller *ctrl, + void __iomem *smbios_start, + void __iomem *smbios_table) +{ + struct slot *slot; + struct hotplug_slot *hotplug_slot; + struct hotplug_slot_info *hotplug_slot_info; + u8 number_of_slots; + u8 slot_device; + u8 slot_number; + u8 ctrl_slot; + u32 tempdword; + char name[SLOT_NAME_SIZE]; + void __iomem *slot_entry= NULL; + int result = -ENOMEM; + + dbg("%s\n", __func__); + + tempdword = readl(ctrl->hpc_reg + INT_INPUT_CLEAR); + + number_of_slots = readb(ctrl->hpc_reg + SLOT_MASK) & 0x0F; + slot_device = readb(ctrl->hpc_reg + SLOT_MASK) >> 4; + slot_number = ctrl->first_slot; + + while (number_of_slots) { + slot = kzalloc(sizeof(*slot), GFP_KERNEL); + if (!slot) + goto error; + + slot->hotplug_slot = kzalloc(sizeof(*(slot->hotplug_slot)), + GFP_KERNEL); + if (!slot->hotplug_slot) + goto error_slot; + hotplug_slot = slot->hotplug_slot; + + hotplug_slot->info = kzalloc(sizeof(*(hotplug_slot->info)), + GFP_KERNEL); + if (!hotplug_slot->info) + goto error_hpslot; + hotplug_slot_info = hotplug_slot->info; + + slot->ctrl = ctrl; + slot->bus = ctrl->bus; + slot->device = slot_device; + slot->number = slot_number; + dbg("slot->number = %u\n", slot->number); + + slot_entry = get_SMBIOS_entry(smbios_start, smbios_table, 9, + slot_entry); + + while (slot_entry && (readw(slot_entry + SMBIOS_SLOT_NUMBER) != + slot->number)) { + slot_entry = get_SMBIOS_entry(smbios_start, + smbios_table, 9, slot_entry); + } + + slot->p_sm_slot = slot_entry; + + init_timer(&slot->task_event); + slot->task_event.expires = jiffies + 5 * HZ; + slot->task_event.function = cpqhp_pushbutton_thread; + + /*FIXME: these capabilities aren't used but if they are + * they need to be correctly implemented + */ + slot->capabilities |= PCISLOT_REPLACE_SUPPORTED; + slot->capabilities |= PCISLOT_INTERLOCK_SUPPORTED; + + if (is_slot64bit(slot)) + slot->capabilities |= PCISLOT_64_BIT_SUPPORTED; + if (is_slot66mhz(slot)) + slot->capabilities |= PCISLOT_66_MHZ_SUPPORTED; + if (ctrl->speed == PCI_SPEED_66MHz) + slot->capabilities |= PCISLOT_66_MHZ_OPERATION; + + ctrl_slot = + slot_device - (readb(ctrl->hpc_reg + SLOT_MASK) >> 4); + + /* Check presence */ + slot->capabilities |= + ((((~tempdword) >> 23) | + ((~tempdword) >> 15)) >> ctrl_slot) & 0x02; + /* Check the switch state */ + slot->capabilities |= + ((~tempdword & 0xFF) >> ctrl_slot) & 0x01; + /* Check the slot enable */ + slot->capabilities |= + ((read_slot_enable(ctrl) << 2) >> ctrl_slot) & 0x04; + + /* register this slot with the hotplug pci core */ + hotplug_slot->release = &release_slot; + hotplug_slot->private = slot; + snprintf(name, SLOT_NAME_SIZE, "%u", slot->number); + hotplug_slot->ops = &cpqphp_hotplug_slot_ops; + + hotplug_slot_info->power_status = get_slot_enabled(ctrl, slot); + hotplug_slot_info->attention_status = + cpq_get_attention_status(ctrl, slot); + hotplug_slot_info->latch_status = + cpq_get_latch_status(ctrl, slot); + hotplug_slot_info->adapter_status = + get_presence_status(ctrl, slot); + + dbg("registering bus %d, dev %d, number %d, " + "ctrl->slot_device_offset %d, slot %d\n", + slot->bus, slot->device, + slot->number, ctrl->slot_device_offset, + slot_number); + result = pci_hp_register(hotplug_slot, + ctrl->pci_dev->bus, + slot->device, + name); + if (result) { + err("pci_hp_register failed with error %d\n", result); + goto error_info; + } + + slot->next = ctrl->slot; + ctrl->slot = slot; + + number_of_slots--; + slot_device++; + slot_number++; + } + + return 0; +error_info: + kfree(hotplug_slot_info); +error_hpslot: + kfree(hotplug_slot); +error_slot: + kfree(slot); +error: + return result; +} + +static int one_time_init(void) +{ + int loop; + int retval = 0; + + if (initialized) + return 0; + + power_mode = 0; + + retval = init_cpqhp_routing_table(); + if (retval) + goto error; + + if (cpqhp_debug) + pci_print_IRQ_route(); + + dbg("Initialize + Start the notification mechanism \n"); + + retval = cpqhp_event_start_thread(); + if (retval) + goto error; + + dbg("Initialize slot lists\n"); + for (loop = 0; loop < 256; loop++) + cpqhp_slot_list[loop] = NULL; + + /* FIXME: We also need to hook the NMI handler eventually. + * this also needs to be worked with Christoph + * register_NMI_handler(); + */ + /* Map rom address */ + cpqhp_rom_start = ioremap(ROM_PHY_ADDR, ROM_PHY_LEN); + if (!cpqhp_rom_start) { + err ("Could not ioremap memory region for ROM\n"); + retval = -EIO; + goto error; + } + + /* Now, map the int15 entry point if we are on compaq specific + * hardware + */ + compaq_nvram_init(cpqhp_rom_start); + + /* Map smbios table entry point structure */ + smbios_table = detect_SMBIOS_pointer(cpqhp_rom_start, + cpqhp_rom_start + ROM_PHY_LEN); + if (!smbios_table) { + err ("Could not find the SMBIOS pointer in memory\n"); + retval = -EIO; + goto error_rom_start; + } + + smbios_start = ioremap(readl(smbios_table + ST_ADDRESS), + readw(smbios_table + ST_LENGTH)); + if (!smbios_start) { + err ("Could not ioremap memory region taken from SMBIOS values\n"); + retval = -EIO; + goto error_smbios_start; + } + + initialized = 1; + + return retval; + +error_smbios_start: + iounmap(smbios_start); +error_rom_start: + iounmap(cpqhp_rom_start); +error: + return retval; +} + static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { u8 num_of_slots = 0; @@ -815,7 +853,9 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) return err; } - // Need to read VID early b/c it's used to differentiate CPQ and INTC discovery + /* Need to read VID early b/c it's used to differentiate CPQ and INTC + * discovery + */ rc = pci_read_config_word(pdev, PCI_VENDOR_ID, &vendor_id); if (rc || ((vendor_id != PCI_VENDOR_ID_COMPAQ) && (vendor_id != PCI_VENDOR_ID_INTEL))) { err(msg_HPC_non_compaq_or_intel); @@ -832,217 +872,209 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) } /* Check for the proper subsytem ID's - * Intel uses a different SSID programming model than Compaq. + * 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. */ - if ((pdev->revision > 2) || (vendor_id == PCI_VENDOR_ID_INTEL)) { - // TODO: This code can be made to support non-Compaq or Intel subsystem IDs - rc = pci_read_config_word(pdev, PCI_SUBSYSTEM_VENDOR_ID, &subsystem_vid); - if (rc) { - err("%s : pci_read_config_word failed\n", __func__); - goto err_disable_device; - } - dbg("Subsystem Vendor ID: %x\n", subsystem_vid); - if ((subsystem_vid != PCI_VENDOR_ID_COMPAQ) && (subsystem_vid != PCI_VENDOR_ID_INTEL)) { - err(msg_HPC_non_compaq_or_intel); - rc = -ENODEV; - goto err_disable_device; - } + if ((pdev->revision <= 2) && (vendor_id != PCI_VENDOR_ID_INTEL)) { + err(msg_HPC_not_supported); + return -ENODEV; + } - ctrl = kzalloc(sizeof(struct controller), GFP_KERNEL); - if (!ctrl) { - err("%s : out of memory\n", __func__); - rc = -ENOMEM; - goto err_disable_device; - } + /* TODO: This code can be made to support non-Compaq or Intel + * subsystem IDs + */ + rc = pci_read_config_word(pdev, PCI_SUBSYSTEM_VENDOR_ID, &subsystem_vid); + if (rc) { + err("%s : pci_read_config_word failed\n", __func__); + goto err_disable_device; + } + dbg("Subsystem Vendor ID: %x\n", subsystem_vid); + if ((subsystem_vid != PCI_VENDOR_ID_COMPAQ) && (subsystem_vid != PCI_VENDOR_ID_INTEL)) { + err(msg_HPC_non_compaq_or_intel); + rc = -ENODEV; + goto err_disable_device; + } - rc = pci_read_config_word(pdev, PCI_SUBSYSTEM_ID, &subsystem_deviceid); - if (rc) { - err("%s : pci_read_config_word failed\n", __func__); - goto err_free_ctrl; - } + ctrl = kzalloc(sizeof(struct controller), GFP_KERNEL); + if (!ctrl) { + err("%s : out of memory\n", __func__); + rc = -ENOMEM; + goto err_disable_device; + } - info("Hot Plug Subsystem Device ID: %x\n", subsystem_deviceid); - - /* Set Vendor ID, so it can be accessed later from other functions */ - ctrl->vendor_id = vendor_id; - - switch (subsystem_vid) { - case PCI_VENDOR_ID_COMPAQ: - if (pdev->revision >= 0x13) { /* CIOBX */ - ctrl->push_flag = 1; - ctrl->slot_switch_type = 1; - ctrl->push_button = 1; - ctrl->pci_config_space = 1; - ctrl->defeature_PHP = 1; - ctrl->pcix_support = 1; - ctrl->pcix_speed_capability = 1; - pci_read_config_byte(pdev, 0x41, &bus_cap); - if (bus_cap & 0x80) { - dbg("bus max supports 133MHz PCI-X\n"); - ctrl->speed_capability = PCI_SPEED_133MHz_PCIX; - break; - } - if (bus_cap & 0x40) { - dbg("bus max supports 100MHz PCI-X\n"); - ctrl->speed_capability = PCI_SPEED_100MHz_PCIX; - break; - } - if (bus_cap & 20) { - dbg("bus max supports 66MHz PCI-X\n"); - ctrl->speed_capability = PCI_SPEED_66MHz_PCIX; - break; - } - if (bus_cap & 10) { - dbg("bus max supports 66MHz PCI\n"); - ctrl->speed_capability = PCI_SPEED_66MHz; - break; - } - - break; - } - - switch (subsystem_deviceid) { - case PCI_SUB_HPC_ID: - /* Original 6500/7000 implementation */ - ctrl->slot_switch_type = 1; - ctrl->speed_capability = PCI_SPEED_33MHz; - ctrl->push_button = 0; - ctrl->pci_config_space = 1; - ctrl->defeature_PHP = 1; - ctrl->pcix_support = 0; - ctrl->pcix_speed_capability = 0; - break; - case PCI_SUB_HPC_ID2: - /* First Pushbutton implementation */ - ctrl->push_flag = 1; - ctrl->slot_switch_type = 1; - ctrl->speed_capability = PCI_SPEED_33MHz; - ctrl->push_button = 1; - ctrl->pci_config_space = 1; - ctrl->defeature_PHP = 1; - ctrl->pcix_support = 0; - ctrl->pcix_speed_capability = 0; - break; - case PCI_SUB_HPC_ID_INTC: - /* Third party (6500/7000) */ - ctrl->slot_switch_type = 1; - ctrl->speed_capability = PCI_SPEED_33MHz; - ctrl->push_button = 0; - ctrl->pci_config_space = 1; - ctrl->defeature_PHP = 1; - ctrl->pcix_support = 0; - ctrl->pcix_speed_capability = 0; - break; - case PCI_SUB_HPC_ID3: - /* First 66 Mhz implementation */ - ctrl->push_flag = 1; - ctrl->slot_switch_type = 1; - ctrl->speed_capability = PCI_SPEED_66MHz; - ctrl->push_button = 1; - ctrl->pci_config_space = 1; - ctrl->defeature_PHP = 1; - ctrl->pcix_support = 0; - ctrl->pcix_speed_capability = 0; - break; - case PCI_SUB_HPC_ID4: - /* First PCI-X implementation, 100MHz */ - ctrl->push_flag = 1; - ctrl->slot_switch_type = 1; - ctrl->speed_capability = PCI_SPEED_100MHz_PCIX; - ctrl->push_button = 1; - ctrl->pci_config_space = 1; - ctrl->defeature_PHP = 1; - ctrl->pcix_support = 1; - ctrl->pcix_speed_capability = 0; - break; - default: - err(msg_HPC_not_supported); - rc = -ENODEV; - goto err_free_ctrl; - } - break; + rc = pci_read_config_word(pdev, PCI_SUBSYSTEM_ID, &subsystem_deviceid); + if (rc) { + err("%s : pci_read_config_word failed\n", __func__); + goto err_free_ctrl; + } - case PCI_VENDOR_ID_INTEL: - /* Check for speed capability (0=33, 1=66) */ - if (subsystem_deviceid & 0x0001) { - ctrl->speed_capability = PCI_SPEED_66MHz; - } else { - ctrl->speed_capability = PCI_SPEED_33MHz; - } - - /* Check for push button */ - if (subsystem_deviceid & 0x0002) { - /* no push button */ - ctrl->push_button = 0; - } else { - /* push button supported */ - ctrl->push_button = 1; - } - - /* Check for slot switch type (0=mechanical, 1=not mechanical) */ - if (subsystem_deviceid & 0x0004) { - /* no switch */ - ctrl->slot_switch_type = 0; - } else { - /* switch */ - ctrl->slot_switch_type = 1; - } - - /* PHP Status (0=De-feature PHP, 1=Normal operation) */ - if (subsystem_deviceid & 0x0008) { - ctrl->defeature_PHP = 1; // PHP supported - } else { - ctrl->defeature_PHP = 0; // PHP not supported - } - - /* Alternate Base Address Register Interface (0=not supported, 1=supported) */ - if (subsystem_deviceid & 0x0010) { - ctrl->alternate_base_address = 1; // supported - } else { - ctrl->alternate_base_address = 0; // not supported - } - - /* PCI Config Space Index (0=not supported, 1=supported) */ - if (subsystem_deviceid & 0x0020) { - ctrl->pci_config_space = 1; // supported - } else { - ctrl->pci_config_space = 0; // not supported - } - - /* PCI-X support */ - if (subsystem_deviceid & 0x0080) { - /* PCI-X capable */ - ctrl->pcix_support = 1; - /* Frequency of operation in PCI-X mode */ - if (subsystem_deviceid & 0x0040) { - /* 133MHz PCI-X if bit 7 is 1 */ - ctrl->pcix_speed_capability = 1; - } else { - /* 100MHz PCI-X if bit 7 is 1 and bit 0 is 0, */ - /* 66MHz PCI-X if bit 7 is 1 and bit 0 is 1 */ - ctrl->pcix_speed_capability = 0; - } - } else { - /* Conventional PCI */ - ctrl->pcix_support = 0; - ctrl->pcix_speed_capability = 0; - } + info("Hot Plug Subsystem Device ID: %x\n", subsystem_deviceid); + + /* Set Vendor ID, so it can be accessed later from other + * functions + */ + ctrl->vendor_id = vendor_id; + + switch (subsystem_vid) { + case PCI_VENDOR_ID_COMPAQ: + if (pdev->revision >= 0x13) { /* CIOBX */ + ctrl->push_flag = 1; + ctrl->slot_switch_type = 1; + ctrl->push_button = 1; + ctrl->pci_config_space = 1; + ctrl->defeature_PHP = 1; + ctrl->pcix_support = 1; + ctrl->pcix_speed_capability = 1; + pci_read_config_byte(pdev, 0x41, &bus_cap); + if (bus_cap & 0x80) { + dbg("bus max supports 133MHz PCI-X\n"); + ctrl->speed_capability = PCI_SPEED_133MHz_PCIX; break; + } + if (bus_cap & 0x40) { + dbg("bus max supports 100MHz PCI-X\n"); + ctrl->speed_capability = PCI_SPEED_100MHz_PCIX; + break; + } + if (bus_cap & 20) { + dbg("bus max supports 66MHz PCI-X\n"); + ctrl->speed_capability = PCI_SPEED_66MHz_PCIX; + break; + } + if (bus_cap & 10) { + dbg("bus max supports 66MHz PCI\n"); + ctrl->speed_capability = PCI_SPEED_66MHz; + break; + } + + break; + } - default: - err(msg_HPC_not_supported); - rc = -ENODEV; - goto err_free_ctrl; + switch (subsystem_deviceid) { + case PCI_SUB_HPC_ID: + /* Original 6500/7000 implementation */ + ctrl->slot_switch_type = 1; + ctrl->speed_capability = PCI_SPEED_33MHz; + ctrl->push_button = 0; + ctrl->pci_config_space = 1; + ctrl->defeature_PHP = 1; + ctrl->pcix_support = 0; + ctrl->pcix_speed_capability = 0; + break; + case PCI_SUB_HPC_ID2: + /* First Pushbutton implementation */ + ctrl->push_flag = 1; + ctrl->slot_switch_type = 1; + ctrl->speed_capability = PCI_SPEED_33MHz; + ctrl->push_button = 1; + ctrl->pci_config_space = 1; + ctrl->defeature_PHP = 1; + ctrl->pcix_support = 0; + ctrl->pcix_speed_capability = 0; + break; + case PCI_SUB_HPC_ID_INTC: + /* Third party (6500/7000) */ + ctrl->slot_switch_type = 1; + ctrl->speed_capability = PCI_SPEED_33MHz; + ctrl->push_button = 0; + ctrl->pci_config_space = 1; + ctrl->defeature_PHP = 1; + ctrl->pcix_support = 0; + ctrl->pcix_speed_capability = 0; + break; + case PCI_SUB_HPC_ID3: + /* First 66 Mhz implementation */ + ctrl->push_flag = 1; + ctrl->slot_switch_type = 1; + ctrl->speed_capability = PCI_SPEED_66MHz; + ctrl->push_button = 1; + ctrl->pci_config_space = 1; + ctrl->defeature_PHP = 1; + ctrl->pcix_support = 0; + ctrl->pcix_speed_capability = 0; + break; + case PCI_SUB_HPC_ID4: + /* First PCI-X implementation, 100MHz */ + ctrl->push_flag = 1; + ctrl->slot_switch_type = 1; + ctrl->speed_capability = PCI_SPEED_100MHz_PCIX; + ctrl->push_button = 1; + ctrl->pci_config_space = 1; + ctrl->defeature_PHP = 1; + ctrl->pcix_support = 1; + ctrl->pcix_speed_capability = 0; + break; + default: + err(msg_HPC_not_supported); + rc = -ENODEV; + goto err_free_ctrl; } + break; + + case PCI_VENDOR_ID_INTEL: + /* Check for speed capability (0=33, 1=66) */ + if (subsystem_deviceid & 0x0001) + ctrl->speed_capability = PCI_SPEED_66MHz; + else + ctrl->speed_capability = PCI_SPEED_33MHz; + + /* Check for push button */ + if (subsystem_deviceid & 0x0002) + ctrl->push_button = 0; + else + ctrl->push_button = 1; + + /* Check for slot switch type (0=mechanical, 1=not mechanical) */ + if (subsystem_deviceid & 0x0004) + ctrl->slot_switch_type = 0; + else + ctrl->slot_switch_type = 1; + + /* PHP Status (0=De-feature PHP, 1=Normal operation) */ + if (subsystem_deviceid & 0x0008) + ctrl->defeature_PHP = 1; /* PHP supported */ + else + ctrl->defeature_PHP = 0; /* PHP not supported */ + + /* Alternate Base Address Register Interface + * (0=not supported, 1=supported) + */ + if (subsystem_deviceid & 0x0010) + ctrl->alternate_base_address = 1; + else + ctrl->alternate_base_address = 0; + + /* PCI Config Space Index (0=not supported, 1=supported) */ + if (subsystem_deviceid & 0x0020) + ctrl->pci_config_space = 1; + else + ctrl->pci_config_space = 0; + + /* PCI-X support */ + if (subsystem_deviceid & 0x0080) { + ctrl->pcix_support = 1; + if (subsystem_deviceid & 0x0040) + /* 133MHz PCI-X if bit 7 is 1 */ + ctrl->pcix_speed_capability = 1; + else + /* 100MHz PCI-X if bit 7 is 1 and bit 0 is 0, */ + /* 66MHz PCI-X if bit 7 is 1 and bit 0 is 1 */ + ctrl->pcix_speed_capability = 0; + } else { + /* Conventional PCI */ + ctrl->pcix_support = 0; + ctrl->pcix_speed_capability = 0; + } + break; - } else { + default: err(msg_HPC_not_supported); - return -ENODEV; + rc = -ENODEV; + goto err_free_ctrl; } - // Tell the user that we found one. + /* Tell the user that we found one. */ info("Initializing the PCI hot plug controller residing on PCI bus %d\n", pdev->bus->number); @@ -1087,7 +1119,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (rc) { goto err_free_bus; } - + dbg("pdev = %p\n", pdev); dbg("pci resource start %llx\n", (unsigned long long)pci_resource_start(pdev, 0)); dbg("pci resource len %llx\n", (unsigned long long)pci_resource_len(pdev, 0)); @@ -1109,7 +1141,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_free_mem_region; } - // Check for 66Mhz operation + /* Check for 66Mhz operation */ ctrl->speed = get_controller_speed(ctrl); @@ -1120,7 +1152,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) * ********************************************************/ - // find the physical slot number of the first hot plug slot + /* find the physical slot number of the first hot plug slot */ /* Get slot won't work for devices behind bridges, but * in this case it will always be called for the "base" @@ -1137,7 +1169,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_iounmap; } - // Store PCI Config Space for all devices on this bus + /* Store PCI Config Space for all devices on this bus */ rc = cpqhp_save_config(ctrl, ctrl->bus, readb(ctrl->hpc_reg + SLOT_MASK)); if (rc) { err("%s: unable to save PCI configuration data, error %d\n", @@ -1148,7 +1180,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) /* * Get IO, memory, and IRQ resources for new devices */ - // The next line is required for cpqhp_find_available_resources + /* The next line is required for cpqhp_find_available_resources */ ctrl->interrupt = pdev->irq; if (ctrl->interrupt < 0x10) { cpqhp_legacy_mode = 1; @@ -1182,7 +1214,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) __func__, rc); goto err_iounmap; } - + /* Mask all general input interrupts */ writel(0xFFFFFFFFL, ctrl->hpc_reg + INT_MASK); @@ -1196,12 +1228,14 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_iounmap; } - /* Enable Shift Out interrupt and clear it, also enable SERR on power fault */ + /* Enable Shift Out interrupt and clear it, also enable SERR on power + * fault + */ temp_word = readw(ctrl->hpc_reg + MISC); temp_word |= 0x4006; writew(temp_word, ctrl->hpc_reg + MISC); - // Changed 05/05/97 to clear all interrupts at start + /* Changed 05/05/97 to clear all interrupts at start */ writel(0xFFFFFFFFL, ctrl->hpc_reg + INT_INPUT_CLEAR); ctrl->ctrl_int_comp = readl(ctrl->hpc_reg + INT_INPUT_CLEAR); @@ -1216,13 +1250,14 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) cpqhp_ctrl_list = ctrl; } - // turn off empty slots here unless command line option "ON" set - // Wait for exclusive access to hardware + /* turn off empty slots here unless command line option "ON" set + * Wait for exclusive access to hardware + */ mutex_lock(&ctrl->crit_sect); num_of_slots = readb(ctrl->hpc_reg + SLOT_MASK) & 0x0F; - // find first device number for the ctrl + /* find first device number for the ctrl */ device = readb(ctrl->hpc_reg + SLOT_MASK) >> 4; while (num_of_slots) { @@ -1234,23 +1269,21 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) hp_slot = func->device - ctrl->slot_device_offset; dbg("hp_slot: %d\n", hp_slot); - // We have to save the presence info for these slots + /* We have to save the presence info for these slots */ temp_word = ctrl->ctrl_int_comp >> 16; func->presence_save = (temp_word >> hp_slot) & 0x01; func->presence_save |= (temp_word >> (hp_slot + 7)) & 0x02; - if (ctrl->ctrl_int_comp & (0x1L << hp_slot)) { + if (ctrl->ctrl_int_comp & (0x1L << hp_slot)) func->switch_save = 0; - } else { + else func->switch_save = 0x10; - } - if (!power_mode) { + if (!power_mode) if (!func->is_a_board) { green_LED_off(ctrl, hp_slot); slot_disable(ctrl, hp_slot); } - } device++; num_of_slots--; @@ -1258,7 +1291,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (!power_mode) { set_SOGO(ctrl); - // Wait for SOBS to be unset + /* Wait for SOBS to be unset */ wait_for_ctrl_irq(ctrl); } @@ -1269,7 +1302,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_free_irq; } - // Done with exclusive hardware access + /* Done with exclusive hardware access */ mutex_unlock(&ctrl->crit_sect); cpqhp_create_debugfs_files(ctrl); @@ -1291,77 +1324,6 @@ err_disable_device: return rc; } - -static int one_time_init(void) -{ - int loop; - int retval = 0; - - if (initialized) - return 0; - - power_mode = 0; - - retval = pci_print_IRQ_route(); - if (retval) - goto error; - - dbg("Initialize + Start the notification mechanism \n"); - - retval = cpqhp_event_start_thread(); - if (retval) - goto error; - - dbg("Initialize slot lists\n"); - for (loop = 0; loop < 256; loop++) { - cpqhp_slot_list[loop] = NULL; - } - - // FIXME: We also need to hook the NMI handler eventually. - // this also needs to be worked with Christoph - // register_NMI_handler(); - - // Map rom address - cpqhp_rom_start = ioremap(ROM_PHY_ADDR, ROM_PHY_LEN); - if (!cpqhp_rom_start) { - err ("Could not ioremap memory region for ROM\n"); - retval = -EIO; - goto error; - } - - /* Now, map the int15 entry point if we are on compaq specific hardware */ - compaq_nvram_init(cpqhp_rom_start); - - /* Map smbios table entry point structure */ - smbios_table = detect_SMBIOS_pointer(cpqhp_rom_start, - cpqhp_rom_start + ROM_PHY_LEN); - if (!smbios_table) { - err ("Could not find the SMBIOS pointer in memory\n"); - retval = -EIO; - goto error_rom_start; - } - - smbios_start = ioremap(readl(smbios_table + ST_ADDRESS), - readw(smbios_table + ST_LENGTH)); - if (!smbios_start) { - err ("Could not ioremap memory region taken from SMBIOS values\n"); - retval = -EIO; - goto error_smbios_start; - } - - initialized = 1; - - return retval; - -error_smbios_start: - iounmap(smbios_start); -error_rom_start: - iounmap(cpqhp_rom_start); -error: - return retval; -} - - static void __exit unload_cpqphpd(void) { struct pci_func *next; @@ -1381,10 +1343,10 @@ static void __exit unload_cpqphpd(void) if (ctrl->hpc_reg) { u16 misc; rc = read_slot_enable (ctrl); - + writeb(0, ctrl->hpc_reg + SLOT_SERR); writel(0xFFFFFFC0L | ~rc, ctrl->hpc_reg + INT_MASK); - + misc = readw(ctrl->hpc_reg + MISC); misc &= 0xFFFD; writew(misc, ctrl->hpc_reg + MISC); @@ -1464,38 +1426,34 @@ static void __exit unload_cpqphpd(void) } } - // Stop the notification mechanism + /* Stop the notification mechanism */ if (initialized) cpqhp_event_stop_thread(); - //unmap the rom address + /* unmap the rom address */ if (cpqhp_rom_start) iounmap(cpqhp_rom_start); if (smbios_start) iounmap(smbios_start); } - - static struct pci_device_id hpcd_pci_tbl[] = { { /* handle any PCI Hotplug controller */ .class = ((PCI_CLASS_SYSTEM_PCI_HOTPLUG << 8) | 0x00), .class_mask = ~0, - + /* no matter who makes it */ .vendor = PCI_ANY_ID, .device = PCI_ANY_ID, .subvendor = PCI_ANY_ID, .subdevice = PCI_ANY_ID, - + }, { /* end: all zeroes */ } }; MODULE_DEVICE_TABLE(pci, hpcd_pci_tbl); - - static struct pci_driver cpqhpc_driver = { .name = "compaq_pci_hotplug", .id_table = hpcd_pci_tbl, @@ -1503,8 +1461,6 @@ static struct pci_driver cpqhpc_driver = { /* remove: cpqhpc_remove_one, */ }; - - static int __init cpqhpc_init(void) { int result; @@ -1518,7 +1474,6 @@ static int __init cpqhpc_init(void) return result; } - static void __exit cpqhpc_cleanup(void) { dbg("unload_cpqphpd()\n"); @@ -1529,8 +1484,5 @@ static void __exit cpqhpc_cleanup(void) cpqhp_shutdown_debugfs(); } - module_init(cpqhpc_init); module_exit(cpqhpc_cleanup); - - diff --git a/drivers/pci/hotplug/cpqphp_ctrl.c b/drivers/pci/hotplug/cpqphp_ctrl.c index cc227a8c4b1..0ff689afa75 100644 --- a/drivers/pci/hotplug/cpqphp_ctrl.c +++ b/drivers/pci/hotplug/cpqphp_ctrl.c @@ -34,7 +34,6 @@ #include <linux/interrupt.h> #include <linux/delay.h> #include <linux/wait.h> -#include <linux/smp_lock.h> #include <linux/pci.h> #include <linux/pci_hotplug.h> #include <linux/kthread.h> @@ -81,14 +80,15 @@ static u8 handle_switch_change(u8 change, struct controller * ctrl) for (hp_slot = 0; hp_slot < 6; hp_slot++) { if (change & (0x1L << hp_slot)) { - /********************************** + /* * this one changed. - **********************************/ + */ func = cpqhp_slot_find(ctrl->bus, (hp_slot + ctrl->slot_device_offset), 0); /* this is the structure that tells the worker thread - *what to do */ + * what to do + */ taskInfo = &(ctrl->event_queue[ctrl->next_event]); ctrl->next_event = (ctrl->next_event + 1) % 10; taskInfo->hp_slot = hp_slot; @@ -100,17 +100,17 @@ static u8 handle_switch_change(u8 change, struct controller * ctrl) func->presence_save |= (temp_word >> (hp_slot + 7)) & 0x02; if (ctrl->ctrl_int_comp & (0x1L << hp_slot)) { - /********************************** + /* * Switch opened - **********************************/ + */ func->switch_save = 0; taskInfo->event_type = INT_SWITCH_OPEN; } else { - /********************************** + /* * Switch closed - **********************************/ + */ func->switch_save = 0x10; @@ -131,9 +131,8 @@ static struct slot *cpqhp_find_slot(struct controller *ctrl, u8 device) { struct slot *slot = ctrl->slot; - while (slot && (slot->device != device)) { + while (slot && (slot->device != device)) slot = slot->next; - } return slot; } @@ -152,17 +151,17 @@ static u8 handle_presence_change(u16 change, struct controller * ctrl) if (!change) return 0; - /********************************** + /* * Presence Change - **********************************/ + */ dbg("cpqsbd: Presence/Notify input change.\n"); dbg(" Changed bits are 0x%4.4x\n", change ); for (hp_slot = 0; hp_slot < 6; hp_slot++) { if (change & (0x0101 << hp_slot)) { - /********************************** + /* * this one changed. - **********************************/ + */ func = cpqhp_slot_find(ctrl->bus, (hp_slot + ctrl->slot_device_offset), 0); @@ -177,22 +176,23 @@ static u8 handle_presence_change(u16 change, struct controller * ctrl) return 0; /* If the switch closed, must be a button - * If not in button mode, nevermind */ + * If not in button mode, nevermind + */ if (func->switch_save && (ctrl->push_button == 1)) { temp_word = ctrl->ctrl_int_comp >> 16; temp_byte = (temp_word >> hp_slot) & 0x01; temp_byte |= (temp_word >> (hp_slot + 7)) & 0x02; if (temp_byte != func->presence_save) { - /************************************** + /* * button Pressed (doesn't do anything) - **************************************/ + */ dbg("hp_slot %d button pressed\n", hp_slot); taskInfo->event_type = INT_BUTTON_PRESS; } else { - /********************************** + /* * button Released - TAKE ACTION!!!! - **********************************/ + */ dbg("hp_slot %d button released\n", hp_slot); taskInfo->event_type = INT_BUTTON_RELEASE; @@ -210,7 +210,8 @@ static u8 handle_presence_change(u16 change, struct controller * ctrl) } } else { /* Switch is open, assume a presence change - * Save the presence state */ + * Save the presence state + */ temp_word = ctrl->ctrl_int_comp >> 16; func->presence_save = (temp_word >> hp_slot) & 0x01; func->presence_save |= (temp_word >> (hp_slot + 7)) & 0x02; @@ -241,17 +242,17 @@ static u8 handle_power_fault(u8 change, struct controller * ctrl) if (!change) return 0; - /********************************** + /* * power fault - **********************************/ + */ info("power fault interrupt\n"); for (hp_slot = 0; hp_slot < 6; hp_slot++) { if (change & (0x01 << hp_slot)) { - /********************************** + /* * this one changed. - **********************************/ + */ func = cpqhp_slot_find(ctrl->bus, (hp_slot + ctrl->slot_device_offset), 0); @@ -262,16 +263,16 @@ static u8 handle_power_fault(u8 change, struct controller * ctrl) rc++; if (ctrl->ctrl_int_comp & (0x00000100 << hp_slot)) { - /********************************** + /* * power fault Cleared - **********************************/ + */ func->status = 0x00; taskInfo->event_type = INT_POWER_FAULT_CLEAR; } else { - /********************************** + /* * power fault - **********************************/ + */ taskInfo->event_type = INT_POWER_FAULT; if (ctrl->rev < 4) { @@ -432,13 +433,15 @@ static struct pci_resource *do_pre_bridge_resource_split(struct pci_resource **h /* If we got here, there the bridge requires some of the resource, but - * we may be able to split some off of the front */ + * we may be able to split some off of the front + */ node = *head; if (node->length & (alignment -1)) { /* this one isn't an aligned length, so we'll make a new entry - * and split it up. */ + * and split it up. + */ split_node = kmalloc(sizeof(*split_node), GFP_KERNEL); if (!split_node) @@ -544,10 +547,10 @@ static struct pci_resource *get_io_resource(struct pci_resource **head, u32 size if (!(*head)) return NULL; - if ( cpqhp_resource_sort_and_combine(head) ) + if (cpqhp_resource_sort_and_combine(head)) return NULL; - if ( sort_by_size(head) ) + if (sort_by_size(head)) return NULL; for (node = *head; node; node = node->next) { @@ -556,7 +559,8 @@ static struct pci_resource *get_io_resource(struct pci_resource **head, u32 size if (node->base & (size - 1)) { /* this one isn't base aligned properly - * so we'll make a new entry and split it up */ + * so we'll make a new entry and split it up + */ temp_dword = (node->base | (size-1)) + 1; /* Short circuit if adjusted size is too small */ @@ -581,7 +585,8 @@ static struct pci_resource *get_io_resource(struct pci_resource **head, u32 size /* Don't need to check if too small since we already did */ if (node->length > size) { /* this one is longer than we need - * so we'll make a new entry and split it up */ + * so we'll make a new entry and split it up + */ split_node = kmalloc(sizeof(*split_node), GFP_KERNEL); if (!split_node) @@ -601,7 +606,8 @@ static struct pci_resource *get_io_resource(struct pci_resource **head, u32 size continue; /* If we got here, then it is the right size - * Now take it out of the list and break */ + * Now take it out of the list and break + */ if (*head == node) { *head = node->next; } else { @@ -642,14 +648,16 @@ static struct pci_resource *get_max_resource(struct pci_resource **head, u32 siz return NULL; for (max = *head; max; max = max->next) { - /* If not big enough we could probably just bail, - * instead we'll continue to the next. */ + /* If not big enough we could probably just bail, + * instead we'll continue to the next. + */ if (max->length < size) continue; if (max->base & (size - 1)) { /* this one isn't base aligned properly - * so we'll make a new entry and split it up */ + * so we'll make a new entry and split it up + */ temp_dword = (max->base | (size-1)) + 1; /* Short circuit if adjusted size is too small */ @@ -672,7 +680,8 @@ static struct pci_resource *get_max_resource(struct pci_resource **head, u32 siz if ((max->base + max->length) & (size - 1)) { /* this one isn't end aligned properly at the top - * so we'll make a new entry and split it up */ + * so we'll make a new entry and split it up + */ split_node = kmalloc(sizeof(*split_node), GFP_KERNEL); if (!split_node) @@ -744,7 +753,8 @@ static struct pci_resource *get_resource(struct pci_resource **head, u32 size) if (node->base & (size - 1)) { dbg("%s: not aligned\n", __func__); /* this one isn't base aligned properly - * so we'll make a new entry and split it up */ + * so we'll make a new entry and split it up + */ temp_dword = (node->base | (size-1)) + 1; /* Short circuit if adjusted size is too small */ @@ -769,7 +779,8 @@ static struct pci_resource *get_resource(struct pci_resource **head, u32 size) if (node->length > size) { dbg("%s: too big\n", __func__); /* this one is longer than we need - * so we'll make a new entry and split it up */ + * so we'll make a new entry and split it up + */ split_node = kmalloc(sizeof(*split_node), GFP_KERNEL); if (!split_node) @@ -886,19 +897,19 @@ irqreturn_t cpqhp_ctrl_intr(int IRQ, void *data) u32 Diff; u32 temp_dword; - + misc = readw(ctrl->hpc_reg + MISC); - /*************************************** + /* * Check to see if it was our interrupt - ***************************************/ + */ if (!(misc & 0x000C)) { return IRQ_NONE; } if (misc & 0x0004) { - /********************************** + /* * Serial Output interrupt Pending - **********************************/ + */ /* Clear the interrupt */ misc |= 0x0004; @@ -961,11 +972,8 @@ struct pci_func *cpqhp_slot_create(u8 busnumber) struct pci_func *next; new_slot = kzalloc(sizeof(*new_slot), GFP_KERNEL); - if (new_slot == NULL) { - /* I'm not dead yet! - * You will be. */ + if (new_slot == NULL) return new_slot; - } new_slot->next = NULL; new_slot->configured = 1; @@ -996,10 +1004,8 @@ static int slot_remove(struct pci_func * old_slot) return 1; next = cpqhp_slot_list[old_slot->bus]; - - if (next == NULL) { + if (next == NULL) return 1; - } if (next == old_slot) { cpqhp_slot_list[old_slot->bus] = old_slot->next; @@ -1008,9 +1014,8 @@ static int slot_remove(struct pci_func * old_slot) return 0; } - while ((next->next != old_slot) && (next->next != NULL)) { + while ((next->next != old_slot) && (next->next != NULL)) next = next->next; - } if (next->next == old_slot) { next->next = old_slot->next; @@ -1040,9 +1045,8 @@ static int bridge_slot_remove(struct pci_func *bridge) for (tempBus = secondaryBus; tempBus <= subordinateBus; tempBus++) { next = cpqhp_slot_list[tempBus]; - while (!slot_remove(next)) { + while (!slot_remove(next)) next = cpqhp_slot_list[tempBus]; - } } next = cpqhp_slot_list[bridge->bus]; @@ -1130,39 +1134,43 @@ static u8 set_controller_speed(struct controller *ctrl, u8 adapter_speed, u8 hp_ u8 slot_power = readb(ctrl->hpc_reg + SLOT_POWER); u16 reg16; u32 leds = readl(ctrl->hpc_reg + LED_CONTROL); - + if (ctrl->speed == adapter_speed) return 0; - + /* We don't allow freq/mode changes if we find another adapter running - * in another slot on this controller */ + * in another slot on this controller + */ for(slot = ctrl->slot; slot; slot = slot->next) { - if (slot->device == (hp_slot + ctrl->slot_device_offset)) + if (slot->device == (hp_slot + ctrl->slot_device_offset)) continue; if (!slot->hotplug_slot || !slot->hotplug_slot->info) continue; - if (slot->hotplug_slot->info->adapter_status == 0) + if (slot->hotplug_slot->info->adapter_status == 0) continue; /* If another adapter is running on the same segment but at a * lower speed/mode, we allow the new adapter to function at - * this rate if supported */ - if (ctrl->speed < adapter_speed) + * this rate if supported + */ + if (ctrl->speed < adapter_speed) return 0; return 1; } - + /* If the controller doesn't support freq/mode changes and the - * controller is running at a higher mode, we bail */ + * controller is running at a higher mode, we bail + */ if ((ctrl->speed > adapter_speed) && (!ctrl->pcix_speed_capability)) return 1; - + /* But we allow the adapter to run at a lower rate if possible */ if ((ctrl->speed < adapter_speed) && (!ctrl->pcix_speed_capability)) return 0; /* We try to set the max speed supported by both the adapter and - * controller */ + * controller + */ if (ctrl->speed_capability < adapter_speed) { if (ctrl->speed == ctrl->speed_capability) return 0; @@ -1171,22 +1179,22 @@ static u8 set_controller_speed(struct controller *ctrl, u8 adapter_speed, u8 hp_ writel(0x0L, ctrl->hpc_reg + LED_CONTROL); writeb(0x00, ctrl->hpc_reg + SLOT_ENABLE); - - set_SOGO(ctrl); + + set_SOGO(ctrl); wait_for_ctrl_irq(ctrl); - + if (adapter_speed != PCI_SPEED_133MHz_PCIX) reg = 0xF5; else - reg = 0xF4; + reg = 0xF4; pci_write_config_byte(ctrl->pci_dev, 0x41, reg); - + reg16 = readw(ctrl->hpc_reg + NEXT_CURR_FREQ); reg16 &= ~0x000F; switch(adapter_speed) { - case(PCI_SPEED_133MHz_PCIX): + case(PCI_SPEED_133MHz_PCIX): reg = 0x75; - reg16 |= 0xB; + reg16 |= 0xB; break; case(PCI_SPEED_100MHz_PCIX): reg = 0x74; @@ -1203,48 +1211,48 @@ static u8 set_controller_speed(struct controller *ctrl, u8 adapter_speed, u8 hp_ default: /* 33MHz PCI 2.2 */ reg = 0x71; break; - + } reg16 |= 0xB << 12; writew(reg16, ctrl->hpc_reg + NEXT_CURR_FREQ); - - mdelay(5); - + + mdelay(5); + /* Reenable interrupts */ writel(0, ctrl->hpc_reg + INT_MASK); - pci_write_config_byte(ctrl->pci_dev, 0x41, reg); - + pci_write_config_byte(ctrl->pci_dev, 0x41, reg); + /* Restart state machine */ reg = ~0xF; pci_read_config_byte(ctrl->pci_dev, 0x43, ®); pci_write_config_byte(ctrl->pci_dev, 0x43, reg); - + /* Only if mode change...*/ if (((ctrl->speed == PCI_SPEED_66MHz) && (adapter_speed == PCI_SPEED_66MHz_PCIX)) || ((ctrl->speed == PCI_SPEED_66MHz_PCIX) && (adapter_speed == PCI_SPEED_66MHz))) set_SOGO(ctrl); - + wait_for_ctrl_irq(ctrl); mdelay(1100); - + /* Restore LED/Slot state */ writel(leds, ctrl->hpc_reg + LED_CONTROL); writeb(slot_power, ctrl->hpc_reg + SLOT_ENABLE); - + set_SOGO(ctrl); wait_for_ctrl_irq(ctrl); ctrl->speed = adapter_speed; slot = cpqhp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset); - info("Successfully changed frequency/mode for adapter in slot %d\n", + info("Successfully changed frequency/mode for adapter in slot %d\n", slot->number); return 0; } -/* the following routines constitute the bulk of the - hotplug controller logic +/* the following routines constitute the bulk of the + * hotplug controller logic */ @@ -1268,17 +1276,17 @@ static u32 board_replaced(struct pci_func *func, struct controller *ctrl) hp_slot = func->device - ctrl->slot_device_offset; - if (readl(ctrl->hpc_reg + INT_INPUT_CLEAR) & (0x01L << hp_slot)) { - /********************************** - * The switch is open. - **********************************/ + /* + * The switch is open. + */ + if (readl(ctrl->hpc_reg + INT_INPUT_CLEAR) & (0x01L << hp_slot)) rc = INTERLOCK_OPEN; - } else if (is_slot_enabled (ctrl, hp_slot)) { - /********************************** - * The board is already on - **********************************/ + /* + * The board is already on + */ + else if (is_slot_enabled (ctrl, hp_slot)) rc = CARD_FUNCTIONING; - } else { + else { mutex_lock(&ctrl->crit_sect); /* turn on board without attaching to the bus */ @@ -1299,7 +1307,7 @@ static u32 board_replaced(struct pci_func *func, struct controller *ctrl) /* Wait for SOBS to be unset */ wait_for_ctrl_irq (ctrl); - + adapter_speed = get_adapter_speed(ctrl, hp_slot); if (ctrl->speed != adapter_speed) if (set_controller_speed(ctrl, adapter_speed, hp_slot)) @@ -1352,7 +1360,8 @@ static u32 board_replaced(struct pci_func *func, struct controller *ctrl) * Get slot won't work for devices behind * bridges, but in this case it will always be * called for the "base" bus/dev/func of an - * adapter. */ + * adapter. + */ mutex_lock(&ctrl->crit_sect); @@ -1377,7 +1386,8 @@ static u32 board_replaced(struct pci_func *func, struct controller *ctrl) * Get slot won't work for devices behind bridges, but * in this case it will always be called for the "base" - * bus/dev/func of an adapter. */ + * bus/dev/func of an adapter. + */ mutex_lock(&ctrl->crit_sect); @@ -1434,7 +1444,8 @@ static u32 board_added(struct pci_func *func, struct controller *ctrl) wait_for_ctrl_irq (ctrl); /* Change bits in slot power register to force another shift out - * NOTE: this is to work around the timer bug */ + * NOTE: this is to work around the timer bug + */ temp_byte = readb(ctrl->hpc_reg + SLOT_POWER); writeb(0x00, ctrl->hpc_reg + SLOT_POWER); writeb(temp_byte, ctrl->hpc_reg + SLOT_POWER); @@ -1443,12 +1454,12 @@ static u32 board_added(struct pci_func *func, struct controller *ctrl) /* Wait for SOBS to be unset */ wait_for_ctrl_irq (ctrl); - + adapter_speed = get_adapter_speed(ctrl, hp_slot); if (ctrl->speed != adapter_speed) if (set_controller_speed(ctrl, adapter_speed, hp_slot)) rc = WRONG_BUS_FREQUENCY; - + /* turn off board without attaching to the bus */ disable_slot_power (ctrl, hp_slot); @@ -1461,7 +1472,7 @@ static u32 board_added(struct pci_func *func, struct controller *ctrl) if (rc) return rc; - + p_slot = cpqhp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset); /* turn on board and blink green LED */ @@ -1521,7 +1532,7 @@ static u32 board_added(struct pci_func *func, struct controller *ctrl) } /* All F's is an empty slot or an invalid board */ - if (temp_register != 0xFFFFFFFF) { /* Check for a board in the slot */ + if (temp_register != 0xFFFFFFFF) { res_lists.io_head = ctrl->io_head; res_lists.mem_head = ctrl->mem_head; res_lists.p_mem_head = ctrl->p_mem_head; @@ -1570,9 +1581,8 @@ static u32 board_added(struct pci_func *func, struct controller *ctrl) index = 0; do { new_slot = cpqhp_slot_find(ctrl->bus, func->device, index++); - if (new_slot && !new_slot->pci_dev) { + if (new_slot && !new_slot->pci_dev) cpqhp_configure_device(ctrl, new_slot); - } } while (new_slot); mutex_lock(&ctrl->crit_sect); @@ -1859,12 +1869,12 @@ static void interrupt_event_handler(struct controller *ctrl) info(msg_button_on, p_slot->number); } mutex_lock(&ctrl->crit_sect); - + dbg("blink green LED and turn off amber\n"); - + amber_LED_off (ctrl, hp_slot); green_LED_blink (ctrl, hp_slot); - + set_SOGO(ctrl); /* Wait for SOBS to be unset */ @@ -1958,7 +1968,7 @@ void cpqhp_pushbutton_thread(unsigned long slot) if (cpqhp_process_SI(ctrl, func) != 0) { amber_LED_on(ctrl, hp_slot); green_LED_off(ctrl, hp_slot); - + set_SOGO(ctrl); /* Wait for SOBS to be unset */ @@ -2079,7 +2089,7 @@ int cpqhp_process_SS(struct controller *ctrl, struct pci_func *func) struct pci_bus *pci_bus = ctrl->pci_bus; int physical_slot=0; - device = func->device; + device = func->device; func = cpqhp_slot_find(ctrl->bus, device, index++); p_slot = cpqhp_find_slot(ctrl, device); if (p_slot) { @@ -2113,9 +2123,8 @@ int cpqhp_process_SS(struct controller *ctrl, struct pci_func *func) /* If the VGA Enable bit is set, remove isn't * supported */ - if (BCR & PCI_BRIDGE_CTL_VGA) { + if (BCR & PCI_BRIDGE_CTL_VGA) rc = REMOVE_NOT_SUPPORTED; - } } } @@ -2183,67 +2192,67 @@ int cpqhp_hardware_test(struct controller *ctrl, int test_num) num_of_slots = readb(ctrl->hpc_reg + SLOT_MASK) & 0x0f; switch (test_num) { - case 1: - /* Do stuff here! */ - - /* Do that funky LED thing */ - /* so we can restore them later */ - save_LED = readl(ctrl->hpc_reg + LED_CONTROL); - work_LED = 0x01010101; - switch_leds(ctrl, num_of_slots, &work_LED, 0); - switch_leds(ctrl, num_of_slots, &work_LED, 1); - switch_leds(ctrl, num_of_slots, &work_LED, 0); - switch_leds(ctrl, num_of_slots, &work_LED, 1); - - work_LED = 0x01010000; - writel(work_LED, ctrl->hpc_reg + LED_CONTROL); - switch_leds(ctrl, num_of_slots, &work_LED, 0); - switch_leds(ctrl, num_of_slots, &work_LED, 1); - work_LED = 0x00000101; - writel(work_LED, ctrl->hpc_reg + LED_CONTROL); - switch_leds(ctrl, num_of_slots, &work_LED, 0); - switch_leds(ctrl, num_of_slots, &work_LED, 1); + case 1: + /* Do stuff here! */ + + /* Do that funky LED thing */ + /* so we can restore them later */ + save_LED = readl(ctrl->hpc_reg + LED_CONTROL); + work_LED = 0x01010101; + switch_leds(ctrl, num_of_slots, &work_LED, 0); + switch_leds(ctrl, num_of_slots, &work_LED, 1); + switch_leds(ctrl, num_of_slots, &work_LED, 0); + switch_leds(ctrl, num_of_slots, &work_LED, 1); + + work_LED = 0x01010000; + writel(work_LED, ctrl->hpc_reg + LED_CONTROL); + switch_leds(ctrl, num_of_slots, &work_LED, 0); + switch_leds(ctrl, num_of_slots, &work_LED, 1); + work_LED = 0x00000101; + writel(work_LED, ctrl->hpc_reg + LED_CONTROL); + switch_leds(ctrl, num_of_slots, &work_LED, 0); + switch_leds(ctrl, num_of_slots, &work_LED, 1); + + work_LED = 0x01010000; + writel(work_LED, ctrl->hpc_reg + LED_CONTROL); + for (loop = 0; loop < num_of_slots; loop++) { + set_SOGO(ctrl); - work_LED = 0x01010000; - writel(work_LED, ctrl->hpc_reg + LED_CONTROL); - for (loop = 0; loop < num_of_slots; loop++) { - set_SOGO(ctrl); + /* Wait for SOGO interrupt */ + wait_for_ctrl_irq (ctrl); - /* Wait for SOGO interrupt */ - wait_for_ctrl_irq (ctrl); + /* Get ready for next iteration */ + long_delay((3*HZ)/10); + work_LED = work_LED >> 16; + writel(work_LED, ctrl->hpc_reg + LED_CONTROL); - /* Get ready for next iteration */ - long_delay((3*HZ)/10); - work_LED = work_LED >> 16; - writel(work_LED, ctrl->hpc_reg + LED_CONTROL); - - set_SOGO(ctrl); + set_SOGO(ctrl); - /* Wait for SOGO interrupt */ - wait_for_ctrl_irq (ctrl); + /* Wait for SOGO interrupt */ + wait_for_ctrl_irq (ctrl); - /* Get ready for next iteration */ - long_delay((3*HZ)/10); - work_LED = work_LED << 16; - writel(work_LED, ctrl->hpc_reg + LED_CONTROL); - work_LED = work_LED << 1; - writel(work_LED, ctrl->hpc_reg + LED_CONTROL); - } + /* Get ready for next iteration */ + long_delay((3*HZ)/10); + work_LED = work_LED << 16; + writel(work_LED, ctrl->hpc_reg + LED_CONTROL); + work_LED = work_LED << 1; + writel(work_LED, ctrl->hpc_reg + LED_CONTROL); + } - /* put it back the way it was */ - writel(save_LED, ctrl->hpc_reg + LED_CONTROL); + /* put it back the way it was */ + writel(save_LED, ctrl->hpc_reg + LED_CONTROL); - set_SOGO(ctrl); + set_SOGO(ctrl); - /* Wait for SOBS to be unset */ - wait_for_ctrl_irq (ctrl); - break; - case 2: - /* Do other stuff here! */ - break; - case 3: - /* and more... */ - break; + /* Wait for SOBS to be unset */ + wait_for_ctrl_irq (ctrl); + break; + case 2: + /* Do other stuff here! */ + break; + case 3: + /* and more... */ + break; } return 0; } @@ -2312,9 +2321,9 @@ static u32 configure_new_device(struct controller * ctrl, struct pci_func * func while ((function < max_functions) && (!stop_it)) { pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(func->device, function), 0x00, &ID); - if (ID == 0xFFFFFFFF) { /* There's nothing there. */ + if (ID == 0xFFFFFFFF) { function++; - } else { /* There's something there */ + } else { /* Setup slot structure. */ new_slot = cpqhp_slot_create(func->bus); @@ -2339,8 +2348,8 @@ static u32 configure_new_device(struct controller * ctrl, struct pci_func * func /* - Configuration logic that involves the hotplug data structures and - their bookkeeping + * Configuration logic that involves the hotplug data structures and + * their bookkeeping */ @@ -2393,7 +2402,7 @@ static int configure_new_function(struct controller *ctrl, struct pci_func *func if (rc) return rc; - if ((temp_byte & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { /* PCI-PCI Bridge */ + if ((temp_byte & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { /* set Primary bus */ dbg("set Primary bus = %d\n", func->bus); rc = pci_bus_write_config_byte(pci_bus, devfn, PCI_PRIMARY_BUS, func->bus); @@ -2484,7 +2493,8 @@ static int configure_new_function(struct controller *ctrl, struct pci_func *func temp_resources.irqs = &irqs; /* Make copies of the nodes we are going to pass down so that - * if there is a problem,we can just use these to free resources */ + * if there is a problem,we can just use these to free resources + */ hold_bus_node = kmalloc(sizeof(*hold_bus_node), GFP_KERNEL); hold_IO_node = kmalloc(sizeof(*hold_IO_node), GFP_KERNEL); hold_mem_node = kmalloc(sizeof(*hold_mem_node), GFP_KERNEL); @@ -2556,7 +2566,8 @@ static int configure_new_function(struct controller *ctrl, struct pci_func *func temp_word = (p_mem_node->base + p_mem_node->length - 1) >> 16; rc = pci_bus_write_config_word (pci_bus, devfn, PCI_PREF_MEMORY_LIMIT, temp_word); - /* Adjust this to compensate for extra adjustment in first loop */ + /* Adjust this to compensate for extra adjustment in first loop + */ irqs.barber_pole--; rc = 0; @@ -2917,27 +2928,26 @@ static int configure_new_function(struct controller *ctrl, struct pci_func *func } /* End of base register loop */ if (cpqhp_legacy_mode) { /* Figure out which interrupt pin this function uses */ - rc = pci_bus_read_config_byte (pci_bus, devfn, + rc = pci_bus_read_config_byte (pci_bus, devfn, PCI_INTERRUPT_PIN, &temp_byte); /* If this function needs an interrupt and we are behind * a bridge and the pin is tied to something that's * alread mapped, set this one the same */ - if (temp_byte && resources->irqs && - (resources->irqs->valid_INT & + if (temp_byte && resources->irqs && + (resources->irqs->valid_INT & (0x01 << ((temp_byte + resources->irqs->barber_pole - 1) & 0x03)))) { /* We have to share with something already set up */ - IRQ = resources->irqs->interrupt[(temp_byte + + IRQ = resources->irqs->interrupt[(temp_byte + resources->irqs->barber_pole - 1) & 0x03]; } else { /* Program IRQ based on card type */ rc = pci_bus_read_config_byte (pci_bus, devfn, 0x0B, &class_code); - if (class_code == PCI_BASE_CLASS_STORAGE) { + if (class_code == PCI_BASE_CLASS_STORAGE) IRQ = cpqhp_disk_irq; - } else { + else IRQ = cpqhp_nic_irq; - } } /* IRQ Line */ diff --git a/drivers/pci/hotplug/cpqphp_nvram.c b/drivers/pci/hotplug/cpqphp_nvram.c index cb174888002..76ba8a1c774 100644 --- a/drivers/pci/hotplug/cpqphp_nvram.c +++ b/drivers/pci/hotplug/cpqphp_nvram.c @@ -94,12 +94,13 @@ static u8 evbuffer[1024]; static void __iomem *compaq_int15_entry_point; -static spinlock_t int15_lock; /* lock for ordering int15_bios_call() */ +/* lock for ordering int15_bios_call() */ +static spinlock_t int15_lock; /* This is a series of function that deals with - setting & getting the hotplug resource table in some environment variable. -*/ + * setting & getting the hotplug resource table in some environment variable. + */ /* * We really shouldn't be doing this unless there is a _very_ good reason to!!! @@ -113,7 +114,7 @@ static u32 add_byte( u32 **p_buffer, u8 value, u32 *used, u32 *avail) if ((*used + 1) > *avail) return(1); - + *((u8*)*p_buffer) = value; tByte = (u8**)p_buffer; (*tByte)++; @@ -170,10 +171,10 @@ static u32 access_EV (u16 operation, u8 *ev_name, u8 *buffer, u32 *buf_size) unsigned long flags; int op = operation; int ret_val; - + if (!compaq_int15_entry_point) return -ENODEV; - + spin_lock_irqsave(&int15_lock, flags); __asm__ ( "xorl %%ebx,%%ebx\n" \ @@ -187,7 +188,7 @@ static u32 access_EV (u16 operation, u8 *ev_name, u8 *buffer, u32 *buf_size) "D" (buffer), "m" (compaq_int15_entry_point) : "%ebx", "%edx"); spin_unlock_irqrestore(&int15_lock, flags); - + return((ret_val & 0xFF00) >> 8); } @@ -210,14 +211,16 @@ static int load_HRT (void __iomem *rom_start) available = 1024; - // Now load the EV + /* Now load the EV */ temp_dword = available; rc = access_EV(READ_EV, "CQTHPS", evbuffer, &temp_dword); evbuffer_length = temp_dword; - // We're maintaining the resource lists so write FF to invalidate old info + /* We're maintaining the resource lists so write FF to invalidate old + * info + */ temp_dword = 1; rc = access_EV(WRITE_EV, "CQTHPS", &temp_byte, &temp_dword); @@ -263,13 +266,13 @@ static u32 store_HRT (void __iomem *rom_start) p_EV_header = (struct ev_hrt_header *) pFill; ctrl = cpqhp_ctrl_list; - - // The revision of this structure + + /* The revision of this structure */ rc = add_byte( &pFill, 1 + ctrl->push_flag, &usedbytes, &available); if (rc) return(rc); - // The number of controllers + /* The number of controllers */ rc = add_byte( &pFill, 1, &usedbytes, &available); if (rc) return(rc); @@ -279,27 +282,27 @@ static u32 store_HRT (void __iomem *rom_start) numCtrl++; - // The bus number + /* The bus number */ rc = add_byte( &pFill, ctrl->bus, &usedbytes, &available); if (rc) return(rc); - // The device Number + /* The device Number */ rc = add_byte( &pFill, PCI_SLOT(ctrl->pci_dev->devfn), &usedbytes, &available); if (rc) return(rc); - // The function Number + /* The function Number */ rc = add_byte( &pFill, PCI_FUNC(ctrl->pci_dev->devfn), &usedbytes, &available); if (rc) return(rc); - // Skip the number of available entries + /* Skip the number of available entries */ rc = add_dword( &pFill, 0, &usedbytes, &available); if (rc) return(rc); - // Figure out memory Available + /* Figure out memory Available */ resNode = ctrl->mem_head; @@ -308,12 +311,12 @@ static u32 store_HRT (void __iomem *rom_start) while (resNode) { loop ++; - // base + /* base */ rc = add_dword( &pFill, resNode->base, &usedbytes, &available); if (rc) return(rc); - // length + /* length */ rc = add_dword( &pFill, resNode->length, &usedbytes, &available); if (rc) return(rc); @@ -321,10 +324,10 @@ static u32 store_HRT (void __iomem *rom_start) resNode = resNode->next; } - // Fill in the number of entries + /* Fill in the number of entries */ p_ev_ctrl->mem_avail = loop; - // Figure out prefetchable memory Available + /* Figure out prefetchable memory Available */ resNode = ctrl->p_mem_head; @@ -333,12 +336,12 @@ static u32 store_HRT (void __iomem *rom_start) while (resNode) { loop ++; - // base + /* base */ rc = add_dword( &pFill, resNode->base, &usedbytes, &available); if (rc) return(rc); - // length + /* length */ rc = add_dword( &pFill, resNode->length, &usedbytes, &available); if (rc) return(rc); @@ -346,10 +349,10 @@ static u32 store_HRT (void __iomem *rom_start) resNode = resNode->next; } - // Fill in the number of entries + /* Fill in the number of entries */ p_ev_ctrl->p_mem_avail = loop; - // Figure out IO Available + /* Figure out IO Available */ resNode = ctrl->io_head; @@ -358,12 +361,12 @@ static u32 store_HRT (void __iomem *rom_start) while (resNode) { loop ++; - // base + /* base */ rc = add_dword( &pFill, resNode->base, &usedbytes, &available); if (rc) return(rc); - // length + /* length */ rc = add_dword( &pFill, resNode->length, &usedbytes, &available); if (rc) return(rc); @@ -371,10 +374,10 @@ static u32 store_HRT (void __iomem *rom_start) resNode = resNode->next; } - // Fill in the number of entries + /* Fill in the number of entries */ p_ev_ctrl->io_avail = loop; - // Figure out bus Available + /* Figure out bus Available */ resNode = ctrl->bus_head; @@ -383,12 +386,12 @@ static u32 store_HRT (void __iomem *rom_start) while (resNode) { loop ++; - // base + /* base */ rc = add_dword( &pFill, resNode->base, &usedbytes, &available); if (rc) return(rc); - // length + /* length */ rc = add_dword( &pFill, resNode->length, &usedbytes, &available); if (rc) return(rc); @@ -396,15 +399,15 @@ static u32 store_HRT (void __iomem *rom_start) resNode = resNode->next; } - // Fill in the number of entries + /* Fill in the number of entries */ p_ev_ctrl->bus_avail = loop; ctrl = ctrl->next; } - + p_EV_header->num_of_ctrl = numCtrl; - // Now store the EV + /* Now store the EV */ temp_dword = usedbytes; @@ -449,20 +452,21 @@ int compaq_nvram_load (void __iomem *rom_start, struct controller *ctrl) struct ev_hrt_header *p_EV_header; if (!evbuffer_init) { - // Read the resource list information in from NVRAM + /* Read the resource list information in from NVRAM */ if (load_HRT(rom_start)) memset (evbuffer, 0, 1024); evbuffer_init = 1; } - // If we saved information in NVRAM, use it now + /* If we saved information in NVRAM, use it now */ p_EV_header = (struct ev_hrt_header *) evbuffer; - // The following code is for systems where version 1.0 of this - // driver has been loaded, but doesn't support the hardware. - // In that case, the driver would incorrectly store something - // in NVRAM. + /* The following code is for systems where version 1.0 of this + * driver has been loaded, but doesn't support the hardware. + * In that case, the driver would incorrectly store something + * in NVRAM. + */ if ((p_EV_header->Version == 2) || ((p_EV_header->Version == 1) && !ctrl->push_flag)) { p_byte = &(p_EV_header->next); @@ -479,7 +483,7 @@ int compaq_nvram_load (void __iomem *rom_start, struct controller *ctrl) function = p_ev_ctrl->function; while ((bus != ctrl->bus) || - (device != PCI_SLOT(ctrl->pci_dev->devfn)) || + (device != PCI_SLOT(ctrl->pci_dev->devfn)) || (function != PCI_FUNC(ctrl->pci_dev->devfn))) { nummem = p_ev_ctrl->mem_avail; numpmem = p_ev_ctrl->p_mem_avail; @@ -491,7 +495,7 @@ int compaq_nvram_load (void __iomem *rom_start, struct controller *ctrl) if (p_byte > ((u8*)p_EV_header + evbuffer_length)) return 2; - // Skip forward to the next entry + /* Skip forward to the next entry */ p_byte += (nummem + numpmem + numio + numbus) * 8; if (p_byte > ((u8*)p_EV_header + evbuffer_length)) @@ -629,8 +633,9 @@ int compaq_nvram_load (void __iomem *rom_start, struct controller *ctrl) ctrl->bus_head = bus_node; } - // If all of the following fail, we don't have any resources for - // hot plug add + /* If all of the following fail, we don't have any resources for + * hot plug add + */ rc = 1; rc &= cpqhp_resource_sort_and_combine(&(ctrl->mem_head)); rc &= cpqhp_resource_sort_and_combine(&(ctrl->p_mem_head)); @@ -640,14 +645,14 @@ int compaq_nvram_load (void __iomem *rom_start, struct controller *ctrl) if (rc) return(rc); } else { - if ((evbuffer[0] != 0) && (!ctrl->push_flag)) + if ((evbuffer[0] != 0) && (!ctrl->push_flag)) return 1; } return 0; } - + int compaq_nvram_store (void __iomem *rom_start) { int rc = 1; diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c index 6c0ed0fcb8e..6173b9a4544 100644 --- a/drivers/pci/hotplug/cpqphp_pci.c +++ b/drivers/pci/hotplug/cpqphp_pci.c @@ -37,7 +37,6 @@ #include "../pci.h" #include "cpqphp.h" #include "cpqphp_nvram.h" -#include <asm/pci_x86.h> u8 cpqhp_nic_irq; @@ -82,14 +81,14 @@ static void __iomem *detect_HRT_floating_pointer(void __iomem *begin, void __iom } -int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func) +int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func) { unsigned char bus; struct pci_bus *child; int num; if (func->pci_dev == NULL) - func->pci_dev = pci_find_slot(func->bus, PCI_DEVFN(func->device, func->function)); + func->pci_dev = pci_get_bus_and_slot(func->bus,PCI_DEVFN(func->device, func->function)); /* No pci device, we need to create it then */ if (func->pci_dev == NULL) { @@ -99,7 +98,7 @@ int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func) if (num) pci_bus_add_devices(ctrl->pci_dev->bus); - func->pci_dev = pci_find_slot(func->bus, PCI_DEVFN(func->device, func->function)); + func->pci_dev = pci_get_bus_and_slot(func->bus, PCI_DEVFN(func->device, func->function)); if (func->pci_dev == NULL) { dbg("ERROR: pci_dev still null\n"); return 0; @@ -112,20 +111,24 @@ int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func) pci_do_scan_bus(child); } + pci_dev_put(func->pci_dev); + return 0; } -int cpqhp_unconfigure_device(struct pci_func* func) +int cpqhp_unconfigure_device(struct pci_func* func) { int j; - + dbg("%s: bus/dev/func = %x/%x/%x\n", __func__, func->bus, func->device, func->function); for (j=0; j<8 ; j++) { - struct pci_dev* temp = pci_find_slot(func->bus, PCI_DEVFN(func->device, j)); - if (temp) + struct pci_dev* temp = pci_get_bus_and_slot(func->bus, PCI_DEVFN(func->device, j)); + if (temp) { + pci_dev_put(temp); pci_remove_bus_device(temp); + } } return 0; } @@ -178,32 +181,22 @@ int cpqhp_set_irq (u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num) if (!rc) return !rc; - // set the Edge Level Control Register (ELCR) + /* set the Edge Level Control Register (ELCR) */ temp_word = inb(0x4d0); temp_word |= inb(0x4d1) << 8; temp_word |= 0x01 << irq_num; - // This should only be for x86 as it sets the Edge Level Control Register - outb((u8) (temp_word & 0xFF), 0x4d0); - outb((u8) ((temp_word & 0xFF00) >> 8), 0x4d1); - rc = 0; - } + /* This should only be for x86 as it sets the Edge Level + * Control Register + */ + outb((u8) (temp_word & 0xFF), 0x4d0); outb((u8) ((temp_word & + 0xFF00) >> 8), 0x4d1); rc = 0; } return rc; } -/* - * WTF??? This function isn't in the code, yet a function calls it, but the - * compiler optimizes it away? strange. Here as a placeholder to keep the - * compiler happy. - */ -static int PCI_ScanBusNonBridge (u8 bus, u8 device) -{ - return 0; -} - static int PCI_ScanBusForNonBridge(struct controller *ctrl, u8 bus_num, u8 * dev_num) { u16 tdevice; @@ -213,11 +206,11 @@ static int PCI_ScanBusForNonBridge(struct controller *ctrl, u8 bus_num, u8 * dev ctrl->pci_bus->number = bus_num; for (tdevice = 0; tdevice < 0xFF; tdevice++) { - //Scan for access first + /* Scan for access first */ if (PCI_RefinedAccessConfig(ctrl->pci_bus, tdevice, 0x08, &work) == -1) continue; dbg("Looking for nonbridge bus_num %d dev_num %d\n", bus_num, tdevice); - //Yep we got one. Not a bridge ? + /* Yep we got one. Not a bridge ? */ if ((work >> 8) != PCI_TO_PCI_BRIDGE_CLASS) { *dev_num = tdevice; dbg("found it !\n"); @@ -225,16 +218,16 @@ static int PCI_ScanBusForNonBridge(struct controller *ctrl, u8 bus_num, u8 * dev } } for (tdevice = 0; tdevice < 0xFF; tdevice++) { - //Scan for access first + /* Scan for access first */ if (PCI_RefinedAccessConfig(ctrl->pci_bus, tdevice, 0x08, &work) == -1) continue; dbg("Looking for bridge bus_num %d dev_num %d\n", bus_num, tdevice); - //Yep we got one. bridge ? + /* Yep we got one. bridge ? */ if ((work >> 8) == PCI_TO_PCI_BRIDGE_CLASS) { pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(tdevice, 0), PCI_SECONDARY_BUS, &tbus); + /* XXX: no recursion, wtf? */ dbg("Recurse on bus_num %d tdevice %d\n", tbus, tdevice); - if (PCI_ScanBusNonBridge(tbus, tdevice) == 0) - return 0; + return 0; } } @@ -244,39 +237,23 @@ static int PCI_ScanBusForNonBridge(struct controller *ctrl, u8 bus_num, u8 * dev static int PCI_GetBusDevHelper(struct controller *ctrl, u8 *bus_num, u8 *dev_num, u8 slot, u8 nobridge) { - struct irq_routing_table *PCIIRQRoutingInfoLength; - long len; - long loop; + int loop, len; u32 work; - u8 tbus, tdevice, tslot; - PCIIRQRoutingInfoLength = pcibios_get_irq_routing_table(); - if (!PCIIRQRoutingInfoLength) - return -1; - - len = (PCIIRQRoutingInfoLength->size - - sizeof(struct irq_routing_table)) / sizeof(struct irq_info); - // Make sure I got at least one entry - if (len == 0) { - kfree(PCIIRQRoutingInfoLength ); - return -1; - } - + len = cpqhp_routing_table_length(); for (loop = 0; loop < len; ++loop) { - tbus = PCIIRQRoutingInfoLength->slots[loop].bus; - tdevice = PCIIRQRoutingInfoLength->slots[loop].devfn; - tslot = PCIIRQRoutingInfoLength->slots[loop].slot; + tbus = cpqhp_routing_table->slots[loop].bus; + tdevice = cpqhp_routing_table->slots[loop].devfn; + tslot = cpqhp_routing_table->slots[loop].slot; if (tslot == slot) { *bus_num = tbus; *dev_num = tdevice; ctrl->pci_bus->number = tbus; pci_bus_read_config_dword (ctrl->pci_bus, *dev_num, PCI_VENDOR_ID, &work); - if (!nobridge || (work == 0xffffffff)) { - kfree(PCIIRQRoutingInfoLength ); + if (!nobridge || (work == 0xffffffff)) return 0; - } dbg("bus_num %d devfn %d\n", *bus_num, *dev_num); pci_bus_read_config_dword (ctrl->pci_bus, *dev_num, PCI_CLASS_REVISION, &work); @@ -287,28 +264,26 @@ static int PCI_GetBusDevHelper(struct controller *ctrl, u8 *bus_num, u8 *dev_num dbg("Scan bus for Non Bridge: bus %d\n", tbus); if (PCI_ScanBusForNonBridge(ctrl, tbus, dev_num) == 0) { *bus_num = tbus; - kfree(PCIIRQRoutingInfoLength ); return 0; } - } else { - kfree(PCIIRQRoutingInfoLength ); + } else return 0; - } - } } - kfree(PCIIRQRoutingInfoLength ); return -1; } int cpqhp_get_bus_dev (struct controller *ctrl, u8 * bus_num, u8 * dev_num, u8 slot) { - return PCI_GetBusDevHelper(ctrl, bus_num, dev_num, slot, 0); //plain (bridges allowed) + /* plain (bridges allowed) */ + return PCI_GetBusDevHelper(ctrl, bus_num, dev_num, slot, 0); } -/* More PCI configuration routines; this time centered around hotplug controller */ +/* More PCI configuration routines; this time centered around hotplug + * controller + */ /* @@ -339,12 +314,12 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug) int stop_it; int index; - // Decide which slots are supported + /* Decide which slots are supported */ if (is_hot_plug) { - //********************************* - // is_hot_plug is the slot mask - //********************************* + /* + * is_hot_plug is the slot mask + */ FirstSupported = is_hot_plug >> 4; LastSupported = FirstSupported + (is_hot_plug & 0x0F) - 1; } else { @@ -352,123 +327,127 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug) LastSupported = 0x1F; } - // Save PCI configuration space for all devices in supported slots + /* Save PCI configuration space for all devices in supported slots */ ctrl->pci_bus->number = busnumber; for (device = FirstSupported; device <= LastSupported; device++) { ID = 0xFFFFFFFF; - rc = pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(device, 0), PCI_VENDOR_ID, &ID); + rc = pci_bus_read_config_dword(ctrl->pci_bus, PCI_DEVFN(device, 0), PCI_VENDOR_ID, &ID); + + if (ID == 0xFFFFFFFF) { + if (is_hot_plug) { + /* Setup slot structure with entry for empty + * slot + */ + new_slot = cpqhp_slot_create(busnumber); + if (new_slot == NULL) + return 1; - if (ID != 0xFFFFFFFF) { // device in slot - rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, 0), 0x0B, &class_code); - if (rc) - return rc; + new_slot->bus = (u8) busnumber; + new_slot->device = (u8) device; + new_slot->function = 0; + new_slot->is_a_board = 0; + new_slot->presence_save = 0; + new_slot->switch_save = 0; + } + continue; + } - rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, 0), PCI_HEADER_TYPE, &header_type); - if (rc) - return rc; + rc = pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(device, 0), 0x0B, &class_code); + if (rc) + return rc; - // If multi-function device, set max_functions to 8 - if (header_type & 0x80) - max_functions = 8; - else - max_functions = 1; + rc = pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(device, 0), PCI_HEADER_TYPE, &header_type); + if (rc) + return rc; - function = 0; + /* If multi-function device, set max_functions to 8 */ + if (header_type & 0x80) + max_functions = 8; + else + max_functions = 1; - do { - DevError = 0; + function = 0; - if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { // P-P Bridge - // Recurse the subordinate bus - // get the subordinate bus number - rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, function), PCI_SECONDARY_BUS, &secondary_bus); - if (rc) { + do { + DevError = 0; + if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { + /* Recurse the subordinate bus + * get the subordinate bus number + */ + rc = pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(device, function), PCI_SECONDARY_BUS, &secondary_bus); + if (rc) { + return rc; + } else { + sub_bus = (int) secondary_bus; + + /* Save secondary bus cfg spc + * with this recursive call. + */ + rc = cpqhp_save_config(ctrl, sub_bus, 0); + if (rc) return rc; - } else { - sub_bus = (int) secondary_bus; - - // Save secondary bus cfg spc - // with this recursive call. - rc = cpqhp_save_config(ctrl, sub_bus, 0); - if (rc) - return rc; - ctrl->pci_bus->number = busnumber; - } + ctrl->pci_bus->number = busnumber; } + } - index = 0; + index = 0; + new_slot = cpqhp_slot_find(busnumber, device, index++); + while (new_slot && + (new_slot->function != (u8) function)) new_slot = cpqhp_slot_find(busnumber, device, index++); - while (new_slot && - (new_slot->function != (u8) function)) - new_slot = cpqhp_slot_find(busnumber, device, index++); - if (!new_slot) { - // Setup slot structure. - new_slot = cpqhp_slot_create(busnumber); - - if (new_slot == NULL) - return(1); - } - - new_slot->bus = (u8) busnumber; - new_slot->device = (u8) device; - new_slot->function = (u8) function; - new_slot->is_a_board = 1; - new_slot->switch_save = 0x10; - // In case of unsupported board - new_slot->status = DevError; - new_slot->pci_dev = pci_find_slot(new_slot->bus, (new_slot->device << 3) | new_slot->function); - - for (cloop = 0; cloop < 0x20; cloop++) { - rc = pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(device, function), cloop << 2, (u32 *) & (new_slot-> config_space [cloop])); - if (rc) - return rc; - } + if (!new_slot) { + /* Setup slot structure. */ + new_slot = cpqhp_slot_create(busnumber); + if (new_slot == NULL) + return 1; + } - function++; + new_slot->bus = (u8) busnumber; + new_slot->device = (u8) device; + new_slot->function = (u8) function; + new_slot->is_a_board = 1; + new_slot->switch_save = 0x10; + /* In case of unsupported board */ + new_slot->status = DevError; + new_slot->pci_dev = pci_get_bus_and_slot(new_slot->bus, (new_slot->device << 3) | new_slot->function); - stop_it = 0; + for (cloop = 0; cloop < 0x20; cloop++) { + rc = pci_bus_read_config_dword(ctrl->pci_bus, PCI_DEVFN(device, function), cloop << 2, (u32 *) & (new_slot-> config_space [cloop])); + if (rc) + return rc; + } - // this loop skips to the next present function - // reading in Class Code and Header type. + pci_dev_put(new_slot->pci_dev); - while ((function < max_functions)&&(!stop_it)) { - rc = pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(device, function), PCI_VENDOR_ID, &ID); - if (ID == 0xFFFFFFFF) { // nothing there. - function++; - } else { // Something there - rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, function), 0x0B, &class_code); - if (rc) - return rc; + function++; - rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, function), PCI_HEADER_TYPE, &header_type); - if (rc) - return rc; + stop_it = 0; - stop_it++; - } + /* this loop skips to the next present function + * reading in Class Code and Header type. + */ + while ((function < max_functions) && (!stop_it)) { + rc = pci_bus_read_config_dword(ctrl->pci_bus, PCI_DEVFN(device, function), PCI_VENDOR_ID, &ID); + if (ID == 0xFFFFFFFF) { + function++; + continue; } + rc = pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(device, function), 0x0B, &class_code); + if (rc) + return rc; - } while (function < max_functions); - } // End of IF (device in slot?) - else if (is_hot_plug) { - // Setup slot structure with entry for empty slot - new_slot = cpqhp_slot_create(busnumber); + rc = pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(device, function), PCI_HEADER_TYPE, &header_type); + if (rc) + return rc; - if (new_slot == NULL) { - return(1); + stop_it++; } - new_slot->bus = (u8) busnumber; - new_slot->device = (u8) device; - new_slot->function = 0; - new_slot->is_a_board = 0; - new_slot->presence_save = 0; - new_slot->switch_save = 0; - } - } // End of FOR loop + } while (function < max_functions); + } /* End of FOR loop */ - return(0); + return 0; } @@ -489,7 +468,7 @@ int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot) u8 secondary_bus; int sub_bus; int max_functions; - int function; + int function = 0; int cloop = 0; int stop_it; @@ -498,63 +477,58 @@ int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot) ctrl->pci_bus->number = new_slot->bus; pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), PCI_VENDOR_ID, &ID); - if (ID != 0xFFFFFFFF) { // device in slot - pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), 0x0B, &class_code); - pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), PCI_HEADER_TYPE, &header_type); - - if (header_type & 0x80) // Multi-function device - max_functions = 8; - else - max_functions = 1; - - function = 0; - - do { - if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { // PCI-PCI Bridge - // Recurse the subordinate bus - pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_SECONDARY_BUS, &secondary_bus); + if (ID == 0xFFFFFFFF) + return 2; - sub_bus = (int) secondary_bus; + pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), 0x0B, &class_code); + pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), PCI_HEADER_TYPE, &header_type); - // Save the config headers for the secondary bus. - rc = cpqhp_save_config(ctrl, sub_bus, 0); - if (rc) - return(rc); - ctrl->pci_bus->number = new_slot->bus; + if (header_type & 0x80) /* Multi-function device */ + max_functions = 8; + else + max_functions = 1; - } // End of IF + while (function < max_functions) { + if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { + /* Recurse the subordinate bus */ + pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_SECONDARY_BUS, &secondary_bus); - new_slot->status = 0; + sub_bus = (int) secondary_bus; - for (cloop = 0; cloop < 0x20; cloop++) { - pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), cloop << 2, (u32 *) & (new_slot-> config_space [cloop])); - } + /* Save the config headers for the secondary + * bus. + */ + rc = cpqhp_save_config(ctrl, sub_bus, 0); + if (rc) + return(rc); + ctrl->pci_bus->number = new_slot->bus; - function++; + } - stop_it = 0; + new_slot->status = 0; - // this loop skips to the next present function - // reading in the Class Code and the Header type. + for (cloop = 0; cloop < 0x20; cloop++) + pci_bus_read_config_dword(ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), cloop << 2, (u32 *) & (new_slot-> config_space [cloop])); - while ((function < max_functions) && (!stop_it)) { - pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_VENDOR_ID, &ID); + function++; - if (ID == 0xFFFFFFFF) { // nothing there. - function++; - } else { // Something there - pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), 0x0B, &class_code); + stop_it = 0; - pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_HEADER_TYPE, &header_type); + /* this loop skips to the next present function + * reading in the Class Code and the Header type. + */ + while ((function < max_functions) && (!stop_it)) { + pci_bus_read_config_dword(ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_VENDOR_ID, &ID); - stop_it++; - } + if (ID == 0xFFFFFFFF) + function++; + else { + pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), 0x0B, &class_code); + pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_HEADER_TYPE, &header_type); + stop_it++; } + } - } while (function < max_functions); - } // End of IF (device in slot?) - else { - return 2; } return 0; @@ -590,11 +564,10 @@ int cpqhp_save_base_addr_length(struct controller *ctrl, struct pci_func * func) pci_bus->number = func->bus; devfn = PCI_DEVFN(func->device, func->function); - // Check for Bridge + /* Check for Bridge */ pci_bus_read_config_byte (pci_bus, devfn, PCI_HEADER_TYPE, &header_type); if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { - // PCI-PCI Bridge pci_bus_read_config_byte (pci_bus, devfn, PCI_SECONDARY_BUS, &secondary_bus); sub_bus = (int) secondary_bus; @@ -610,23 +583,27 @@ int cpqhp_save_base_addr_length(struct controller *ctrl, struct pci_func * func) } pci_bus->number = func->bus; - //FIXME: this loop is duplicated in the non-bridge case. The two could be rolled together - // Figure out IO and memory base lengths + /* FIXME: this loop is duplicated in the non-bridge + * case. The two could be rolled together Figure out + * IO and memory base lengths + */ for (cloop = 0x10; cloop <= 0x14; cloop += 4) { temp_register = 0xFFFFFFFF; pci_bus_write_config_dword (pci_bus, devfn, cloop, temp_register); pci_bus_read_config_dword (pci_bus, devfn, cloop, &base); - - if (base) { // If this register is implemented + /* If this register is implemented */ + if (base) { if (base & 0x01L) { - // IO base - // set base = amount of IO space requested + /* IO base + * set base = amount of IO space + * requested + */ base = base & 0xFFFFFFFE; base = (~base) + 1; type = 1; } else { - // memory base + /* memory base */ base = base & 0xFFFFFFF0; base = (~base) + 1; @@ -637,32 +614,36 @@ int cpqhp_save_base_addr_length(struct controller *ctrl, struct pci_func * func) type = 0; } - // Save information in slot structure + /* Save information in slot structure */ func->base_length[(cloop - 0x10) >> 2] = base; func->base_type[(cloop - 0x10) >> 2] = type; - } // End of base register loop + } /* End of base register loop */ - - } else if ((header_type & 0x7F) == 0x00) { // PCI-PCI Bridge - // Figure out IO and memory base lengths + } else if ((header_type & 0x7F) == 0x00) { + /* Figure out IO and memory base lengths */ for (cloop = 0x10; cloop <= 0x24; cloop += 4) { temp_register = 0xFFFFFFFF; pci_bus_write_config_dword (pci_bus, devfn, cloop, temp_register); pci_bus_read_config_dword (pci_bus, devfn, cloop, &base); - if (base) { // If this register is implemented + /* If this register is implemented */ + if (base) { if (base & 0x01L) { - // IO base - // base = amount of IO space requested + /* IO base + * base = amount of IO space + * requested + */ base = base & 0xFFFFFFFE; base = (~base) + 1; type = 1; } else { - // memory base - // base = amount of memory space requested + /* memory base + * base = amount of memory + * space requested + */ base = base & 0xFFFFFFF0; base = (~base) + 1; @@ -673,16 +654,16 @@ int cpqhp_save_base_addr_length(struct controller *ctrl, struct pci_func * func) type = 0; } - // Save information in slot structure + /* Save information in slot structure */ func->base_length[(cloop - 0x10) >> 2] = base; func->base_type[(cloop - 0x10) >> 2] = type; - } // End of base register loop + } /* End of base register loop */ - } else { // Some other unknown header type + } else { /* Some other unknown header type */ } - // find the next device in this slot + /* find the next device in this slot */ func = cpqhp_slot_find(func->bus, func->device, index++); } @@ -728,18 +709,18 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) pci_bus->number = func->bus; devfn = PCI_DEVFN(func->device, func->function); - // Save the command register + /* Save the command register */ pci_bus_read_config_word(pci_bus, devfn, PCI_COMMAND, &save_command); - // disable card + /* disable card */ command = 0x00; pci_bus_write_config_word(pci_bus, devfn, PCI_COMMAND, command); - // Check for Bridge + /* Check for Bridge */ pci_bus_read_config_byte(pci_bus, devfn, PCI_HEADER_TYPE, &header_type); - if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { // PCI-PCI Bridge - // Clear Bridge Control Register + if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { + /* Clear Bridge Control Register */ command = 0x00; pci_bus_write_config_word(pci_bus, devfn, PCI_BRIDGE_CONTROL, command); pci_bus_read_config_byte(pci_bus, devfn, PCI_SECONDARY_BUS, &secondary_bus); @@ -755,7 +736,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) bus_node->next = func->bus_head; func->bus_head = bus_node; - // Save IO base and Limit registers + /* Save IO base and Limit registers */ pci_bus_read_config_byte(pci_bus, devfn, PCI_IO_BASE, &b_base); pci_bus_read_config_byte(pci_bus, devfn, PCI_IO_LIMIT, &b_length); @@ -771,7 +752,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) func->io_head = io_node; } - // Save memory base and Limit registers + /* Save memory base and Limit registers */ pci_bus_read_config_word(pci_bus, devfn, PCI_MEMORY_BASE, &w_base); pci_bus_read_config_word(pci_bus, devfn, PCI_MEMORY_LIMIT, &w_length); @@ -787,7 +768,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) func->mem_head = mem_node; } - // Save prefetchable memory base and Limit registers + /* Save prefetchable memory base and Limit registers */ pci_bus_read_config_word(pci_bus, devfn, PCI_PREF_MEMORY_BASE, &w_base); pci_bus_read_config_word(pci_bus, devfn, PCI_PREF_MEMORY_LIMIT, &w_length); @@ -802,7 +783,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) p_mem_node->next = func->p_mem_head; func->p_mem_head = p_mem_node; } - // Figure out IO and memory base lengths + /* Figure out IO and memory base lengths */ for (cloop = 0x10; cloop <= 0x14; cloop += 4) { pci_bus_read_config_dword (pci_bus, devfn, cloop, &save_base); @@ -812,11 +793,14 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) temp_register = base; - if (base) { // If this register is implemented + /* If this register is implemented */ + if (base) { if (((base & 0x03L) == 0x01) && (save_command & 0x01)) { - // IO base - // set temp_register = amount of IO space requested + /* IO base + * set temp_register = amount + * of IO space requested + */ temp_register = base & 0xFFFFFFFE; temp_register = (~temp_register) + 1; @@ -834,7 +818,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) } else if (((base & 0x0BL) == 0x08) && (save_command & 0x02)) { - // prefetchable memory base + /* prefetchable memory base */ temp_register = base & 0xFFFFFFF0; temp_register = (~temp_register) + 1; @@ -851,7 +835,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) } else if (((base & 0x0BL) == 0x00) && (save_command & 0x02)) { - // prefetchable memory base + /* prefetchable memory base */ temp_register = base & 0xFFFFFFF0; temp_register = (~temp_register) + 1; @@ -868,9 +852,10 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) } else return(1); } - } // End of base register loop - } else if ((header_type & 0x7F) == 0x00) { // Standard header - // Figure out IO and memory base lengths + } /* End of base register loop */ + /* Standard header */ + } else if ((header_type & 0x7F) == 0x00) { + /* Figure out IO and memory base lengths */ for (cloop = 0x10; cloop <= 0x24; cloop += 4) { pci_bus_read_config_dword(pci_bus, devfn, cloop, &save_base); @@ -880,11 +865,14 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) temp_register = base; - if (base) { // If this register is implemented + /* If this register is implemented */ + if (base) { if (((base & 0x03L) == 0x01) && (save_command & 0x01)) { - // IO base - // set temp_register = amount of IO space requested + /* IO base + * set temp_register = amount + * of IO space requested + */ temp_register = base & 0xFFFFFFFE; temp_register = (~temp_register) + 1; @@ -901,7 +889,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) } else if (((base & 0x0BL) == 0x08) && (save_command & 0x02)) { - // prefetchable memory base + /* prefetchable memory base */ temp_register = base & 0xFFFFFFF0; temp_register = (~temp_register) + 1; @@ -918,7 +906,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) } else if (((base & 0x0BL) == 0x00) && (save_command & 0x02)) { - // prefetchable memory base + /* prefetchable memory base */ temp_register = base & 0xFFFFFFF0; temp_register = (~temp_register) + 1; @@ -935,15 +923,14 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func) } else return(1); } - } // End of base register loop - } else { // Some other unknown header type + } /* End of base register loop */ } - // find the next device in this slot + /* find the next device in this slot */ func = cpqhp_slot_find(func->bus, func->device, index++); } - return(0); + return 0; } @@ -975,16 +962,16 @@ int cpqhp_configure_board(struct controller *ctrl, struct pci_func * func) pci_bus->number = func->bus; devfn = PCI_DEVFN(func->device, func->function); - // Start at the top of config space so that the control - // registers are programmed last - for (cloop = 0x3C; cloop > 0; cloop -= 4) { + /* Start at the top of config space so that the control + * registers are programmed last + */ + for (cloop = 0x3C; cloop > 0; cloop -= 4) pci_bus_write_config_dword (pci_bus, devfn, cloop, func->config_space[cloop >> 2]); - } pci_bus_read_config_byte (pci_bus, devfn, PCI_HEADER_TYPE, &header_type); - // If this is a bridge device, restore subordinate devices - if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { // PCI-PCI Bridge + /* If this is a bridge device, restore subordinate devices */ + if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { pci_bus_read_config_byte (pci_bus, devfn, PCI_SECONDARY_BUS, &secondary_bus); sub_bus = (int) secondary_bus; @@ -1000,8 +987,9 @@ int cpqhp_configure_board(struct controller *ctrl, struct pci_func * func) } } else { - // Check all the base Address Registers to make sure - // they are the same. If not, the board is different. + /* Check all the base Address Registers to make sure + * they are the same. If not, the board is different. + */ for (cloop = 16; cloop < 40; cloop += 4) { pci_bus_read_config_dword (pci_bus, devfn, cloop, &temp); @@ -1058,27 +1046,28 @@ int cpqhp_valid_replace(struct controller *ctrl, struct pci_func * func) pci_bus_read_config_dword (pci_bus, devfn, PCI_VENDOR_ID, &temp_register); - // No adapter present + /* No adapter present */ if (temp_register == 0xFFFFFFFF) return(NO_ADAPTER_PRESENT); if (temp_register != func->config_space[0]) return(ADAPTER_NOT_SAME); - // Check for same revision number and class code + /* Check for same revision number and class code */ pci_bus_read_config_dword (pci_bus, devfn, PCI_CLASS_REVISION, &temp_register); - // Adapter not the same + /* Adapter not the same */ if (temp_register != func->config_space[0x08 >> 2]) return(ADAPTER_NOT_SAME); - // Check for Bridge + /* Check for Bridge */ pci_bus_read_config_byte (pci_bus, devfn, PCI_HEADER_TYPE, &header_type); - if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { // PCI-PCI Bridge - // In order to continue checking, we must program the - // bus registers in the bridge to respond to accesses - // for it's subordinate bus(es) + if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { + /* In order to continue checking, we must program the + * bus registers in the bridge to respond to accesses + * for its subordinate bus(es) + */ temp_register = func->config_space[0x18 >> 2]; pci_bus_write_config_dword (pci_bus, devfn, PCI_PRIMARY_BUS, temp_register); @@ -1096,35 +1085,39 @@ int cpqhp_valid_replace(struct controller *ctrl, struct pci_func * func) } } - // Check to see if it is a standard config header + /* Check to see if it is a standard config header */ else if ((header_type & 0x7F) == PCI_HEADER_TYPE_NORMAL) { - // Check subsystem vendor and ID + /* Check subsystem vendor and ID */ pci_bus_read_config_dword (pci_bus, devfn, PCI_SUBSYSTEM_VENDOR_ID, &temp_register); if (temp_register != func->config_space[0x2C >> 2]) { - // If it's a SMART-2 and the register isn't filled - // in, ignore the difference because - // they just have an old rev of the firmware - + /* If it's a SMART-2 and the register isn't + * filled in, ignore the difference because + * they just have an old rev of the firmware + */ if (!((func->config_space[0] == 0xAE100E11) && (temp_register == 0x00L))) return(ADAPTER_NOT_SAME); } - // Figure out IO and memory base lengths + /* Figure out IO and memory base lengths */ for (cloop = 0x10; cloop <= 0x24; cloop += 4) { temp_register = 0xFFFFFFFF; pci_bus_write_config_dword (pci_bus, devfn, cloop, temp_register); pci_bus_read_config_dword (pci_bus, devfn, cloop, &base); - if (base) { // If this register is implemented + + /* If this register is implemented */ + if (base) { if (base & 0x01L) { - // IO base - // set base = amount of IO space requested + /* IO base + * set base = amount of IO + * space requested + */ base = base & 0xFFFFFFFE; base = (~base) + 1; type = 1; } else { - // memory base + /* memory base */ base = base & 0xFFFFFFF0; base = (~base) + 1; @@ -1135,23 +1128,24 @@ int cpqhp_valid_replace(struct controller *ctrl, struct pci_func * func) type = 0; } - // Check information in slot structure + /* Check information in slot structure */ if (func->base_length[(cloop - 0x10) >> 2] != base) return(ADAPTER_NOT_SAME); if (func->base_type[(cloop - 0x10) >> 2] != type) return(ADAPTER_NOT_SAME); - } // End of base register loop + } /* End of base register loop */ - } // End of (type 0 config space) else + } /* End of (type 0 config space) else */ else { - // this is not a type 0 or 1 config space header so - // we don't know how to do it + /* this is not a type 0 or 1 config space header so + * we don't know how to do it + */ return(DEVICE_TYPE_NOT_SUPPORTED); } - // Get the next function + /* Get the next function */ func = cpqhp_slot_find(func->bus, func->device, index++); } @@ -1168,7 +1162,7 @@ int cpqhp_valid_replace(struct controller *ctrl, struct pci_func * func) * this function is for hot plug ADD! * * returns 0 if success - */ + */ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_start) { u8 temp; @@ -1187,10 +1181,10 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st rom_resource_table = detect_HRT_floating_pointer(rom_start, rom_start+0xffff); dbg("rom_resource_table = %p\n", rom_resource_table); - if (rom_resource_table == NULL) { + if (rom_resource_table == NULL) return -ENODEV; - } - // Sum all resources and setup resource maps + + /* Sum all resources and setup resource maps */ unused_IRQ = readl(rom_resource_table + UNUSED_IRQ); dbg("unused_IRQ = %x\n", unused_IRQ); @@ -1222,13 +1216,11 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st temp = 0; - if (!cpqhp_nic_irq) { + if (!cpqhp_nic_irq) cpqhp_nic_irq = ctrl->cfgspc_irq; - } - if (!cpqhp_disk_irq) { + if (!cpqhp_disk_irq) cpqhp_disk_irq = ctrl->cfgspc_irq; - } dbg("cpqhp_disk_irq, cpqhp_nic_irq= %d, %d\n", cpqhp_disk_irq, cpqhp_nic_irq); @@ -1262,13 +1254,13 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st dev_func, io_base, io_length, mem_base, mem_length, pre_mem_base, pre_mem_length, primary_bus, secondary_bus, max_bus); - // If this entry isn't for our controller's bus, ignore it + /* If this entry isn't for our controller's bus, ignore it */ if (primary_bus != ctrl->bus) { i--; one_slot += sizeof (struct slot_rt); continue; } - // find out if this entry is for an occupied slot + /* find out if this entry is for an occupied slot */ ctrl->pci_bus->number = primary_bus; pci_bus_read_config_dword (ctrl->pci_bus, dev_func, PCI_VENDOR_ID, &temp_dword); dbg("temp_D_word = %x\n", temp_dword); @@ -1282,13 +1274,13 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st func = cpqhp_slot_find(primary_bus, dev_func >> 3, index++); } - // If we can't find a match, skip this table entry + /* If we can't find a match, skip this table entry */ if (!func) { i--; one_slot += sizeof (struct slot_rt); continue; } - // this may not work and shouldn't be used + /* this may not work and shouldn't be used */ if (secondary_bus != primary_bus) bridged_slot = 1; else @@ -1301,7 +1293,7 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st } - // If we've got a valid IO base, use it + /* If we've got a valid IO base, use it */ temp_dword = io_base + io_length; @@ -1325,7 +1317,7 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st } } - // If we've got a valid memory base, use it + /* If we've got a valid memory base, use it */ temp_dword = mem_base + mem_length; if ((mem_base) && (temp_dword < 0x10000)) { mem_node = kmalloc(sizeof(*mem_node), GFP_KERNEL); @@ -1348,8 +1340,9 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st } } - // If we've got a valid prefetchable memory base, and - // the base + length isn't greater than 0xFFFF + /* If we've got a valid prefetchable memory base, and + * the base + length isn't greater than 0xFFFF + */ temp_dword = pre_mem_base + pre_mem_length; if ((pre_mem_base) && (temp_dword < 0x10000)) { p_mem_node = kmalloc(sizeof(*p_mem_node), GFP_KERNEL); @@ -1372,9 +1365,10 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st } } - // If we've got a valid bus number, use it - // The second condition is to ignore bus numbers on - // populated slots that don't have PCI-PCI bridges + /* If we've got a valid bus number, use it + * The second condition is to ignore bus numbers on + * populated slots that don't have PCI-PCI bridges + */ if (secondary_bus && (secondary_bus != primary_bus)) { bus_node = kmalloc(sizeof(*bus_node), GFP_KERNEL); if (!bus_node) @@ -1398,8 +1392,9 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st one_slot += sizeof (struct slot_rt); } - // If all of the following fail, we don't have any resources for - // hot plug add + /* If all of the following fail, we don't have any resources for + * hot plug add + */ rc = 1; rc &= cpqhp_resource_sort_and_combine(&(ctrl->mem_head)); rc &= cpqhp_resource_sort_and_combine(&(ctrl->p_mem_head)); diff --git a/drivers/pci/hotplug/cpqphp_sysfs.c b/drivers/pci/hotplug/cpqphp_sysfs.c index 8450f4a6568..e6089bdb6e5 100644 --- a/drivers/pci/hotplug/cpqphp_sysfs.c +++ b/drivers/pci/hotplug/cpqphp_sysfs.c @@ -33,6 +33,7 @@ #include <linux/workqueue.h> #include <linux/pci.h> #include <linux/pci_hotplug.h> +#include <linux/smp_lock.h> #include <linux/debugfs.h> #include "cpqphp.h" diff --git a/drivers/pci/hotplug/ibmphp_core.c b/drivers/pci/hotplug/ibmphp_core.c index dd18f857dfb..7485ffda950 100644 --- a/drivers/pci/hotplug/ibmphp_core.c +++ b/drivers/pci/hotplug/ibmphp_core.c @@ -153,45 +153,47 @@ int ibmphp_init_devno(struct slot **cur_slot) return -1; } for (loop = 0; loop < len; loop++) { - if ((*cur_slot)->number == rtable->slots[loop].slot) { - if ((*cur_slot)->bus == rtable->slots[loop].bus) { + if ((*cur_slot)->number == rtable->slots[loop].slot && + (*cur_slot)->bus == rtable->slots[loop].bus) { + struct io_apic_irq_attr irq_attr; + (*cur_slot)->device = PCI_SLOT(rtable->slots[loop].devfn); for (i = 0; i < 4; i++) (*cur_slot)->irq[i] = IO_APIC_get_PCI_irq_vector((int) (*cur_slot)->bus, - (int) (*cur_slot)->device, i); - - debug("(*cur_slot)->irq[0] = %x\n", - (*cur_slot)->irq[0]); - debug("(*cur_slot)->irq[1] = %x\n", - (*cur_slot)->irq[1]); - debug("(*cur_slot)->irq[2] = %x\n", - (*cur_slot)->irq[2]); - debug("(*cur_slot)->irq[3] = %x\n", - (*cur_slot)->irq[3]); - - debug("rtable->exlusive_irqs = %x\n", + (int) (*cur_slot)->device, i, + &irq_attr); + + debug("(*cur_slot)->irq[0] = %x\n", + (*cur_slot)->irq[0]); + debug("(*cur_slot)->irq[1] = %x\n", + (*cur_slot)->irq[1]); + debug("(*cur_slot)->irq[2] = %x\n", + (*cur_slot)->irq[2]); + debug("(*cur_slot)->irq[3] = %x\n", + (*cur_slot)->irq[3]); + + debug("rtable->exlusive_irqs = %x\n", rtable->exclusive_irqs); - debug("rtable->slots[loop].irq[0].bitmap = %x\n", + debug("rtable->slots[loop].irq[0].bitmap = %x\n", rtable->slots[loop].irq[0].bitmap); - debug("rtable->slots[loop].irq[1].bitmap = %x\n", + debug("rtable->slots[loop].irq[1].bitmap = %x\n", rtable->slots[loop].irq[1].bitmap); - debug("rtable->slots[loop].irq[2].bitmap = %x\n", + debug("rtable->slots[loop].irq[2].bitmap = %x\n", rtable->slots[loop].irq[2].bitmap); - debug("rtable->slots[loop].irq[3].bitmap = %x\n", + debug("rtable->slots[loop].irq[3].bitmap = %x\n", rtable->slots[loop].irq[3].bitmap); - debug("rtable->slots[loop].irq[0].link = %x\n", + debug("rtable->slots[loop].irq[0].link = %x\n", rtable->slots[loop].irq[0].link); - debug("rtable->slots[loop].irq[1].link = %x\n", + debug("rtable->slots[loop].irq[1].link = %x\n", rtable->slots[loop].irq[1].link); - debug("rtable->slots[loop].irq[2].link = %x\n", + debug("rtable->slots[loop].irq[2].link = %x\n", rtable->slots[loop].irq[2].link); - debug("rtable->slots[loop].irq[3].link = %x\n", + debug("rtable->slots[loop].irq[3].link = %x\n", rtable->slots[loop].irq[3].link); - debug("end of init_devno\n"); - kfree(rtable); - return 0; - } + debug("end of init_devno\n"); + kfree(rtable); + return 0; } } @@ -1316,7 +1318,6 @@ error: } struct hotplug_slot_ops ibmphp_hotplug_slot_ops = { - .owner = THIS_MODULE, .set_attention_status = set_attention_status, .enable_slot = enable_slot, .disable_slot = ibmphp_disable_slot, @@ -1419,3 +1420,4 @@ static void __exit ibmphp_exit(void) } module_init(ibmphp_init); +module_exit(ibmphp_exit); diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c index 535fce0f07f..5c5043f239c 100644 --- a/drivers/pci/hotplug/pci_hotplug_core.c +++ b/drivers/pci/hotplug/pci_hotplug_core.c @@ -347,125 +347,129 @@ static struct pci_slot_attribute hotplug_slot_attr_test = { .store = test_write_file }; -static int has_power_file(struct pci_slot *pci_slot) +static bool has_power_file(struct pci_slot *pci_slot) { struct hotplug_slot *slot = pci_slot->hotplug; if ((!slot) || (!slot->ops)) - return -ENODEV; + return false; if ((slot->ops->enable_slot) || (slot->ops->disable_slot) || (slot->ops->get_power_status)) - return 0; - return -ENOENT; + return true; + return false; } -static int has_attention_file(struct pci_slot *pci_slot) +static bool has_attention_file(struct pci_slot *pci_slot) { struct hotplug_slot *slot = pci_slot->hotplug; if ((!slot) || (!slot->ops)) - return -ENODEV; + return false; if ((slot->ops->set_attention_status) || (slot->ops->get_attention_status)) - return 0; - return -ENOENT; + return true; + return false; } -static int has_latch_file(struct pci_slot *pci_slot) +static bool has_latch_file(struct pci_slot *pci_slot) { struct hotplug_slot *slot = pci_slot->hotplug; if ((!slot) || (!slot->ops)) - return -ENODEV; + return false; if (slot->ops->get_latch_status) - return 0; - return -ENOENT; + return true; + return false; } -static int has_adapter_file(struct pci_slot *pci_slot) +static bool has_adapter_file(struct pci_slot *pci_slot) { struct hotplug_slot *slot = pci_slot->hotplug; if ((!slot) || (!slot->ops)) - return -ENODEV; + return false; if (slot->ops->get_adapter_status) - return 0; - return -ENOENT; + return true; + return false; } -static int has_max_bus_speed_file(struct pci_slot *pci_slot) +static bool has_max_bus_speed_file(struct pci_slot *pci_slot) { struct hotplug_slot *slot = pci_slot->hotplug; if ((!slot) || (!slot->ops)) - return -ENODEV; + return false; if (slot->ops->get_max_bus_speed) - return 0; - return -ENOENT; + return true; + return false; } -static int has_cur_bus_speed_file(struct pci_slot *pci_slot) +static bool has_cur_bus_speed_file(struct pci_slot *pci_slot) { struct hotplug_slot *slot = pci_slot->hotplug; if ((!slot) || (!slot->ops)) - return -ENODEV; + return false; if (slot->ops->get_cur_bus_speed) - return 0; - return -ENOENT; + return true; + return false; } -static int has_test_file(struct pci_slot *pci_slot) +static bool has_test_file(struct pci_slot *pci_slot) { struct hotplug_slot *slot = pci_slot->hotplug; if ((!slot) || (!slot->ops)) - return -ENODEV; + return false; if (slot->ops->hardware_test) - return 0; - return -ENOENT; + return true; + return false; } static int fs_add_slot(struct pci_slot *slot) { int retval = 0; - if (has_power_file(slot) == 0) { - retval = sysfs_create_file(&slot->kobj, &hotplug_slot_attr_power.attr); + /* Create symbolic link to the hotplug driver module */ + pci_hp_create_module_link(slot); + + if (has_power_file(slot)) { + retval = sysfs_create_file(&slot->kobj, + &hotplug_slot_attr_power.attr); if (retval) goto exit_power; } - if (has_attention_file(slot) == 0) { + if (has_attention_file(slot)) { retval = sysfs_create_file(&slot->kobj, &hotplug_slot_attr_attention.attr); if (retval) goto exit_attention; } - if (has_latch_file(slot) == 0) { + if (has_latch_file(slot)) { retval = sysfs_create_file(&slot->kobj, &hotplug_slot_attr_latch.attr); if (retval) goto exit_latch; } - if (has_adapter_file(slot) == 0) { + if (has_adapter_file(slot)) { retval = sysfs_create_file(&slot->kobj, &hotplug_slot_attr_presence.attr); if (retval) goto exit_adapter; } - if (has_max_bus_speed_file(slot) == 0) { + if (has_max_bus_speed_file(slot)) { retval = sysfs_create_file(&slot->kobj, - &hotplug_slot_attr_max_bus_speed.attr); + &hotplug_slot_attr_max_bus_speed.attr); if (retval) goto exit_max_speed; } - if (has_cur_bus_speed_file(slot) == 0) { + if (has_cur_bus_speed_file(slot)) { retval = sysfs_create_file(&slot->kobj, - &hotplug_slot_attr_cur_bus_speed.attr); + &hotplug_slot_attr_cur_bus_speed.attr); if (retval) goto exit_cur_speed; } - if (has_test_file(slot) == 0) { + if (has_test_file(slot)) { retval = sysfs_create_file(&slot->kobj, &hotplug_slot_attr_test.attr); if (retval) @@ -475,55 +479,61 @@ static int fs_add_slot(struct pci_slot *slot) goto exit; exit_test: - if (has_cur_bus_speed_file(slot) == 0) - sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_cur_bus_speed.attr); - + if (has_cur_bus_speed_file(slot)) + sysfs_remove_file(&slot->kobj, + &hotplug_slot_attr_cur_bus_speed.attr); exit_cur_speed: - if (has_max_bus_speed_file(slot) == 0) - sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_max_bus_speed.attr); - + if (has_max_bus_speed_file(slot)) + sysfs_remove_file(&slot->kobj, + &hotplug_slot_attr_max_bus_speed.attr); exit_max_speed: - if (has_adapter_file(slot) == 0) - sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_presence.attr); - + if (has_adapter_file(slot)) + sysfs_remove_file(&slot->kobj, + &hotplug_slot_attr_presence.attr); exit_adapter: - if (has_latch_file(slot) == 0) + if (has_latch_file(slot)) sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_latch.attr); - exit_latch: - if (has_attention_file(slot) == 0) - sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_attention.attr); - + if (has_attention_file(slot)) + sysfs_remove_file(&slot->kobj, + &hotplug_slot_attr_attention.attr); exit_attention: - if (has_power_file(slot) == 0) + if (has_power_file(slot)) sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_power.attr); exit_power: + pci_hp_remove_module_link(slot); exit: return retval; } static void fs_remove_slot(struct pci_slot *slot) { - if (has_power_file(slot) == 0) + if (has_power_file(slot)) sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_power.attr); - if (has_attention_file(slot) == 0) - sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_attention.attr); + if (has_attention_file(slot)) + sysfs_remove_file(&slot->kobj, + &hotplug_slot_attr_attention.attr); - if (has_latch_file(slot) == 0) + if (has_latch_file(slot)) sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_latch.attr); - if (has_adapter_file(slot) == 0) - sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_presence.attr); + if (has_adapter_file(slot)) + sysfs_remove_file(&slot->kobj, + &hotplug_slot_attr_presence.attr); - if (has_max_bus_speed_file(slot) == 0) - sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_max_bus_speed.attr); + if (has_max_bus_speed_file(slot)) + sysfs_remove_file(&slot->kobj, + &hotplug_slot_attr_max_bus_speed.attr); - if (has_cur_bus_speed_file(slot) == 0) - sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_cur_bus_speed.attr); + if (has_cur_bus_speed_file(slot)) + sysfs_remove_file(&slot->kobj, + &hotplug_slot_attr_cur_bus_speed.attr); - if (has_test_file(slot) == 0) + if (has_test_file(slot)) sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_test.attr); + + pci_hp_remove_module_link(slot); } static struct hotplug_slot *get_slot_from_name (const char *name) @@ -540,19 +550,22 @@ static struct hotplug_slot *get_slot_from_name (const char *name) } /** - * pci_hp_register - register a hotplug_slot with the PCI hotplug subsystem + * __pci_hp_register - register a hotplug_slot with the PCI hotplug subsystem * @bus: bus this slot is on * @slot: pointer to the &struct hotplug_slot to register - * @slot_nr: slot number + * @devnr: device number * @name: name registered with kobject core + * @owner: caller module owner + * @mod_name: caller module name * * Registers a hotplug slot with the pci hotplug subsystem, which will allow * userspace interaction to the slot. * * Returns 0 if successful, anything else for an error. */ -int pci_hp_register(struct hotplug_slot *slot, struct pci_bus *bus, int slot_nr, - const char *name) +int __pci_hp_register(struct hotplug_slot *slot, struct pci_bus *bus, + int devnr, const char *name, + struct module *owner, const char *mod_name) { int result; struct pci_slot *pci_slot; @@ -567,14 +580,16 @@ int pci_hp_register(struct hotplug_slot *slot, struct pci_bus *bus, int slot_nr, return -EINVAL; } - mutex_lock(&pci_hp_mutex); + slot->ops->owner = owner; + slot->ops->mod_name = mod_name; + mutex_lock(&pci_hp_mutex); /* * No problems if we call this interface from both ACPI_PCI_SLOT * driver and call it here again. If we've already created the * pci_slot, the interface will simply bump the refcount. */ - pci_slot = pci_create_slot(bus, slot_nr, name, slot); + pci_slot = pci_create_slot(bus, devnr, name, slot); if (IS_ERR(pci_slot)) { result = PTR_ERR(pci_slot); goto out; @@ -684,6 +699,6 @@ MODULE_LICENSE("GPL"); module_param(debug, bool, 0644); MODULE_PARM_DESC(debug, "Debugging mode enabled or not"); -EXPORT_SYMBOL_GPL(pci_hp_register); +EXPORT_SYMBOL_GPL(__pci_hp_register); EXPORT_SYMBOL_GPL(pci_hp_deregister); EXPORT_SYMBOL_GPL(pci_hp_change_slot_info); diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index 0a368547e63..e6cf096498b 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -81,7 +81,6 @@ struct slot { struct hpc_ops *hpc_ops; struct hotplug_slot *hotplug_slot; struct list_head slot_list; - unsigned long last_emi_toggle; struct delayed_work work; /* work for button event */ struct mutex lock; }; @@ -203,8 +202,6 @@ struct hpc_ops { int (*set_attention_status)(struct slot *slot, u8 status); int (*get_latch_status)(struct slot *slot, u8 *status); int (*get_adapter_status)(struct slot *slot, u8 *status); - int (*get_emi_status)(struct slot *slot, u8 *status); - int (*toggle_emi)(struct slot *slot); int (*get_max_bus_speed)(struct slot *slot, enum pci_bus_speed *speed); int (*get_cur_bus_speed)(struct slot *slot, enum pci_bus_speed *speed); int (*get_max_lnk_width)(struct slot *slot, enum pcie_link_width *val); diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c index fb254b2454d..2317557fdee 100644 --- a/drivers/pci/hotplug/pciehp_core.c +++ b/drivers/pci/hotplug/pciehp_core.c @@ -73,7 +73,6 @@ static int get_max_bus_speed (struct hotplug_slot *slot, enum pci_bus_speed *val static int get_cur_bus_speed (struct hotplug_slot *slot, enum pci_bus_speed *value); static struct hotplug_slot_ops pciehp_hotplug_slot_ops = { - .owner = THIS_MODULE, .set_attention_status = set_attention_status, .enable_slot = enable_slot, .disable_slot = disable_slot, @@ -85,99 +84,6 @@ static struct hotplug_slot_ops pciehp_hotplug_slot_ops = { .get_cur_bus_speed = get_cur_bus_speed, }; -/* - * Check the status of the Electro Mechanical Interlock (EMI) - */ -static int get_lock_status(struct hotplug_slot *hotplug_slot, u8 *value) -{ - struct slot *slot = hotplug_slot->private; - return (slot->hpc_ops->get_emi_status(slot, value)); -} - -/* - * sysfs interface for the Electro Mechanical Interlock (EMI) - * 1 == locked, 0 == unlocked - */ -static ssize_t lock_read_file(struct hotplug_slot *slot, char *buf) -{ - int retval; - u8 value; - - retval = get_lock_status(slot, &value); - if (retval) - goto lock_read_exit; - retval = sprintf (buf, "%d\n", value); - -lock_read_exit: - return retval; -} - -/* - * Change the status of the Electro Mechanical Interlock (EMI) - * This is a toggle - in addition there must be at least 1 second - * in between toggles. - */ -static int set_lock_status(struct hotplug_slot *hotplug_slot, u8 status) -{ - struct slot *slot = hotplug_slot->private; - int retval; - u8 value; - - mutex_lock(&slot->ctrl->crit_sect); - - /* has it been >1 sec since our last toggle? */ - if ((get_seconds() - slot->last_emi_toggle) < 1) { - mutex_unlock(&slot->ctrl->crit_sect); - return -EINVAL; - } - - /* see what our current state is */ - retval = get_lock_status(hotplug_slot, &value); - if (retval || (value == status)) - goto set_lock_exit; - - slot->hpc_ops->toggle_emi(slot); -set_lock_exit: - mutex_unlock(&slot->ctrl->crit_sect); - return 0; -} - -/* - * sysfs interface which allows the user to toggle the Electro Mechanical - * Interlock. Valid values are either 0 or 1. 0 == unlock, 1 == lock - */ -static ssize_t lock_write_file(struct hotplug_slot *hotplug_slot, - const char *buf, size_t count) -{ - struct slot *slot = hotplug_slot->private; - unsigned long llock; - u8 lock; - int retval = 0; - - llock = simple_strtoul(buf, NULL, 10); - lock = (u8)(llock & 0xff); - - switch (lock) { - case 0: - case 1: - retval = set_lock_status(hotplug_slot, lock); - break; - default: - ctrl_err(slot->ctrl, "%d is an invalid lock value\n", - lock); - retval = -EINVAL; - } - if (retval) - return retval; - return count; -} - -static struct hotplug_slot_attribute hotplug_slot_attr_lock = { - .attr = {.name = "lock", .mode = S_IFREG | S_IRUGO | S_IWUSR}, - .show = lock_read_file, - .store = lock_write_file -}; - /** * release_slot - free up the memory used by a slot * @hotplug_slot: slot to free @@ -236,17 +142,6 @@ static int init_slots(struct controller *ctrl) get_attention_status(hotplug_slot, &info->attention_status); get_latch_status(hotplug_slot, &info->latch_status); get_adapter_status(hotplug_slot, &info->adapter_status); - /* create additional sysfs entries */ - if (EMI(ctrl)) { - retval = sysfs_create_file(&hotplug_slot->pci_slot->kobj, - &hotplug_slot_attr_lock.attr); - if (retval) { - pci_hp_deregister(hotplug_slot); - ctrl_err(ctrl, "Cannot create additional sysfs " - "entries\n"); - goto error_info; - } - } } return 0; @@ -261,13 +156,8 @@ error: static void cleanup_slots(struct controller *ctrl) { struct slot *slot; - - list_for_each_entry(slot, &ctrl->slot_list, slot_list) { - if (EMI(ctrl)) - sysfs_remove_file(&slot->hotplug_slot->pci_slot->kobj, - &hotplug_slot_attr_lock.attr); + list_for_each_entry(slot, &ctrl->slot_list, slot_list) pci_hp_deregister(slot->hotplug_slot); - } } /* diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c index ff4034502d2..8aab8edf123 100644 --- a/drivers/pci/hotplug/pciehp_ctrl.c +++ b/drivers/pci/hotplug/pciehp_ctrl.c @@ -30,7 +30,6 @@ #include <linux/module.h> #include <linux/kernel.h> #include <linux/types.h> -#include <linux/smp_lock.h> #include <linux/pci.h> #include <linux/workqueue.h> #include "../pci.h" diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index 07bd3215114..52813257e5b 100644 --- a/drivers/pci/hotplug/pciehp_hpc.c +++ b/drivers/pci/hotplug/pciehp_hpc.c @@ -422,35 +422,6 @@ static int hpc_query_power_fault(struct slot *slot) return !!(slot_status & PCI_EXP_SLTSTA_PFD); } -static int hpc_get_emi_status(struct slot *slot, u8 *status) -{ - struct controller *ctrl = slot->ctrl; - u16 slot_status; - int retval; - - retval = pciehp_readw(ctrl, PCI_EXP_SLTSTA, &slot_status); - if (retval) { - ctrl_err(ctrl, "Cannot check EMI status\n"); - return retval; - } - *status = !!(slot_status & PCI_EXP_SLTSTA_EIS); - return retval; -} - -static int hpc_toggle_emi(struct slot *slot) -{ - u16 slot_cmd; - u16 cmd_mask; - int rc; - - slot_cmd = PCI_EXP_SLTCTL_EIC; - cmd_mask = PCI_EXP_SLTCTL_EIC; - rc = pcie_write_cmd(slot->ctrl, slot_cmd, cmd_mask); - slot->last_emi_toggle = get_seconds(); - - return rc; -} - static int hpc_set_attention_status(struct slot *slot, u8 value) { struct controller *ctrl = slot->ctrl; @@ -874,8 +845,6 @@ static struct hpc_ops pciehp_hpc_ops = { .get_attention_status = hpc_get_attention_status, .get_latch_status = hpc_get_latch_status, .get_adapter_status = hpc_get_adapter_status, - .get_emi_status = hpc_get_emi_status, - .toggle_emi = hpc_toggle_emi, .get_max_bus_speed = hpc_get_max_lnk_speed, .get_cur_bus_speed = hpc_get_cur_lnk_speed, diff --git a/drivers/pci/hotplug/pcihp_skeleton.c b/drivers/pci/hotplug/pcihp_skeleton.c index e3dd6cf9e89..5175d9b26f0 100644 --- a/drivers/pci/hotplug/pcihp_skeleton.c +++ b/drivers/pci/hotplug/pcihp_skeleton.c @@ -82,7 +82,6 @@ static int get_latch_status (struct hotplug_slot *slot, u8 *value); static int get_adapter_status (struct hotplug_slot *slot, u8 *value); static struct hotplug_slot_ops skel_hotplug_slot_ops = { - .owner = THIS_MODULE, .enable_slot = enable_slot, .disable_slot = disable_slot, .set_attention_status = set_attention_status, diff --git a/drivers/pci/hotplug/rpaphp_core.c b/drivers/pci/hotplug/rpaphp_core.c index 95d02a08fdc..c159223389e 100644 --- a/drivers/pci/hotplug/rpaphp_core.c +++ b/drivers/pci/hotplug/rpaphp_core.c @@ -423,7 +423,6 @@ static int disable_slot(struct hotplug_slot *hotplug_slot) } struct hotplug_slot_ops rpaphp_hotplug_slot_ops = { - .owner = THIS_MODULE, .enable_slot = enable_slot, .disable_slot = disable_slot, .set_attention_status = set_attention_status, diff --git a/drivers/pci/hotplug/sgi_hotplug.c b/drivers/pci/hotplug/sgi_hotplug.c index 3eee70928d4..8aebe1e9d3d 100644 --- a/drivers/pci/hotplug/sgi_hotplug.c +++ b/drivers/pci/hotplug/sgi_hotplug.c @@ -83,7 +83,6 @@ static int disable_slot(struct hotplug_slot *slot); static inline int get_power_status(struct hotplug_slot *slot, u8 *value); static struct hotplug_slot_ops sn_hotplug_slot_ops = { - .owner = THIS_MODULE, .enable_slot = enable_slot, .disable_slot = disable_slot, .get_power_status = get_power_status, @@ -91,11 +90,10 @@ static struct hotplug_slot_ops sn_hotplug_slot_ops = { static DEFINE_MUTEX(sn_hotplug_mutex); -static ssize_t path_show (struct hotplug_slot *bss_hotplug_slot, - char *buf) +static ssize_t path_show(struct pci_slot *pci_slot, char *buf) { int retval = -ENOENT; - struct slot *slot = bss_hotplug_slot->private; + struct slot *slot = pci_slot->hotplug->private; if (!slot) return retval; @@ -104,7 +102,7 @@ static ssize_t path_show (struct hotplug_slot *bss_hotplug_slot, return retval; } -static struct hotplug_slot_attribute sn_slot_path_attr = __ATTR_RO(path); +static struct pci_slot_attribute sn_slot_path_attr = __ATTR_RO(path); static int sn_pci_slot_valid(struct pci_bus *pci_bus, int device) { @@ -679,7 +677,7 @@ alloc_err: return rc; } -static int sn_pci_hotplug_init(void) +static int __init sn_pci_hotplug_init(void) { struct pci_bus *pci_bus = NULL; int rc; @@ -716,7 +714,7 @@ static int sn_pci_hotplug_init(void) return registered == 1 ? 0 : -ENODEV; } -static void sn_pci_hotplug_exit(void) +static void __exit sn_pci_hotplug_exit(void) { struct hotplug_slot *bss_hotplug_slot; diff --git a/drivers/pci/hotplug/shpchp_core.c b/drivers/pci/hotplug/shpchp_core.c index fe8d149c229..8a520a3d0f5 100644 --- a/drivers/pci/hotplug/shpchp_core.c +++ b/drivers/pci/hotplug/shpchp_core.c @@ -69,7 +69,6 @@ static int get_max_bus_speed (struct hotplug_slot *slot, enum pci_bus_speed *val static int get_cur_bus_speed (struct hotplug_slot *slot, enum pci_bus_speed *value); static struct hotplug_slot_ops shpchp_hotplug_slot_ops = { - .owner = THIS_MODULE, .set_attention_status = set_attention_status, .enable_slot = enable_slot, .disable_slot = disable_slot, diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c index 6808d8333ec..737a1c44b07 100644 --- a/drivers/pci/htirq.c +++ b/drivers/pci/htirq.c @@ -98,6 +98,7 @@ int __ht_create_irq(struct pci_dev *dev, int idx, ht_irq_update_t *update) int max_irq; int pos; int irq; + int node; pos = pci_find_ht_capability(dev, HT_CAPTYPE_IRQ); if (!pos) @@ -125,7 +126,8 @@ int __ht_create_irq(struct pci_dev *dev, int idx, ht_irq_update_t *update) cfg->msg.address_lo = 0xffffffff; cfg->msg.address_hi = 0xffffffff; - irq = create_irq(); + node = dev_to_node(&dev->dev); + irq = create_irq_nr(0, node); if (irq <= 0) { kfree(cfg); diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c index a563fbe559d..2314ad7ee5f 100644 --- a/drivers/pci/intel-iommu.c +++ b/drivers/pci/intel-iommu.c @@ -53,15 +53,35 @@ #define DEFAULT_DOMAIN_ADDRESS_WIDTH 48 +#define MAX_AGAW_WIDTH 64 + #define DOMAIN_MAX_ADDR(gaw) ((((u64)1) << gaw) - 1) +#define DOMAIN_MAX_PFN(gaw) ((((u64)1) << (gaw-VTD_PAGE_SHIFT)) - 1) #define IOVA_PFN(addr) ((addr) >> PAGE_SHIFT) #define DMA_32BIT_PFN IOVA_PFN(DMA_BIT_MASK(32)) #define DMA_64BIT_PFN IOVA_PFN(DMA_BIT_MASK(64)) -#ifndef PHYSICAL_PAGE_MASK -#define PHYSICAL_PAGE_MASK PAGE_MASK -#endif + +/* VT-d pages must always be _smaller_ than MM pages. Otherwise things + are never going to work. */ +static inline unsigned long dma_to_mm_pfn(unsigned long dma_pfn) +{ + return dma_pfn >> (PAGE_SHIFT - VTD_PAGE_SHIFT); +} + +static inline unsigned long mm_to_dma_pfn(unsigned long mm_pfn) +{ + return mm_pfn << (PAGE_SHIFT - VTD_PAGE_SHIFT); +} +static inline unsigned long page_to_dma_pfn(struct page *pg) +{ + return mm_to_dma_pfn(page_to_pfn(pg)); +} +static inline unsigned long virt_to_dma_pfn(void *p) +{ + return page_to_dma_pfn(virt_to_page(p)); +} /* global iommu list, set NULL for ignored DMAR units */ static struct intel_iommu **g_iommus; @@ -131,8 +151,6 @@ static inline void context_set_fault_enable(struct context_entry *context) context->lo &= (((u64)-1) << 2) | 1; } -#define CONTEXT_TT_MULTI_LEVEL 0 - static inline void context_set_translation_type(struct context_entry *context, unsigned long value) { @@ -204,12 +222,17 @@ static inline void dma_set_pte_prot(struct dma_pte *pte, unsigned long prot) static inline u64 dma_pte_addr(struct dma_pte *pte) { - return (pte->val & VTD_PAGE_MASK); +#ifdef CONFIG_64BIT + return pte->val & VTD_PAGE_MASK; +#else + /* Must have a full atomic 64-bit read */ + return __cmpxchg64(pte, 0ULL, 0ULL) & VTD_PAGE_MASK; +#endif } -static inline void dma_set_pte_addr(struct dma_pte *pte, u64 addr) +static inline void dma_set_pte_pfn(struct dma_pte *pte, unsigned long pfn) { - pte->val |= (addr & VTD_PAGE_MASK); + pte->val |= (uint64_t)pfn << VTD_PAGE_SHIFT; } static inline bool dma_pte_present(struct dma_pte *pte) @@ -217,6 +240,19 @@ static inline bool dma_pte_present(struct dma_pte *pte) return (pte->val & 3) != 0; } +static inline int first_pte_in_page(struct dma_pte *pte) +{ + return !((unsigned long)pte & ~VTD_PAGE_MASK); +} + +/* + * This domain is a statically identity mapping domain. + * 1. This domain creats a static 1:1 mapping to all usable memory. + * 2. It maps to each iommu if successful. + * 3. Each iommu mapps to this domain if successful. + */ +struct dmar_domain *si_domain; + /* devices under the same p2p bridge are owned in one domain */ #define DOMAIN_FLAG_P2P_MULTIPLE_DEVICES (1 << 0) @@ -225,6 +261,9 @@ static inline bool dma_pte_present(struct dma_pte *pte) */ #define DOMAIN_FLAG_VIRTUAL_MACHINE (1 << 1) +/* si_domain contains mulitple devices */ +#define DOMAIN_FLAG_STATIC_IDENTITY (1 << 2) + struct dmar_domain { int id; /* domain id */ unsigned long iommu_bmp; /* bitmap of iommus this domain uses*/ @@ -233,7 +272,6 @@ struct dmar_domain { struct iova_domain iovad; /* iova's that belong to this domain */ struct dma_pte *pgd; /* virtual address */ - spinlock_t mapping_lock; /* page table lock */ int gaw; /* max guest address width */ /* adjusted guest address width, 0 is level 2 30-bit */ @@ -256,6 +294,7 @@ struct device_domain_info { u8 bus; /* PCI bus number */ u8 devfn; /* PCI devfn number */ struct pci_dev *dev; /* it's NULL for PCIE-to-PCI bridge */ + struct intel_iommu *iommu; /* IOMMU used by this device */ struct dmar_domain *domain; /* pointer to domain */ }; @@ -401,17 +440,13 @@ void free_iova_mem(struct iova *iova) static inline int width_to_agaw(int width); -/* calculate agaw for each iommu. - * "SAGAW" may be different across iommus, use a default agaw, and - * get a supported less agaw for iommus that don't support the default agaw. - */ -int iommu_calculate_agaw(struct intel_iommu *iommu) +static int __iommu_calculate_agaw(struct intel_iommu *iommu, int max_gaw) { unsigned long sagaw; int agaw = -1; sagaw = cap_sagaw(iommu->cap); - for (agaw = width_to_agaw(DEFAULT_DOMAIN_ADDRESS_WIDTH); + for (agaw = width_to_agaw(max_gaw); agaw >= 0; agaw--) { if (test_bit(agaw, &sagaw)) break; @@ -420,12 +455,32 @@ int iommu_calculate_agaw(struct intel_iommu *iommu) return agaw; } -/* in native case, each domain is related to only one iommu */ +/* + * Calculate max SAGAW for each iommu. + */ +int iommu_calculate_max_sagaw(struct intel_iommu *iommu) +{ + return __iommu_calculate_agaw(iommu, MAX_AGAW_WIDTH); +} + +/* + * calculate agaw for each iommu. + * "SAGAW" may be different across iommus, use a default agaw, and + * get a supported less agaw for iommus that don't support the default agaw. + */ +int iommu_calculate_agaw(struct intel_iommu *iommu) +{ + return __iommu_calculate_agaw(iommu, DEFAULT_DOMAIN_ADDRESS_WIDTH); +} + +/* This functionin only returns single iommu in a domain */ static struct intel_iommu *domain_get_iommu(struct dmar_domain *domain) { int iommu_id; + /* si_domain and vm domain should not get here. */ BUG_ON(domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE); + BUG_ON(domain->flags & DOMAIN_FLAG_STATIC_IDENTITY); iommu_id = find_first_bit(&domain->iommu_bmp, g_num_of_iommus); if (iommu_id < 0 || iommu_id >= g_num_of_iommus) @@ -620,80 +675,78 @@ static inline int width_to_agaw(int width) static inline unsigned int level_to_offset_bits(int level) { - return (12 + (level - 1) * LEVEL_STRIDE); + return (level - 1) * LEVEL_STRIDE; } -static inline int address_level_offset(u64 addr, int level) +static inline int pfn_level_offset(unsigned long pfn, int level) { - return ((addr >> level_to_offset_bits(level)) & LEVEL_MASK); + return (pfn >> level_to_offset_bits(level)) & LEVEL_MASK; } -static inline u64 level_mask(int level) +static inline unsigned long level_mask(int level) { - return ((u64)-1 << level_to_offset_bits(level)); + return -1UL << level_to_offset_bits(level); } -static inline u64 level_size(int level) +static inline unsigned long level_size(int level) { - return ((u64)1 << level_to_offset_bits(level)); + return 1UL << level_to_offset_bits(level); } -static inline u64 align_to_level(u64 addr, int level) +static inline unsigned long align_to_level(unsigned long pfn, int level) { - return ((addr + level_size(level) - 1) & level_mask(level)); + return (pfn + level_size(level) - 1) & level_mask(level); } -static struct dma_pte * addr_to_dma_pte(struct dmar_domain *domain, u64 addr) +static struct dma_pte *pfn_to_dma_pte(struct dmar_domain *domain, + unsigned long pfn) { - int addr_width = agaw_to_width(domain->agaw); + int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT; struct dma_pte *parent, *pte = NULL; int level = agaw_to_level(domain->agaw); int offset; - unsigned long flags; BUG_ON(!domain->pgd); - - addr &= (((u64)1) << addr_width) - 1; + BUG_ON(addr_width < BITS_PER_LONG && pfn >> addr_width); parent = domain->pgd; - spin_lock_irqsave(&domain->mapping_lock, flags); while (level > 0) { void *tmp_page; - offset = address_level_offset(addr, level); + offset = pfn_level_offset(pfn, level); pte = &parent[offset]; if (level == 1) break; if (!dma_pte_present(pte)) { + uint64_t pteval; + tmp_page = alloc_pgtable_page(); - if (!tmp_page) { - spin_unlock_irqrestore(&domain->mapping_lock, - flags); + if (!tmp_page) return NULL; + + domain_flush_cache(domain, tmp_page, VTD_PAGE_SIZE); + pteval = (virt_to_dma_pfn(tmp_page) << VTD_PAGE_SHIFT) | DMA_PTE_READ | DMA_PTE_WRITE; + if (cmpxchg64(&pte->val, 0ULL, pteval)) { + /* Someone else set it while we were thinking; use theirs. */ + free_pgtable_page(tmp_page); + } else { + dma_pte_addr(pte); + domain_flush_cache(domain, pte, sizeof(*pte)); } - domain_flush_cache(domain, tmp_page, PAGE_SIZE); - dma_set_pte_addr(pte, virt_to_phys(tmp_page)); - /* - * high level table always sets r/w, last level page - * table control read/write - */ - dma_set_pte_readable(pte); - dma_set_pte_writable(pte); - domain_flush_cache(domain, pte, sizeof(*pte)); } parent = phys_to_virt(dma_pte_addr(pte)); level--; } - spin_unlock_irqrestore(&domain->mapping_lock, flags); return pte; } /* return address's pte at specific level */ -static struct dma_pte *dma_addr_level_pte(struct dmar_domain *domain, u64 addr, - int level) +static struct dma_pte *dma_pfn_level_pte(struct dmar_domain *domain, + unsigned long pfn, + int level) { struct dma_pte *parent, *pte = NULL; int total = agaw_to_level(domain->agaw); @@ -701,7 +754,7 @@ static struct dma_pte *dma_addr_level_pte(struct dmar_domain *domain, u64 addr, parent = domain->pgd; while (level <= total) { - offset = address_level_offset(addr, total); + offset = pfn_level_offset(pfn, total); pte = &parent[offset]; if (level == total) return pte; @@ -714,74 +767,82 @@ static struct dma_pte *dma_addr_level_pte(struct dmar_domain *domain, u64 addr, return NULL; } -/* clear one page's page table */ -static void dma_pte_clear_one(struct dmar_domain *domain, u64 addr) -{ - struct dma_pte *pte = NULL; - - /* get last level pte */ - pte = dma_addr_level_pte(domain, addr, 1); - - if (pte) { - dma_clear_pte(pte); - domain_flush_cache(domain, pte, sizeof(*pte)); - } -} - /* clear last level pte, a tlb flush should be followed */ -static void dma_pte_clear_range(struct dmar_domain *domain, u64 start, u64 end) +static void dma_pte_clear_range(struct dmar_domain *domain, + unsigned long start_pfn, + unsigned long last_pfn) { - int addr_width = agaw_to_width(domain->agaw); - int npages; + int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT; + struct dma_pte *first_pte, *pte; - start &= (((u64)1) << addr_width) - 1; - end &= (((u64)1) << addr_width) - 1; - /* in case it's partial page */ - start &= PAGE_MASK; - end = PAGE_ALIGN(end); - npages = (end - start) / VTD_PAGE_SIZE; + BUG_ON(addr_width < BITS_PER_LONG && start_pfn >> addr_width); + BUG_ON(addr_width < BITS_PER_LONG && last_pfn >> addr_width); - /* we don't need lock here, nobody else touches the iova range */ - while (npages--) { - dma_pte_clear_one(domain, start); - start += VTD_PAGE_SIZE; + /* we don't need lock here; nobody else touches the iova range */ + while (start_pfn <= last_pfn) { + first_pte = pte = dma_pfn_level_pte(domain, start_pfn, 1); + if (!pte) { + start_pfn = align_to_level(start_pfn + 1, 2); + continue; + } + do { + dma_clear_pte(pte); + start_pfn++; + pte++; + } while (start_pfn <= last_pfn && !first_pte_in_page(pte)); + + domain_flush_cache(domain, first_pte, + (void *)pte - (void *)first_pte); } } /* free page table pages. last level pte should already be cleared */ static void dma_pte_free_pagetable(struct dmar_domain *domain, - u64 start, u64 end) + unsigned long start_pfn, + unsigned long last_pfn) { - int addr_width = agaw_to_width(domain->agaw); - struct dma_pte *pte; + int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT; + struct dma_pte *first_pte, *pte; int total = agaw_to_level(domain->agaw); int level; - u64 tmp; + unsigned long tmp; - start &= (((u64)1) << addr_width) - 1; - end &= (((u64)1) << addr_width) - 1; + BUG_ON(addr_width < BITS_PER_LONG && start_pfn >> addr_width); + BUG_ON(addr_width < BITS_PER_LONG && last_pfn >> addr_width); - /* we don't need lock here, nobody else touches the iova range */ + /* We don't need lock here; nobody else touches the iova range */ level = 2; while (level <= total) { - tmp = align_to_level(start, level); - if (tmp >= end || (tmp + level_size(level) > end)) + tmp = align_to_level(start_pfn, level); + + /* If we can't even clear one PTE at this level, we're done */ + if (tmp + level_size(level) - 1 > last_pfn) return; - while (tmp < end) { - pte = dma_addr_level_pte(domain, tmp, level); - if (pte) { - free_pgtable_page( - phys_to_virt(dma_pte_addr(pte))); - dma_clear_pte(pte); - domain_flush_cache(domain, pte, sizeof(*pte)); + while (tmp + level_size(level) - 1 <= last_pfn) { + first_pte = pte = dma_pfn_level_pte(domain, tmp, level); + if (!pte) { + tmp = align_to_level(tmp + 1, level + 1); + continue; } - tmp += level_size(level); + do { + if (dma_pte_present(pte)) { + free_pgtable_page(phys_to_virt(dma_pte_addr(pte))); + dma_clear_pte(pte); + } + pte++; + tmp += level_size(level); + } while (!first_pte_in_page(pte) && + tmp + level_size(level) - 1 <= last_pfn); + + domain_flush_cache(domain, first_pte, + (void *)pte - (void *)first_pte); + } level++; } /* free pgd */ - if (start == 0 && end >= ((((u64)1) << addr_width) - 1)) { + if (start_pfn == 0 && last_pfn == DOMAIN_MAX_PFN(domain->gaw)) { free_pgtable_page(domain->pgd); domain->pgd = NULL; } @@ -809,7 +870,7 @@ static int iommu_alloc_root_entry(struct intel_iommu *iommu) static void iommu_set_root_entry(struct intel_iommu *iommu) { void *addr; - u32 cmd, sts; + u32 sts; unsigned long flag; addr = iommu->root_entry; @@ -817,12 +878,11 @@ static void iommu_set_root_entry(struct intel_iommu *iommu) spin_lock_irqsave(&iommu->register_lock, flag); dmar_writeq(iommu->reg + DMAR_RTADDR_REG, virt_to_phys(addr)); - cmd = iommu->gcmd | DMA_GCMD_SRTP; - writel(cmd, iommu->reg + DMAR_GCMD_REG); + writel(iommu->gcmd | DMA_GCMD_SRTP, iommu->reg + DMAR_GCMD_REG); /* Make sure hardware complete it */ IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, - readl, (sts & DMA_GSTS_RTPS), sts); + readl, (sts & DMA_GSTS_RTPS), sts); spin_unlock_irqrestore(&iommu->register_lock, flag); } @@ -834,39 +894,25 @@ static void iommu_flush_write_buffer(struct intel_iommu *iommu) if (!rwbf_quirk && !cap_rwbf(iommu->cap)) return; - val = iommu->gcmd | DMA_GCMD_WBF; spin_lock_irqsave(&iommu->register_lock, flag); - writel(val, iommu->reg + DMAR_GCMD_REG); + writel(iommu->gcmd | DMA_GCMD_WBF, iommu->reg + DMAR_GCMD_REG); /* Make sure hardware complete it */ IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, - readl, (!(val & DMA_GSTS_WBFS)), val); + readl, (!(val & DMA_GSTS_WBFS)), val); spin_unlock_irqrestore(&iommu->register_lock, flag); } /* return value determine if we need a write buffer flush */ -static int __iommu_flush_context(struct intel_iommu *iommu, - u16 did, u16 source_id, u8 function_mask, u64 type, - int non_present_entry_flush) +static void __iommu_flush_context(struct intel_iommu *iommu, + u16 did, u16 source_id, u8 function_mask, + u64 type) { u64 val = 0; unsigned long flag; - /* - * In the non-present entry flush case, if hardware doesn't cache - * non-present entry we do nothing and if hardware cache non-present - * entry, we flush entries of domain 0 (the domain id is used to cache - * any non-present entries) - */ - if (non_present_entry_flush) { - if (!cap_caching_mode(iommu->cap)) - return 1; - else - did = 0; - } - switch (type) { case DMA_CCMD_GLOBAL_INVL: val = DMA_CCMD_GLOBAL_INVL; @@ -891,33 +937,16 @@ static int __iommu_flush_context(struct intel_iommu *iommu, dmar_readq, (!(val & DMA_CCMD_ICC)), val); spin_unlock_irqrestore(&iommu->register_lock, flag); - - /* flush context entry will implicitly flush write buffer */ - return 0; } /* return value determine if we need a write buffer flush */ -static int __iommu_flush_iotlb(struct intel_iommu *iommu, u16 did, - u64 addr, unsigned int size_order, u64 type, - int non_present_entry_flush) +static void __iommu_flush_iotlb(struct intel_iommu *iommu, u16 did, + u64 addr, unsigned int size_order, u64 type) { int tlb_offset = ecap_iotlb_offset(iommu->ecap); u64 val = 0, val_iva = 0; unsigned long flag; - /* - * In the non-present entry flush case, if hardware doesn't cache - * non-present entry we do nothing and if hardware cache non-present - * entry, we flush entries of domain 0 (the domain id is used to cache - * any non-present entries) - */ - if (non_present_entry_flush) { - if (!cap_caching_mode(iommu->cap)) - return 1; - else - did = 0; - } - switch (type) { case DMA_TLB_GLOBAL_FLUSH: /* global flush doesn't need set IVA_REG */ @@ -965,37 +994,106 @@ static int __iommu_flush_iotlb(struct intel_iommu *iommu, u16 did, pr_debug("IOMMU: tlb flush request %Lx, actual %Lx\n", (unsigned long long)DMA_TLB_IIRG(type), (unsigned long long)DMA_TLB_IAIG(val)); - /* flush iotlb entry will implicitly flush write buffer */ - return 0; } -static int iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did, - u64 addr, unsigned int pages, int non_present_entry_flush) +static struct device_domain_info *iommu_support_dev_iotlb( + struct dmar_domain *domain, int segment, u8 bus, u8 devfn) { - unsigned int mask; + int found = 0; + unsigned long flags; + struct device_domain_info *info; + struct intel_iommu *iommu = device_to_iommu(segment, bus, devfn); - BUG_ON(addr & (~VTD_PAGE_MASK)); - BUG_ON(pages == 0); + if (!ecap_dev_iotlb_support(iommu->ecap)) + return NULL; + + if (!iommu->qi) + return NULL; + + spin_lock_irqsave(&device_domain_lock, flags); + list_for_each_entry(info, &domain->devices, link) + if (info->bus == bus && info->devfn == devfn) { + found = 1; + break; + } + spin_unlock_irqrestore(&device_domain_lock, flags); + + if (!found || !info->dev) + return NULL; + + if (!pci_find_ext_capability(info->dev, PCI_EXT_CAP_ID_ATS)) + return NULL; + + if (!dmar_find_matched_atsr_unit(info->dev)) + return NULL; + + info->iommu = iommu; + + return info; +} + +static void iommu_enable_dev_iotlb(struct device_domain_info *info) +{ + if (!info) + return; + + pci_enable_ats(info->dev, VTD_PAGE_SHIFT); +} + +static void iommu_disable_dev_iotlb(struct device_domain_info *info) +{ + if (!info->dev || !pci_ats_enabled(info->dev)) + return; + + pci_disable_ats(info->dev); +} - /* Fallback to domain selective flush if no PSI support */ - if (!cap_pgsel_inv(iommu->cap)) - return iommu->flush.flush_iotlb(iommu, did, 0, 0, - DMA_TLB_DSI_FLUSH, - non_present_entry_flush); +static void iommu_flush_dev_iotlb(struct dmar_domain *domain, + u64 addr, unsigned mask) +{ + u16 sid, qdep; + unsigned long flags; + struct device_domain_info *info; + + spin_lock_irqsave(&device_domain_lock, flags); + list_for_each_entry(info, &domain->devices, link) { + if (!info->dev || !pci_ats_enabled(info->dev)) + continue; + + sid = info->bus << 8 | info->devfn; + qdep = pci_ats_queue_depth(info->dev); + qi_flush_dev_iotlb(info->iommu, sid, qdep, addr, mask); + } + spin_unlock_irqrestore(&device_domain_lock, flags); +} + +static void iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did, + unsigned long pfn, unsigned int pages) +{ + unsigned int mask = ilog2(__roundup_pow_of_two(pages)); + uint64_t addr = (uint64_t)pfn << VTD_PAGE_SHIFT; + + BUG_ON(pages == 0); /* + * Fallback to domain selective flush if no PSI support or the size is + * too big. * PSI requires page size to be 2 ^ x, and the base address is naturally * aligned to the size */ - mask = ilog2(__roundup_pow_of_two(pages)); - /* Fallback to domain selective flush if size is too big */ - if (mask > cap_max_amask_val(iommu->cap)) - return iommu->flush.flush_iotlb(iommu, did, 0, 0, - DMA_TLB_DSI_FLUSH, non_present_entry_flush); + if (!cap_pgsel_inv(iommu->cap) || mask > cap_max_amask_val(iommu->cap)) + iommu->flush.flush_iotlb(iommu, did, 0, 0, + DMA_TLB_DSI_FLUSH); + else + iommu->flush.flush_iotlb(iommu, did, addr, mask, + DMA_TLB_PSI_FLUSH); - return iommu->flush.flush_iotlb(iommu, did, addr, mask, - DMA_TLB_PSI_FLUSH, - non_present_entry_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. + */ + if (!cap_caching_mode(iommu->cap) || did) + iommu_flush_dev_iotlb(iommu->domains[did], addr, mask); } static void iommu_disable_protect_mem_regions(struct intel_iommu *iommu) @@ -1021,13 +1119,13 @@ static int iommu_enable_translation(struct intel_iommu *iommu) unsigned long flags; spin_lock_irqsave(&iommu->register_lock, flags); - writel(iommu->gcmd|DMA_GCMD_TE, iommu->reg + DMAR_GCMD_REG); + iommu->gcmd |= DMA_GCMD_TE; + writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG); /* Make sure hardware complete it */ IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, - readl, (sts & DMA_GSTS_TES), sts); + readl, (sts & DMA_GSTS_TES), sts); - iommu->gcmd |= DMA_GCMD_TE; spin_unlock_irqrestore(&iommu->register_lock, flags); return 0; } @@ -1043,7 +1141,7 @@ static int iommu_disable_translation(struct intel_iommu *iommu) /* Make sure hardware complete it */ IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, - readl, (!(sts & DMA_GSTS_TES)), sts); + readl, (!(sts & DMA_GSTS_TES)), sts); spin_unlock_irqrestore(&iommu->register_lock, flag); return 0; @@ -1142,48 +1240,71 @@ void free_dmar_iommu(struct intel_iommu *iommu) free_context_table(iommu); } -static struct dmar_domain * iommu_alloc_domain(struct intel_iommu *iommu) +static struct dmar_domain *alloc_domain(void) { - unsigned long num; - unsigned long ndomains; struct dmar_domain *domain; - unsigned long flags; domain = alloc_domain_mem(); if (!domain) return NULL; + memset(&domain->iommu_bmp, 0, sizeof(unsigned long)); + domain->flags = 0; + + return domain; +} + +static int iommu_attach_domain(struct dmar_domain *domain, + struct intel_iommu *iommu) +{ + int num; + unsigned long ndomains; + unsigned long flags; + ndomains = cap_ndoms(iommu->cap); spin_lock_irqsave(&iommu->lock, flags); + num = find_first_zero_bit(iommu->domain_ids, ndomains); if (num >= ndomains) { spin_unlock_irqrestore(&iommu->lock, flags); - free_domain_mem(domain); printk(KERN_ERR "IOMMU: no free domain ids\n"); - return NULL; + return -ENOMEM; } - set_bit(num, iommu->domain_ids); domain->id = num; - memset(&domain->iommu_bmp, 0, sizeof(unsigned long)); + set_bit(num, iommu->domain_ids); set_bit(iommu->seq_id, &domain->iommu_bmp); - domain->flags = 0; iommu->domains[num] = domain; spin_unlock_irqrestore(&iommu->lock, flags); - return domain; + return 0; } -static void iommu_free_domain(struct dmar_domain *domain) +static void iommu_detach_domain(struct dmar_domain *domain, + struct intel_iommu *iommu) { unsigned long flags; - struct intel_iommu *iommu; - - iommu = domain_get_iommu(domain); + int num, ndomains; + int found = 0; spin_lock_irqsave(&iommu->lock, flags); - clear_bit(domain->id, iommu->domain_ids); + ndomains = cap_ndoms(iommu->cap); + num = find_first_bit(iommu->domain_ids, ndomains); + for (; num < ndomains; ) { + if (iommu->domains[num] == domain) { + found = 1; + break; + } + num = find_next_bit(iommu->domain_ids, + cap_ndoms(iommu->cap), num+1); + } + + if (found) { + clear_bit(num, iommu->domain_ids); + clear_bit(iommu->seq_id, &domain->iommu_bmp); + iommu->domains[num] = NULL; + } spin_unlock_irqrestore(&iommu->lock, flags); } @@ -1196,7 +1317,6 @@ static void dmar_init_reserved_ranges(void) struct pci_dev *pdev = NULL; struct iova *iova; int i; - u64 addr, size; init_iova_domain(&reserved_iova_list, DMA_32BIT_PFN); @@ -1219,12 +1339,9 @@ static void dmar_init_reserved_ranges(void) r = &pdev->resource[i]; if (!r->flags || !(r->flags & IORESOURCE_MEM)) continue; - addr = r->start; - addr &= PHYSICAL_PAGE_MASK; - size = r->end - addr; - size = PAGE_ALIGN(size); - iova = reserve_iova(&reserved_iova_list, IOVA_PFN(addr), - IOVA_PFN(size + addr) - 1); + iova = reserve_iova(&reserved_iova_list, + IOVA_PFN(r->start), + IOVA_PFN(r->end)); if (!iova) printk(KERN_ERR "Reserve iova failed\n"); } @@ -1258,7 +1375,6 @@ static int domain_init(struct dmar_domain *domain, int guest_width) unsigned long sagaw; init_iova_domain(&domain->iovad, DMA_32BIT_PFN); - spin_lock_init(&domain->mapping_lock); spin_lock_init(&domain->iommu_lock); domain_reserve_special_ranges(domain); @@ -1303,7 +1419,8 @@ static int domain_init(struct dmar_domain *domain, int guest_width) static void domain_exit(struct dmar_domain *domain) { - u64 end; + struct dmar_drhd_unit *drhd; + struct intel_iommu *iommu; /* Domain 0 is reserved, so dont process it */ if (!domain) @@ -1312,21 +1429,22 @@ static void domain_exit(struct dmar_domain *domain) domain_remove_dev_info(domain); /* destroy iovas */ put_iova_domain(&domain->iovad); - end = DOMAIN_MAX_ADDR(domain->gaw); - end = end & (~PAGE_MASK); /* clear ptes */ - dma_pte_clear_range(domain, 0, end); + dma_pte_clear_range(domain, 0, DOMAIN_MAX_PFN(domain->gaw)); /* free page tables */ - dma_pte_free_pagetable(domain, 0, end); + dma_pte_free_pagetable(domain, 0, DOMAIN_MAX_PFN(domain->gaw)); + + for_each_active_iommu(iommu, drhd) + if (test_bit(iommu->seq_id, &domain->iommu_bmp)) + iommu_detach_domain(domain, iommu); - iommu_free_domain(domain); free_domain_mem(domain); } -static int domain_context_mapping_one(struct dmar_domain *domain, - int segment, u8 bus, u8 devfn) +static int domain_context_mapping_one(struct dmar_domain *domain, int segment, + u8 bus, u8 devfn, int translation) { struct context_entry *context; unsigned long flags; @@ -1336,10 +1454,14 @@ static int domain_context_mapping_one(struct dmar_domain *domain, unsigned long ndomains; int id; int agaw; + struct device_domain_info *info = NULL; pr_debug("Set context mapping for %02x:%02x.%d\n", bus, PCI_SLOT(devfn), PCI_FUNC(devfn)); + BUG_ON(!domain->pgd); + BUG_ON(translation != CONTEXT_TT_PASS_THROUGH && + translation != CONTEXT_TT_MULTI_LEVEL); iommu = device_to_iommu(segment, bus, devfn); if (!iommu) @@ -1357,7 +1479,8 @@ static int domain_context_mapping_one(struct dmar_domain *domain, id = domain->id; pgd = domain->pgd; - if (domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE) { + if (domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE || + domain->flags & DOMAIN_FLAG_STATIC_IDENTITY) { int found = 0; /* find an available domain id for this device in iommu */ @@ -1399,21 +1522,44 @@ static int domain_context_mapping_one(struct dmar_domain *domain, } context_set_domain_id(context, id); - context_set_address_width(context, iommu->agaw); - context_set_address_root(context, virt_to_phys(pgd)); - context_set_translation_type(context, CONTEXT_TT_MULTI_LEVEL); + + if (translation != CONTEXT_TT_PASS_THROUGH) { + info = iommu_support_dev_iotlb(domain, segment, bus, devfn); + translation = info ? CONTEXT_TT_DEV_IOTLB : + CONTEXT_TT_MULTI_LEVEL; + } + /* + * In pass through mode, AW must be programmed to indicate the largest + * AGAW value supported by hardware. And ASR is ignored by hardware. + */ + if (unlikely(translation == CONTEXT_TT_PASS_THROUGH)) + context_set_address_width(context, iommu->msagaw); + else { + context_set_address_root(context, virt_to_phys(pgd)); + context_set_address_width(context, iommu->agaw); + } + + context_set_translation_type(context, translation); context_set_fault_enable(context); context_set_present(context); domain_flush_cache(domain, context, sizeof(*context)); - /* it's a non-present to present mapping */ - if (iommu->flush.flush_context(iommu, domain->id, - (((u16)bus) << 8) | devfn, DMA_CCMD_MASK_NOBIT, - DMA_CCMD_DEVICE_INVL, 1)) + /* + * It's a non-present to present mapping. If hardware doesn't cache + * non-present entry we only need to flush the write-buffer. If the + * _does_ cache non-present entries, then it does so in the special + * domain #0, which we have to flush: + */ + if (cap_caching_mode(iommu->cap)) { + iommu->flush.flush_context(iommu, 0, + (((u16)bus) << 8) | devfn, + DMA_CCMD_MASK_NOBIT, + DMA_CCMD_DEVICE_INVL); + iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_DSI_FLUSH); + } else { iommu_flush_write_buffer(iommu); - else - iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_DSI_FLUSH, 0); - + } + iommu_enable_dev_iotlb(info); spin_unlock_irqrestore(&iommu->lock, flags); spin_lock_irqsave(&domain->iommu_lock, flags); @@ -1426,13 +1572,15 @@ static int domain_context_mapping_one(struct dmar_domain *domain, } static int -domain_context_mapping(struct dmar_domain *domain, struct pci_dev *pdev) +domain_context_mapping(struct dmar_domain *domain, struct pci_dev *pdev, + int translation) { int ret; struct pci_dev *tmp, *parent; ret = domain_context_mapping_one(domain, pci_domain_nr(pdev->bus), - pdev->bus->number, pdev->devfn); + pdev->bus->number, pdev->devfn, + translation); if (ret) return ret; @@ -1446,7 +1594,7 @@ domain_context_mapping(struct dmar_domain *domain, struct pci_dev *pdev) ret = domain_context_mapping_one(domain, pci_domain_nr(parent->bus), parent->bus->number, - parent->devfn); + parent->devfn, translation); if (ret) return ret; parent = parent->bus->self; @@ -1454,12 +1602,14 @@ domain_context_mapping(struct dmar_domain *domain, struct pci_dev *pdev) if (tmp->is_pcie) /* this is a PCIE-to-PCI bridge */ return domain_context_mapping_one(domain, pci_domain_nr(tmp->subordinate), - tmp->subordinate->number, 0); + tmp->subordinate->number, 0, + translation); else /* this is a legacy PCI bridge */ return domain_context_mapping_one(domain, pci_domain_nr(tmp->bus), tmp->bus->number, - tmp->devfn); + tmp->devfn, + translation); } static int domain_context_mapped(struct pci_dev *pdev) @@ -1497,42 +1647,94 @@ static int domain_context_mapped(struct pci_dev *pdev) tmp->devfn); } -static int -domain_page_mapping(struct dmar_domain *domain, dma_addr_t iova, - u64 hpa, size_t size, int prot) +/* Returns a number of VTD pages, but aligned to MM page size */ +static inline unsigned long aligned_nrpages(unsigned long host_addr, + size_t size) { - u64 start_pfn, end_pfn; - struct dma_pte *pte; - int index; - int addr_width = agaw_to_width(domain->agaw); + host_addr &= ~PAGE_MASK; + return PAGE_ALIGN(host_addr + size) >> VTD_PAGE_SHIFT; +} + +static int __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn, + struct scatterlist *sg, unsigned long phys_pfn, + unsigned long nr_pages, int prot) +{ + struct dma_pte *first_pte = NULL, *pte = NULL; + phys_addr_t uninitialized_var(pteval); + int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT; + unsigned long sg_res; - hpa &= (((u64)1) << addr_width) - 1; + BUG_ON(addr_width < BITS_PER_LONG && (iov_pfn + nr_pages - 1) >> addr_width); if ((prot & (DMA_PTE_READ|DMA_PTE_WRITE)) == 0) return -EINVAL; - iova &= PAGE_MASK; - start_pfn = ((u64)hpa) >> VTD_PAGE_SHIFT; - end_pfn = (VTD_PAGE_ALIGN(((u64)hpa) + size)) >> VTD_PAGE_SHIFT; - index = 0; - while (start_pfn < end_pfn) { - pte = addr_to_dma_pte(domain, iova + VTD_PAGE_SIZE * index); - if (!pte) - return -ENOMEM; + + prot &= DMA_PTE_READ | DMA_PTE_WRITE | DMA_PTE_SNP; + + if (sg) + sg_res = 0; + else { + sg_res = nr_pages + 1; + pteval = ((phys_addr_t)phys_pfn << VTD_PAGE_SHIFT) | prot; + } + + while (nr_pages--) { + uint64_t tmp; + + if (!sg_res) { + sg_res = aligned_nrpages(sg->offset, sg->length); + sg->dma_address = ((dma_addr_t)iov_pfn << VTD_PAGE_SHIFT) + sg->offset; + sg->dma_length = sg->length; + pteval = page_to_phys(sg_page(sg)) | prot; + } + if (!pte) { + first_pte = pte = pfn_to_dma_pte(domain, iov_pfn); + if (!pte) + return -ENOMEM; + } /* We don't need lock here, nobody else * touches the iova range */ - BUG_ON(dma_pte_addr(pte)); - dma_set_pte_addr(pte, start_pfn << VTD_PAGE_SHIFT); - dma_set_pte_prot(pte, prot); - if (prot & DMA_PTE_SNP) - dma_set_pte_snp(pte); - domain_flush_cache(domain, pte, sizeof(*pte)); - start_pfn++; - index++; + tmp = cmpxchg64_local(&pte->val, 0ULL, pteval); + if (tmp) { + static int dumps = 5; + printk(KERN_CRIT "ERROR: DMA PTE for vPFN 0x%lx already set (to %llx not %llx)\n", + iov_pfn, tmp, (unsigned long long)pteval); + if (dumps) { + dumps--; + debug_dma_dump_mappings(NULL); + } + WARN_ON(1); + } + pte++; + if (!nr_pages || first_pte_in_page(pte)) { + domain_flush_cache(domain, first_pte, + (void *)pte - (void *)first_pte); + pte = NULL; + } + iov_pfn++; + pteval += VTD_PAGE_SIZE; + sg_res--; + if (!sg_res) + sg = sg_next(sg); } return 0; } +static inline int domain_sg_mapping(struct dmar_domain *domain, unsigned long iov_pfn, + struct scatterlist *sg, unsigned long nr_pages, + int prot) +{ + return __domain_mapping(domain, iov_pfn, sg, 0, nr_pages, prot); +} + +static inline int domain_pfn_mapping(struct dmar_domain *domain, unsigned long iov_pfn, + unsigned long phys_pfn, unsigned long nr_pages, + int prot) +{ + return __domain_mapping(domain, iov_pfn, NULL, phys_pfn, nr_pages, prot); +} + static void iommu_detach_dev(struct intel_iommu *iommu, u8 bus, u8 devfn) { if (!iommu) @@ -1540,9 +1742,8 @@ static void iommu_detach_dev(struct intel_iommu *iommu, u8 bus, u8 devfn) clear_context_table(iommu, bus, devfn); iommu->flush.flush_context(iommu, 0, 0, 0, - DMA_CCMD_GLOBAL_INVL, 0); - iommu->flush.flush_iotlb(iommu, 0, 0, 0, - DMA_TLB_GLOBAL_FLUSH, 0); + DMA_CCMD_GLOBAL_INVL); + iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_GLOBAL_FLUSH); } static void domain_remove_dev_info(struct dmar_domain *domain) @@ -1561,6 +1762,7 @@ static void domain_remove_dev_info(struct dmar_domain *domain) info->dev->dev.archdata.iommu = NULL; spin_unlock_irqrestore(&device_domain_lock, flags); + iommu_disable_dev_iotlb(info); iommu = device_to_iommu(info->segment, info->bus, info->devfn); iommu_detach_dev(iommu, info->bus, info->devfn); free_devinfo_mem(info); @@ -1597,6 +1799,7 @@ static struct dmar_domain *get_domain_for_dev(struct pci_dev *pdev, int gaw) unsigned long flags; int bus = 0, devfn = 0; int segment; + int ret; domain = find_domain(pdev); if (domain) @@ -1629,6 +1832,10 @@ static struct dmar_domain *get_domain_for_dev(struct pci_dev *pdev, int gaw) } } + domain = alloc_domain(); + if (!domain) + goto error; + /* Allocate new domain for the device */ drhd = dmar_find_matched_drhd_unit(pdev); if (!drhd) { @@ -1638,9 +1845,11 @@ static struct dmar_domain *get_domain_for_dev(struct pci_dev *pdev, int gaw) } iommu = drhd->iommu; - domain = iommu_alloc_domain(iommu); - if (!domain) + ret = iommu_attach_domain(domain, iommu); + if (ret) { + domain_exit(domain); goto error; + } if (domain_init(domain, gaw)) { domain_exit(domain); @@ -1714,55 +1923,63 @@ error: return find_domain(pdev); } +static int iommu_identity_mapping; + +static int iommu_domain_identity_map(struct dmar_domain *domain, + unsigned long long start, + unsigned long long end) +{ + unsigned long first_vpfn = start >> VTD_PAGE_SHIFT; + unsigned long last_vpfn = end >> VTD_PAGE_SHIFT; + + if (!reserve_iova(&domain->iovad, dma_to_mm_pfn(first_vpfn), + dma_to_mm_pfn(last_vpfn))) { + printk(KERN_ERR "IOMMU: reserve iova failed\n"); + return -ENOMEM; + } + + pr_debug("Mapping reserved region %llx-%llx for domain %d\n", + start, end, domain->id); + /* + * RMRR range might have overlap with physical memory range, + * clear it first + */ + dma_pte_clear_range(domain, first_vpfn, last_vpfn); + + return domain_pfn_mapping(domain, first_vpfn, first_vpfn, + last_vpfn - first_vpfn + 1, + DMA_PTE_READ|DMA_PTE_WRITE); +} + static int iommu_prepare_identity_map(struct pci_dev *pdev, unsigned long long start, unsigned long long end) { struct dmar_domain *domain; - unsigned long size; - unsigned long long base; int ret; printk(KERN_INFO - "IOMMU: Setting identity map for device %s [0x%Lx - 0x%Lx]\n", - pci_name(pdev), start, end); - /* page table init */ + "IOMMU: Setting identity map for device %s [0x%Lx - 0x%Lx]\n", + pci_name(pdev), start, end); + domain = get_domain_for_dev(pdev, DEFAULT_DOMAIN_ADDRESS_WIDTH); if (!domain) return -ENOMEM; - /* The address might not be aligned */ - base = start & PAGE_MASK; - size = end - base; - size = PAGE_ALIGN(size); - if (!reserve_iova(&domain->iovad, IOVA_PFN(base), - IOVA_PFN(base + size) - 1)) { - printk(KERN_ERR "IOMMU: reserve iova failed\n"); - ret = -ENOMEM; + ret = iommu_domain_identity_map(domain, start, end); + if (ret) goto error; - } - - pr_debug("Mapping reserved region %lx@%llx for %s\n", - size, base, pci_name(pdev)); - /* - * RMRR range might have overlap with physical memory range, - * clear it first - */ - dma_pte_clear_range(domain, base, base + size); - ret = domain_page_mapping(domain, base, base, size, - DMA_PTE_READ|DMA_PTE_WRITE); + /* context entry init */ + ret = domain_context_mapping(domain, pdev, CONTEXT_TT_MULTI_LEVEL); if (ret) goto error; - /* context entry init */ - ret = domain_context_mapping(domain, pdev); - if (!ret) - return 0; -error: + return 0; + + error: domain_exit(domain); return ret; - } static inline int iommu_prepare_rmrr_dev(struct dmar_rmrr_unit *rmrr, @@ -1774,96 +1991,222 @@ static inline int iommu_prepare_rmrr_dev(struct dmar_rmrr_unit *rmrr, rmrr->end_address + 1); } -#ifdef CONFIG_DMAR_GFX_WA -struct iommu_prepare_data { +#ifdef CONFIG_DMAR_FLOPPY_WA +static inline void iommu_prepare_isa(void) +{ struct pci_dev *pdev; int ret; -}; -static int __init iommu_prepare_work_fn(unsigned long start_pfn, - unsigned long end_pfn, void *datax) -{ - struct iommu_prepare_data *data; + pdev = pci_get_class(PCI_CLASS_BRIDGE_ISA << 8, NULL); + if (!pdev) + return; - data = (struct iommu_prepare_data *)datax; + printk(KERN_INFO "IOMMU: Prepare 0-16MiB unity mapping for LPC\n"); + ret = iommu_prepare_identity_map(pdev, 0, 16*1024*1024); - data->ret = iommu_prepare_identity_map(data->pdev, - start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT); - return data->ret; + if (ret) + printk(KERN_ERR "IOMMU: Failed to create 0-16MiB identity map; " + "floppy might not work\n"); } - -static int __init iommu_prepare_with_active_regions(struct pci_dev *pdev) +#else +static inline void iommu_prepare_isa(void) { - int nid; - struct iommu_prepare_data data; - - data.pdev = pdev; - data.ret = 0; - - for_each_online_node(nid) { - work_with_active_regions(nid, iommu_prepare_work_fn, &data); - if (data.ret) - return data.ret; - } - return data.ret; + return; } +#endif /* !CONFIG_DMAR_FLPY_WA */ -static void __init iommu_prepare_gfx_mapping(void) +/* Initialize each context entry as pass through.*/ +static int __init init_context_pass_through(void) { struct pci_dev *pdev = NULL; + struct dmar_domain *domain; int ret; for_each_pci_dev(pdev) { - if (pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO || - !IS_GFX_DEVICE(pdev)) - continue; - printk(KERN_INFO "IOMMU: gfx device %s 1-1 mapping\n", - pci_name(pdev)); - ret = iommu_prepare_with_active_regions(pdev); + domain = get_domain_for_dev(pdev, DEFAULT_DOMAIN_ADDRESS_WIDTH); + ret = domain_context_mapping(domain, pdev, + CONTEXT_TT_PASS_THROUGH); if (ret) - printk(KERN_ERR "IOMMU: mapping reserved region failed\n"); + return ret; } + return 0; } -#else /* !CONFIG_DMAR_GFX_WA */ -static inline void iommu_prepare_gfx_mapping(void) + +static int md_domain_init(struct dmar_domain *domain, int guest_width); + +static int __init si_domain_work_fn(unsigned long start_pfn, + unsigned long end_pfn, void *datax) { - return; + int *ret = datax; + + *ret = iommu_domain_identity_map(si_domain, + (uint64_t)start_pfn << PAGE_SHIFT, + (uint64_t)end_pfn << PAGE_SHIFT); + return *ret; + } -#endif -#ifdef CONFIG_DMAR_FLOPPY_WA -static inline void iommu_prepare_isa(void) +static int si_domain_init(void) { - struct pci_dev *pdev; - int ret; + struct dmar_drhd_unit *drhd; + struct intel_iommu *iommu; + int nid, ret = 0; - pdev = pci_get_class(PCI_CLASS_BRIDGE_ISA << 8, NULL); - if (!pdev) - return; + si_domain = alloc_domain(); + if (!si_domain) + return -EFAULT; - printk(KERN_INFO "IOMMU: Prepare 0-16M unity mapping for LPC\n"); - ret = iommu_prepare_identity_map(pdev, 0, 16*1024*1024); + pr_debug("Identity mapping domain is domain %d\n", si_domain->id); - if (ret) - printk(KERN_ERR "IOMMU: Failed to create 0-64M identity map, " - "floppy might not work\n"); + for_each_active_iommu(iommu, drhd) { + ret = iommu_attach_domain(si_domain, iommu); + if (ret) { + domain_exit(si_domain); + return -EFAULT; + } + } + if (md_domain_init(si_domain, DEFAULT_DOMAIN_ADDRESS_WIDTH)) { + domain_exit(si_domain); + return -EFAULT; + } + + si_domain->flags = DOMAIN_FLAG_STATIC_IDENTITY; + + for_each_online_node(nid) { + work_with_active_regions(nid, si_domain_work_fn, &ret); + if (ret) + return ret; + } + + return 0; } -#else -static inline void iommu_prepare_isa(void) + +static void domain_remove_one_dev_info(struct dmar_domain *domain, + struct pci_dev *pdev); +static int identity_mapping(struct pci_dev *pdev) { - return; + struct device_domain_info *info; + + if (likely(!iommu_identity_mapping)) + return 0; + + + list_for_each_entry(info, &si_domain->devices, link) + if (info->dev == pdev) + return 1; + return 0; } -#endif /* !CONFIG_DMAR_FLPY_WA */ -static int __init init_dmars(void) +static int domain_add_dev_info(struct dmar_domain *domain, + struct pci_dev *pdev) +{ + struct device_domain_info *info; + unsigned long flags; + + info = alloc_devinfo_mem(); + if (!info) + return -ENOMEM; + + info->segment = pci_domain_nr(pdev->bus); + info->bus = pdev->bus->number; + info->devfn = pdev->devfn; + info->dev = pdev; + info->domain = domain; + + spin_lock_irqsave(&device_domain_lock, flags); + list_add(&info->link, &domain->devices); + list_add(&info->global, &device_domain_list); + pdev->dev.archdata.iommu = info; + spin_unlock_irqrestore(&device_domain_lock, flags); + + return 0; +} + +static int iommu_should_identity_map(struct pci_dev *pdev, int startup) +{ + if (iommu_identity_mapping == 2) + return IS_GFX_DEVICE(pdev); + + /* + * We want to start off with all devices in the 1:1 domain, and + * take them out later if we find they can't access all of memory. + * + * However, we can't do this for PCI devices behind bridges, + * because all PCI devices behind the same bridge will end up + * with the same source-id on their transactions. + * + * Practically speaking, we can't change things around for these + * devices at run-time, because we can't be sure there'll be no + * DMA transactions in flight for any of their siblings. + * + * So PCI devices (unless they're on the root bus) as well as + * their parent PCI-PCI or PCIe-PCI bridges must be left _out_ of + * the 1:1 domain, just in _case_ one of their siblings turns out + * not to be able to map all of memory. + */ + if (!pdev->is_pcie) { + if (!pci_is_root_bus(pdev->bus)) + return 0; + if (pdev->class >> 8 == PCI_CLASS_BRIDGE_PCI) + return 0; + } else if (pdev->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE) + return 0; + + /* + * At boot time, we don't yet know if devices will be 64-bit capable. + * Assume that they will -- if they turn out not to be, then we can + * take them out of the 1:1 domain later. + */ + if (!startup) + return pdev->dma_mask > DMA_BIT_MASK(32); + + return 1; +} + +static int iommu_prepare_static_identity_mapping(void) +{ + struct pci_dev *pdev = NULL; + int ret; + + ret = si_domain_init(); + if (ret) + return -EFAULT; + + for_each_pci_dev(pdev) { + if (iommu_should_identity_map(pdev, 1)) { + printk(KERN_INFO "IOMMU: identity mapping for device %s\n", + pci_name(pdev)); + + ret = domain_context_mapping(si_domain, pdev, + CONTEXT_TT_MULTI_LEVEL); + if (ret) + return ret; + ret = domain_add_dev_info(si_domain, pdev); + if (ret) + return ret; + } + } + + return 0; +} + +int __init init_dmars(void) { struct dmar_drhd_unit *drhd; struct dmar_rmrr_unit *rmrr; struct pci_dev *pdev; struct intel_iommu *iommu; int i, ret; + int pass_through = 1; + + /* + * In case pass through can not be enabled, iommu tries to use identity + * mapping. + */ + if (iommu_pass_through) + iommu_identity_mapping = 1; /* * for each drhd @@ -1917,7 +2260,15 @@ static int __init init_dmars(void) printk(KERN_ERR "IOMMU: allocate root entry failed\n"); goto error; } + if (!ecap_pass_through(iommu->ecap)) + pass_through = 0; } + if (iommu_pass_through) + if (!pass_through) { + printk(KERN_INFO + "Pass Through is not supported by hardware.\n"); + iommu_pass_through = 0; + } /* * Start from the sane iommu hardware state. @@ -1972,45 +2323,63 @@ static int __init init_dmars(void) } } -#ifdef CONFIG_INTR_REMAP - if (!intr_remapping_enabled) { - ret = enable_intr_remapping(0); - if (ret) - printk(KERN_ERR - "IOMMU: enable interrupt remapping failed\n"); + /* + * If pass through is set and enabled, context entries of all pci + * devices are intialized by pass through translation type. + */ + if (iommu_pass_through) { + ret = init_context_pass_through(); + if (ret) { + printk(KERN_ERR "IOMMU: Pass through init failed.\n"); + iommu_pass_through = 0; + } } -#endif /* - * For each rmrr - * for each dev attached to rmrr - * do - * locate drhd for dev, alloc domain for dev - * allocate free domain - * allocate page table entries for rmrr - * if context not allocated for bus - * allocate and init context - * set present in root table for this bus - * init context with domain, translation etc - * endfor - * endfor + * If pass through is not set or not enabled, setup context entries for + * identity mappings for rmrr, gfx, and isa and may fall back to static + * identity mapping if iommu_identity_mapping is set. */ - for_each_rmrr_units(rmrr) { - for (i = 0; i < rmrr->devices_cnt; i++) { - pdev = rmrr->devices[i]; - /* some BIOS lists non-exist devices in DMAR table */ - if (!pdev) - continue; - ret = iommu_prepare_rmrr_dev(rmrr, pdev); - if (ret) - printk(KERN_ERR + if (!iommu_pass_through) { +#ifdef CONFIG_DMAR_BROKEN_GFX_WA + if (!iommu_identity_mapping) + iommu_identity_mapping = 2; +#endif + if (iommu_identity_mapping) + iommu_prepare_static_identity_mapping(); + /* + * For each rmrr + * for each dev attached to rmrr + * do + * locate drhd for dev, alloc domain for dev + * allocate free domain + * allocate page table entries for rmrr + * if context not allocated for bus + * allocate and init context + * set present in root table for this bus + * init context with domain, translation etc + * endfor + * endfor + */ + printk(KERN_INFO "IOMMU: Setting RMRR:\n"); + for_each_rmrr_units(rmrr) { + for (i = 0; i < rmrr->devices_cnt; i++) { + pdev = rmrr->devices[i]; + /* + * some BIOS lists non-exist devices in DMAR + * table. + */ + if (!pdev) + continue; + ret = iommu_prepare_rmrr_dev(rmrr, pdev); + if (ret) + printk(KERN_ERR "IOMMU: mapping reserved region failed\n"); + } } - } - iommu_prepare_gfx_mapping(); - - iommu_prepare_isa(); + iommu_prepare_isa(); + } /* * for each drhd @@ -2032,10 +2401,8 @@ static int __init init_dmars(void) iommu_set_root_entry(iommu); - iommu->flush.flush_context(iommu, 0, 0, 0, DMA_CCMD_GLOBAL_INVL, - 0); - iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_GLOBAL_FLUSH, - 0); + iommu->flush.flush_context(iommu, 0, 0, 0, DMA_CCMD_GLOBAL_INVL); + iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_GLOBAL_FLUSH); iommu_disable_protect_mem_regions(iommu); ret = iommu_enable_translation(iommu); @@ -2055,50 +2422,32 @@ error: return ret; } -static inline u64 aligned_size(u64 host_addr, size_t size) -{ - u64 addr; - addr = (host_addr & (~PAGE_MASK)) + size; - return PAGE_ALIGN(addr); -} - -struct iova * -iommu_alloc_iova(struct dmar_domain *domain, size_t size, u64 end) -{ - struct iova *piova; - - /* Make sure it's in range */ - end = min_t(u64, DOMAIN_MAX_ADDR(domain->gaw), end); - if (!size || (IOVA_START_ADDR + size > end)) - return NULL; - - piova = alloc_iova(&domain->iovad, - size >> PAGE_SHIFT, IOVA_PFN(end), 1); - return piova; -} - -static struct iova * -__intel_alloc_iova(struct device *dev, struct dmar_domain *domain, - size_t size, u64 dma_mask) +/* This takes a number of _MM_ pages, not VTD pages */ +static struct iova *intel_alloc_iova(struct device *dev, + struct dmar_domain *domain, + unsigned long nrpages, uint64_t dma_mask) { struct pci_dev *pdev = to_pci_dev(dev); struct iova *iova = NULL; - if (dma_mask <= DMA_BIT_MASK(32) || dmar_forcedac) - iova = iommu_alloc_iova(domain, size, dma_mask); - else { + /* Restrict dma_mask to the width that the iommu can handle */ + dma_mask = min_t(uint64_t, DOMAIN_MAX_ADDR(domain->gaw), dma_mask); + + if (!dmar_forcedac && dma_mask > DMA_BIT_MASK(32)) { /* * First try to allocate an io virtual address in * DMA_BIT_MASK(32) and if that fails then try allocating * from higher range */ - iova = iommu_alloc_iova(domain, size, DMA_BIT_MASK(32)); - if (!iova) - iova = iommu_alloc_iova(domain, size, dma_mask); - } - - if (!iova) { - printk(KERN_ERR"Allocating iova for %s failed", pci_name(pdev)); + iova = alloc_iova(&domain->iovad, nrpages, + IOVA_PFN(DMA_BIT_MASK(32)), 1); + if (iova) + return iova; + } + iova = alloc_iova(&domain->iovad, nrpages, IOVA_PFN(dma_mask), 1); + if (unlikely(!iova)) { + printk(KERN_ERR "Allocating %ld-page iova for %s failed", + nrpages, pci_name(pdev)); return NULL; } @@ -2121,7 +2470,8 @@ get_valid_domain_for_dev(struct pci_dev *pdev) /* make sure context mapping is ok */ if (unlikely(!domain_context_mapped(pdev))) { - ret = domain_context_mapping(domain, pdev); + ret = domain_context_mapping(domain, pdev, + CONTEXT_TT_MULTI_LEVEL); if (ret) { printk(KERN_ERR "Domain context map for %s failed", @@ -2133,6 +2483,63 @@ get_valid_domain_for_dev(struct pci_dev *pdev) return domain; } +static int iommu_dummy(struct pci_dev *pdev) +{ + return pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO; +} + +/* Check if the pdev needs to go through non-identity map and unmap process.*/ +static int iommu_no_mapping(struct device *dev) +{ + struct pci_dev *pdev; + int found; + + if (unlikely(dev->bus != &pci_bus_type)) + return 1; + + pdev = to_pci_dev(dev); + if (iommu_dummy(pdev)) + return 1; + + if (!iommu_identity_mapping) + return 0; + + found = identity_mapping(pdev); + if (found) { + if (iommu_should_identity_map(pdev, 0)) + return 1; + else { + /* + * 32 bit DMA is removed from si_domain and fall back + * to non-identity mapping. + */ + domain_remove_one_dev_info(si_domain, pdev); + printk(KERN_INFO "32bit %s uses non-identity mapping\n", + pci_name(pdev)); + return 0; + } + } else { + /* + * In case of a detached 64 bit DMA device from vm, the device + * is put into si_domain for identity mapping. + */ + if (iommu_should_identity_map(pdev, 0)) { + int ret; + ret = domain_add_dev_info(si_domain, pdev); + if (ret) + return 0; + ret = domain_context_mapping(si_domain, pdev, CONTEXT_TT_MULTI_LEVEL); + if (!ret) { + printk(KERN_INFO "64bit %s uses identity mapping\n", + pci_name(pdev)); + return 1; + } + } + } + + return 0; +} + static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr, size_t size, int dir, u64 dma_mask) { @@ -2143,9 +2550,11 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr, int prot = 0; int ret; struct intel_iommu *iommu; + unsigned long paddr_pfn = paddr >> PAGE_SHIFT; BUG_ON(dir == DMA_NONE); - if (pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO) + + if (iommu_no_mapping(hwdev)) return paddr; domain = get_valid_domain_for_dev(pdev); @@ -2153,14 +2562,13 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr, return 0; iommu = domain_get_iommu(domain); - size = aligned_size((u64)paddr, size); + size = aligned_nrpages(paddr, size); - iova = __intel_alloc_iova(hwdev, domain, size, pdev->dma_mask); + iova = intel_alloc_iova(hwdev, domain, dma_to_mm_pfn(size), + pdev->dma_mask); if (!iova) goto error; - start_paddr = (phys_addr_t)iova->pfn_lo << PAGE_SHIFT; - /* * Check if DMAR supports zero-length reads on write only * mappings.. @@ -2176,19 +2584,20 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr, * might have two guest_addr mapping to the same host paddr, but this * is not a big problem */ - ret = domain_page_mapping(domain, start_paddr, - ((u64)paddr) & PHYSICAL_PAGE_MASK, - size, prot); + ret = domain_pfn_mapping(domain, mm_to_dma_pfn(iova->pfn_lo), + mm_to_dma_pfn(paddr_pfn), size, prot); if (ret) goto error; - /* it's a non-present to present mapping */ - ret = iommu_flush_iotlb_psi(iommu, domain->id, - start_paddr, size >> VTD_PAGE_SHIFT, 1); - if (ret) + /* 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); + else iommu_flush_write_buffer(iommu); - return start_paddr + ((u64)paddr & (~PAGE_MASK)); + start_paddr = (phys_addr_t)iova->pfn_lo << PAGE_SHIFT; + start_paddr += paddr & ~PAGE_MASK; + return start_paddr; error: if (iova) @@ -2219,15 +2628,22 @@ static void flush_unmaps(void) if (!iommu) continue; - if (deferred_flush[i].next) { - iommu->flush.flush_iotlb(iommu, 0, 0, 0, - DMA_TLB_GLOBAL_FLUSH, 0); - for (j = 0; j < deferred_flush[i].next; j++) { - __free_iova(&deferred_flush[i].domain[j]->iovad, - deferred_flush[i].iova[j]); - } - deferred_flush[i].next = 0; + if (!deferred_flush[i].next) + continue; + + 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 = (iova->pfn_hi - iova->pfn_lo + 1) << PAGE_SHIFT; + mask = ilog2(mask >> VTD_PAGE_SHIFT); + iommu_flush_dev_iotlb(deferred_flush[i].domain[j], + iova->pfn_lo << PAGE_SHIFT, mask); + __free_iova(&deferred_flush[i].domain[j]->iovad, iova); } + deferred_flush[i].next = 0; } list_size = 0; @@ -2274,35 +2690,38 @@ static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr, { struct pci_dev *pdev = to_pci_dev(dev); struct dmar_domain *domain; - unsigned long start_addr; + unsigned long start_pfn, last_pfn; struct iova *iova; struct intel_iommu *iommu; - if (pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO) + if (iommu_no_mapping(dev)) return; + domain = find_domain(pdev); BUG_ON(!domain); iommu = domain_get_iommu(domain); iova = find_iova(&domain->iovad, IOVA_PFN(dev_addr)); - if (!iova) + if (WARN_ONCE(!iova, "Driver unmaps unmatched page at PFN %llx\n", + (unsigned long long)dev_addr)) return; - start_addr = iova->pfn_lo << PAGE_SHIFT; - size = aligned_size((u64)dev_addr, size); + start_pfn = mm_to_dma_pfn(iova->pfn_lo); + last_pfn = mm_to_dma_pfn(iova->pfn_hi + 1) - 1; - pr_debug("Device %s unmapping: %zx@%llx\n", - pci_name(pdev), size, (unsigned long long)start_addr); + pr_debug("Device %s unmapping: pfn %lx-%lx\n", + pci_name(pdev), start_pfn, last_pfn); /* clear the whole page */ - dma_pte_clear_range(domain, start_addr, start_addr + size); + dma_pte_clear_range(domain, start_pfn, last_pfn); + /* free page tables */ - dma_pte_free_pagetable(domain, start_addr, start_addr + size); + dma_pte_free_pagetable(domain, start_pfn, last_pfn); + if (intel_iommu_strict) { - if (iommu_flush_iotlb_psi(iommu, - domain->id, start_addr, size >> VTD_PAGE_SHIFT, 0)) - iommu_flush_write_buffer(iommu); + iommu_flush_iotlb_psi(iommu, domain->id, start_pfn, + last_pfn - start_pfn + 1); /* free iova */ __free_iova(&domain->iovad, iova); } else { @@ -2360,17 +2779,13 @@ static void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, enum dma_data_direction dir, struct dma_attrs *attrs) { - int i; struct pci_dev *pdev = to_pci_dev(hwdev); struct dmar_domain *domain; - unsigned long start_addr; + unsigned long start_pfn, last_pfn; struct iova *iova; - size_t size = 0; - phys_addr_t addr; - struct scatterlist *sg; struct intel_iommu *iommu; - if (pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO) + if (iommu_no_mapping(hwdev)) return; domain = find_domain(pdev); @@ -2379,23 +2794,21 @@ static void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist, iommu = domain_get_iommu(domain); iova = find_iova(&domain->iovad, IOVA_PFN(sglist[0].dma_address)); - if (!iova) + if (WARN_ONCE(!iova, "Driver unmaps unmatched sglist at PFN %llx\n", + (unsigned long long)sglist[0].dma_address)) return; - for_each_sg(sglist, sg, nelems, i) { - addr = page_to_phys(sg_page(sg)) + sg->offset; - size += aligned_size((u64)addr, sg->length); - } - start_addr = iova->pfn_lo << PAGE_SHIFT; + start_pfn = mm_to_dma_pfn(iova->pfn_lo); + last_pfn = mm_to_dma_pfn(iova->pfn_hi + 1) - 1; /* clear the whole page */ - dma_pte_clear_range(domain, start_addr, start_addr + size); + dma_pte_clear_range(domain, start_pfn, last_pfn); + /* free page tables */ - dma_pte_free_pagetable(domain, start_addr, start_addr + size); + dma_pte_free_pagetable(domain, start_pfn, last_pfn); - if (iommu_flush_iotlb_psi(iommu, domain->id, start_addr, - size >> VTD_PAGE_SHIFT, 0)) - iommu_flush_write_buffer(iommu); + iommu_flush_iotlb_psi(iommu, domain->id, start_pfn, + (last_pfn - start_pfn + 1)); /* free iova */ __free_iova(&domain->iovad, iova); @@ -2418,21 +2831,20 @@ static int intel_nontranslate_map_sg(struct device *hddev, static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, enum dma_data_direction dir, struct dma_attrs *attrs) { - phys_addr_t addr; int i; struct pci_dev *pdev = to_pci_dev(hwdev); struct dmar_domain *domain; size_t size = 0; int prot = 0; - size_t offset = 0; + size_t offset_pfn = 0; struct iova *iova = NULL; int ret; struct scatterlist *sg; - unsigned long start_addr; + unsigned long start_vpfn; struct intel_iommu *iommu; BUG_ON(dir == DMA_NONE); - if (pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO) + if (iommu_no_mapping(hwdev)) return intel_nontranslate_map_sg(hwdev, sglist, nelems, dir); domain = get_valid_domain_for_dev(pdev); @@ -2441,12 +2853,11 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne iommu = domain_get_iommu(domain); - for_each_sg(sglist, sg, nelems, i) { - addr = page_to_phys(sg_page(sg)) + sg->offset; - size += aligned_size((u64)addr, sg->length); - } + for_each_sg(sglist, sg, nelems, i) + size += aligned_nrpages(sg->offset, sg->length); - iova = __intel_alloc_iova(hwdev, domain, size, pdev->dma_mask); + iova = intel_alloc_iova(hwdev, domain, dma_to_mm_pfn(size), + pdev->dma_mask); if (!iova) { sglist->dma_length = 0; return 0; @@ -2462,35 +2873,27 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne if (dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL) prot |= DMA_PTE_WRITE; - start_addr = iova->pfn_lo << PAGE_SHIFT; - offset = 0; - for_each_sg(sglist, sg, nelems, i) { - addr = page_to_phys(sg_page(sg)) + sg->offset; - size = aligned_size((u64)addr, sg->length); - ret = domain_page_mapping(domain, start_addr + offset, - ((u64)addr) & PHYSICAL_PAGE_MASK, - size, prot); - if (ret) { - /* clear the page */ - dma_pte_clear_range(domain, start_addr, - start_addr + offset); - /* free page tables */ - dma_pte_free_pagetable(domain, start_addr, - start_addr + offset); - /* free iova */ - __free_iova(&domain->iovad, iova); - return 0; - } - sg->dma_address = start_addr + offset + - ((u64)addr & (~PAGE_MASK)); - sg->dma_length = sg->length; - offset += size; + start_vpfn = mm_to_dma_pfn(iova->pfn_lo); + + ret = domain_sg_mapping(domain, start_vpfn, sglist, size, prot); + if (unlikely(ret)) { + /* clear the page */ + dma_pte_clear_range(domain, start_vpfn, + start_vpfn + size - 1); + /* free page tables */ + dma_pte_free_pagetable(domain, start_vpfn, + start_vpfn + size - 1); + /* free iova */ + __free_iova(&domain->iovad, iova); + return 0; } - /* it's a non-present to present mapping */ - if (iommu_flush_iotlb_psi(iommu, domain->id, - start_addr, offset >> VTD_PAGE_SHIFT, 1)) + /* 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); + else iommu_flush_write_buffer(iommu); + return nelems; } @@ -2649,9 +3052,9 @@ static int init_iommu_hw(void) iommu_set_root_entry(iommu); iommu->flush.flush_context(iommu, 0, 0, 0, - DMA_CCMD_GLOBAL_INVL, 0); + DMA_CCMD_GLOBAL_INVL); iommu->flush.flush_iotlb(iommu, 0, 0, 0, - DMA_TLB_GLOBAL_FLUSH, 0); + DMA_TLB_GLOBAL_FLUSH); iommu_disable_protect_mem_regions(iommu); iommu_enable_translation(iommu); } @@ -2666,9 +3069,9 @@ static void iommu_flush_all(void) for_each_active_iommu(iommu, drhd) { iommu->flush.flush_context(iommu, 0, 0, 0, - DMA_CCMD_GLOBAL_INVL, 0); + DMA_CCMD_GLOBAL_INVL); iommu->flush.flush_iotlb(iommu, 0, 0, 0, - DMA_TLB_GLOBAL_FLUSH, 0); + DMA_TLB_GLOBAL_FLUSH); } } @@ -2791,7 +3194,7 @@ int __init intel_iommu_init(void) * Check the need for DMA-remapping initialization now. * Above initialization will also be used by Interrupt-remapping. */ - if (no_iommu || swiotlb || dmar_disabled) + if (no_iommu || (swiotlb && !iommu_pass_through) || dmar_disabled) return -ENODEV; iommu_init_mempool(); @@ -2811,35 +3214,18 @@ int __init intel_iommu_init(void) init_timer(&unmap_timer); force_iommu = 1; - dma_ops = &intel_dma_ops; - init_iommu_sysfs(); - - register_iommu(&intel_iommu_ops); - - return 0; -} - -static int vm_domain_add_dev_info(struct dmar_domain *domain, - struct pci_dev *pdev) -{ - struct device_domain_info *info; - unsigned long flags; - info = alloc_devinfo_mem(); - if (!info) - return -ENOMEM; + if (!iommu_pass_through) { + printk(KERN_INFO + "Multi-level page-table translation for DMAR.\n"); + dma_ops = &intel_dma_ops; + } else + printk(KERN_INFO + "DMAR: Pass through translation for DMAR.\n"); - info->segment = pci_domain_nr(pdev->bus); - info->bus = pdev->bus->number; - info->devfn = pdev->devfn; - info->dev = pdev; - info->domain = domain; + init_iommu_sysfs(); - spin_lock_irqsave(&device_domain_lock, flags); - list_add(&info->link, &domain->devices); - list_add(&info->global, &device_domain_list); - pdev->dev.archdata.iommu = info; - spin_unlock_irqrestore(&device_domain_lock, flags); + register_iommu(&intel_iommu_ops); return 0; } @@ -2871,7 +3257,7 @@ static void iommu_detach_dependent_devices(struct intel_iommu *iommu, } } -static void vm_domain_remove_one_dev_info(struct dmar_domain *domain, +static void domain_remove_one_dev_info(struct dmar_domain *domain, struct pci_dev *pdev) { struct device_domain_info *info; @@ -2897,6 +3283,7 @@ static void vm_domain_remove_one_dev_info(struct dmar_domain *domain, info->dev->dev.archdata.iommu = NULL; spin_unlock_irqrestore(&device_domain_lock, flags); + iommu_disable_dev_iotlb(info); iommu_detach_dev(iommu, info->bus, info->devfn); iommu_detach_dependent_devices(iommu, pdev); free_devinfo_mem(info); @@ -2947,6 +3334,7 @@ static void vm_domain_remove_all_dev_info(struct dmar_domain *domain) spin_unlock_irqrestore(&device_domain_lock, flags1); + iommu_disable_dev_iotlb(info); iommu = device_to_iommu(info->segment, info->bus, info->devfn); iommu_detach_dev(iommu, info->bus, info->devfn); iommu_detach_dependent_devices(iommu, info->dev); @@ -3002,12 +3390,11 @@ static struct dmar_domain *iommu_alloc_vm_domain(void) return domain; } -static int vm_domain_init(struct dmar_domain *domain, int guest_width) +static int md_domain_init(struct dmar_domain *domain, int guest_width) { int adjust_width; init_iova_domain(&domain->iovad, DMA_32BIT_PFN); - spin_lock_init(&domain->mapping_lock); spin_lock_init(&domain->iommu_lock); domain_reserve_special_ranges(domain); @@ -3021,6 +3408,7 @@ static int vm_domain_init(struct dmar_domain *domain, int guest_width) domain->iommu_count = 0; domain->iommu_coherency = 0; + domain->iommu_snooping = 0; domain->max_addr = 0; /* always allocate the top pgd */ @@ -3061,8 +3449,6 @@ static void iommu_free_vm_domain(struct dmar_domain *domain) static void vm_domain_exit(struct dmar_domain *domain) { - u64 end; - /* Domain 0 is reserved, so dont process it */ if (!domain) return; @@ -3070,14 +3456,12 @@ static void vm_domain_exit(struct dmar_domain *domain) vm_domain_remove_all_dev_info(domain); /* destroy iovas */ put_iova_domain(&domain->iovad); - end = DOMAIN_MAX_ADDR(domain->gaw); - end = end & (~VTD_PAGE_MASK); /* clear ptes */ - dma_pte_clear_range(domain, 0, end); + dma_pte_clear_range(domain, 0, DOMAIN_MAX_PFN(domain->gaw)); /* free page tables */ - dma_pte_free_pagetable(domain, 0, end); + dma_pte_free_pagetable(domain, 0, DOMAIN_MAX_PFN(domain->gaw)); iommu_free_vm_domain(domain); free_domain_mem(domain); @@ -3093,7 +3477,7 @@ static int intel_iommu_domain_init(struct iommu_domain *domain) "intel_iommu_domain_init: dmar_domain == NULL\n"); return -ENOMEM; } - if (vm_domain_init(dmar_domain, DEFAULT_DOMAIN_ADDRESS_WIDTH)) { + if (md_domain_init(dmar_domain, DEFAULT_DOMAIN_ADDRESS_WIDTH)) { printk(KERN_ERR "intel_iommu_domain_init() failed\n"); vm_domain_exit(dmar_domain); @@ -3128,8 +3512,9 @@ static int intel_iommu_attach_device(struct iommu_domain *domain, old_domain = find_domain(pdev); if (old_domain) { - if (dmar_domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE) - vm_domain_remove_one_dev_info(old_domain, pdev); + if (dmar_domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE || + dmar_domain->flags & DOMAIN_FLAG_STATIC_IDENTITY) + domain_remove_one_dev_info(old_domain, pdev); else domain_remove_dev_info(old_domain); } @@ -3151,11 +3536,11 @@ static int intel_iommu_attach_device(struct iommu_domain *domain, return -EFAULT; } - ret = domain_context_mapping(dmar_domain, pdev); + ret = domain_add_dev_info(dmar_domain, pdev); if (ret) return ret; - ret = vm_domain_add_dev_info(dmar_domain, pdev); + ret = domain_context_mapping(dmar_domain, pdev, CONTEXT_TT_MULTI_LEVEL); return ret; } @@ -3165,7 +3550,7 @@ static void intel_iommu_detach_device(struct iommu_domain *domain, struct dmar_domain *dmar_domain = domain->priv; struct pci_dev *pdev = to_pci_dev(dev); - vm_domain_remove_one_dev_info(dmar_domain, pdev); + domain_remove_one_dev_info(dmar_domain, pdev); } static int intel_iommu_map_range(struct iommu_domain *domain, @@ -3185,7 +3570,7 @@ static int intel_iommu_map_range(struct iommu_domain *domain, if ((iommu_prot & IOMMU_CACHE) && dmar_domain->iommu_snooping) prot |= DMA_PTE_SNP; - max_addr = (iova & VTD_PAGE_MASK) + VTD_PAGE_ALIGN(size); + max_addr = iova + size; if (dmar_domain->max_addr < max_addr) { int min_agaw; u64 end; @@ -3203,8 +3588,11 @@ static int intel_iommu_map_range(struct iommu_domain *domain, } dmar_domain->max_addr = max_addr; } - - ret = domain_page_mapping(dmar_domain, iova, hpa, size, prot); + /* Round up size to next multiple of PAGE_SIZE, if it and + the low bits of hpa would take us onto the next page */ + size = aligned_nrpages(hpa, size); + ret = domain_pfn_mapping(dmar_domain, iova >> VTD_PAGE_SHIFT, + hpa >> VTD_PAGE_SHIFT, size, prot); return ret; } @@ -3212,15 +3600,15 @@ static void intel_iommu_unmap_range(struct iommu_domain *domain, unsigned long iova, size_t size) { struct dmar_domain *dmar_domain = domain->priv; - dma_addr_t base; - /* The address might not be aligned */ - base = iova & VTD_PAGE_MASK; - size = VTD_PAGE_ALIGN(size); - dma_pte_clear_range(dmar_domain, base, base + size); + if (!size) + return; + + dma_pte_clear_range(dmar_domain, iova >> VTD_PAGE_SHIFT, + (iova + size - 1) >> VTD_PAGE_SHIFT); - if (dmar_domain->max_addr == base + size) - dmar_domain->max_addr = base; + if (dmar_domain->max_addr == iova + size) + dmar_domain->max_addr = iova; } static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain, @@ -3230,7 +3618,7 @@ static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain, struct dma_pte *pte; u64 phys = 0; - pte = addr_to_dma_pte(dmar_domain, iova); + pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT); if (pte) phys = dma_pte_addr(pte); diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c index f5e0ea724a6..4f5b8712931 100644 --- a/drivers/pci/intr_remapping.c +++ b/drivers/pci/intr_remapping.c @@ -10,11 +10,21 @@ #include <linux/intel-iommu.h> #include "intr_remapping.h" #include <acpi/acpi.h> +#include <asm/pci-direct.h> +#include "pci.h" static struct ioapic_scope ir_ioapic[MAX_IO_APICS]; static int ir_ioapic_num; int intr_remapping_enabled; +static int disable_intremap; +static __init int setup_nointremap(char *str) +{ + disable_intremap = 1; + return 0; +} +early_param("nointremap", setup_nointremap); + struct irq_2_iommu { struct intel_iommu *iommu; u16 irte_index; @@ -23,15 +33,12 @@ struct irq_2_iommu { }; #ifdef CONFIG_GENERIC_HARDIRQS -static struct irq_2_iommu *get_one_free_irq_2_iommu(int cpu) +static struct irq_2_iommu *get_one_free_irq_2_iommu(int node) { struct irq_2_iommu *iommu; - int node; - - node = cpu_to_node(cpu); iommu = kzalloc_node(sizeof(*iommu), GFP_ATOMIC, node); - printk(KERN_DEBUG "alloc irq_2_iommu on cpu %d node %d\n", cpu, node); + printk(KERN_DEBUG "alloc irq_2_iommu on node %d\n", node); return iommu; } @@ -48,7 +55,7 @@ static struct irq_2_iommu *irq_2_iommu(unsigned int irq) return desc->irq_2_iommu; } -static struct irq_2_iommu *irq_2_iommu_alloc_cpu(unsigned int irq, int cpu) +static struct irq_2_iommu *irq_2_iommu_alloc_node(unsigned int irq, int node) { struct irq_desc *desc; struct irq_2_iommu *irq_iommu; @@ -56,7 +63,7 @@ static struct irq_2_iommu *irq_2_iommu_alloc_cpu(unsigned int irq, int cpu) /* * alloc irq desc if not allocated already. */ - desc = irq_to_desc_alloc_cpu(irq, cpu); + desc = irq_to_desc_alloc_node(irq, node); if (!desc) { printk(KERN_INFO "can not get irq_desc for %d\n", irq); return NULL; @@ -65,14 +72,14 @@ static struct irq_2_iommu *irq_2_iommu_alloc_cpu(unsigned int irq, int cpu) irq_iommu = desc->irq_2_iommu; if (!irq_iommu) - desc->irq_2_iommu = get_one_free_irq_2_iommu(cpu); + desc->irq_2_iommu = get_one_free_irq_2_iommu(node); return desc->irq_2_iommu; } static struct irq_2_iommu *irq_2_iommu_alloc(unsigned int irq) { - return irq_2_iommu_alloc_cpu(irq, boot_cpu_id); + return irq_2_iommu_alloc_node(irq, cpu_to_node(boot_cpu_id)); } #else /* !CONFIG_SPARSE_IRQ */ @@ -309,7 +316,8 @@ int modify_irte(int irq, struct irte *irte_modified) index = irq_iommu->irte_index + irq_iommu->sub_handle; irte = &iommu->ir_table->base[index]; - set_64bit((unsigned long *)irte, irte_modified->low); + set_64bit((unsigned long *)&irte->low, irte_modified->low); + set_64bit((unsigned long *)&irte->high, irte_modified->high); __iommu_flush_cache(iommu, irte, sizeof(*irte)); rc = qi_flush_iec(iommu, index, 0); @@ -364,12 +372,32 @@ struct intel_iommu *map_dev_to_ir(struct pci_dev *dev) return drhd->iommu; } +static int clear_entries(struct irq_2_iommu *irq_iommu) +{ + struct irte *start, *entry, *end; + struct intel_iommu *iommu; + int index; + + if (irq_iommu->sub_handle) + return 0; + + iommu = irq_iommu->iommu; + index = irq_iommu->irte_index + irq_iommu->sub_handle; + + start = iommu->ir_table->base + index; + end = start + (1 << irq_iommu->irte_mask); + + for (entry = start; entry < end; entry++) { + set_64bit((unsigned long *)&entry->low, 0); + set_64bit((unsigned long *)&entry->high, 0); + } + + return qi_flush_iec(iommu, index, irq_iommu->irte_mask); +} + int free_irte(int irq) { int rc = 0; - int index, i; - struct irte *irte; - struct intel_iommu *iommu; struct irq_2_iommu *irq_iommu; unsigned long flags; @@ -380,16 +408,7 @@ int free_irte(int irq) return -1; } - iommu = irq_iommu->iommu; - - index = irq_iommu->irte_index + irq_iommu->sub_handle; - irte = &iommu->ir_table->base[index]; - - if (!irq_iommu->sub_handle) { - for (i = 0; i < (1 << irq_iommu->irte_mask); i++) - set_64bit((unsigned long *)(irte + i), 0); - rc = qi_flush_iec(iommu, index, irq_iommu->irte_mask); - } + rc = clear_entries(irq_iommu); irq_iommu->iommu = NULL; irq_iommu->irte_index = 0; @@ -401,10 +420,95 @@ int free_irte(int irq) return rc; } +/* + * source validation type + */ +#define SVT_NO_VERIFY 0x0 /* no verification is required */ +#define SVT_VERIFY_SID_SQ 0x1 /* verify using SID and SQ fiels */ +#define SVT_VERIFY_BUS 0x2 /* verify bus of request-id */ + +/* + * source-id qualifier + */ +#define SQ_ALL_16 0x0 /* verify all 16 bits of request-id */ +#define SQ_13_IGNORE_1 0x1 /* verify most significant 13 bits, ignore + * the third least significant bit + */ +#define SQ_13_IGNORE_2 0x2 /* verify most significant 13 bits, ignore + * the second and third least significant bits + */ +#define SQ_13_IGNORE_3 0x3 /* verify most significant 13 bits, ignore + * the least three significant bits + */ + +/* + * set SVT, SQ and SID fields of irte to verify + * source ids of interrupt requests + */ +static void set_irte_sid(struct irte *irte, unsigned int svt, + unsigned int sq, unsigned int sid) +{ + irte->svt = svt; + irte->sq = sq; + irte->sid = sid; +} + +int set_ioapic_sid(struct irte *irte, int apic) +{ + int i; + u16 sid = 0; + + if (!irte) + return -1; + + for (i = 0; i < MAX_IO_APICS; i++) { + if (ir_ioapic[i].id == apic) { + sid = (ir_ioapic[i].bus << 8) | ir_ioapic[i].devfn; + break; + } + } + + if (sid == 0) { + pr_warning("Failed to set source-id of IOAPIC (%d)\n", apic); + return -1; + } + + set_irte_sid(irte, 1, 0, sid); + + return 0; +} + +int set_msi_sid(struct irte *irte, struct pci_dev *dev) +{ + struct pci_dev *bridge; + + if (!irte || !dev) + return -1; + + /* PCIe device or Root Complex integrated PCI device */ + if (dev->is_pcie || !dev->bus->parent) { + set_irte_sid(irte, SVT_VERIFY_SID_SQ, SQ_ALL_16, + (dev->bus->number << 8) | dev->devfn); + return 0; + } + + bridge = pci_find_upstream_pcie_bridge(dev); + if (bridge) { + if (bridge->is_pcie) /* this is a PCIE-to-PCI/PCIX bridge */ + set_irte_sid(irte, SVT_VERIFY_BUS, SQ_ALL_16, + (bridge->bus->number << 8) | dev->bus->number); + else /* this is a legacy PCI bridge */ + set_irte_sid(irte, SVT_VERIFY_SID_SQ, SQ_ALL_16, + (bridge->bus->number << 8) | bridge->devfn); + } + + return 0; +} + static void iommu_set_intr_remapping(struct intel_iommu *iommu, int mode) { u64 addr; - u32 cmd, sts; + u32 sts; unsigned long flags; addr = virt_to_phys((void *)iommu->ir_table->base); @@ -415,28 +519,13 @@ static void iommu_set_intr_remapping(struct intel_iommu *iommu, int mode) (addr) | IR_X2APIC_MODE(mode) | INTR_REMAP_TABLE_REG_SIZE); /* Set interrupt-remapping table pointer */ - cmd = iommu->gcmd | DMA_GCMD_SIRTP; iommu->gcmd |= DMA_GCMD_SIRTP; - writel(cmd, iommu->reg + DMAR_GCMD_REG); + writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG); IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, readl, (sts & DMA_GSTS_IRTPS), sts); spin_unlock_irqrestore(&iommu->register_lock, flags); - if (mode == 0) { - spin_lock_irqsave(&iommu->register_lock, flags); - - /* enable comaptiblity format interrupt pass through */ - cmd = iommu->gcmd | DMA_GCMD_CFI; - iommu->gcmd |= DMA_GCMD_CFI; - writel(cmd, iommu->reg + DMAR_GCMD_REG); - - IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, - readl, (sts & DMA_GSTS_CFIS), sts); - - spin_unlock_irqrestore(&iommu->register_lock, flags); - } - /* * global invalidation of interrupt entry cache before enabling * interrupt-remapping. @@ -446,9 +535,8 @@ static void iommu_set_intr_remapping(struct intel_iommu *iommu, int mode) spin_lock_irqsave(&iommu->register_lock, flags); /* Enable interrupt-remapping */ - cmd = iommu->gcmd | DMA_GCMD_IRE; iommu->gcmd |= DMA_GCMD_IRE; - writel(cmd, iommu->reg + DMAR_GCMD_REG); + writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG); IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, readl, (sts & DMA_GSTS_IRES), sts); @@ -516,6 +604,23 @@ end: spin_unlock_irqrestore(&iommu->register_lock, flags); } +int __init intr_remapping_supported(void) +{ + struct dmar_drhd_unit *drhd; + + if (disable_intremap) + return 0; + + for_each_drhd_unit(drhd) { + struct intel_iommu *iommu = drhd->iommu; + + if (!ecap_ir_support(iommu->ecap)) + return 0; + } + + return 1; +} + int __init enable_intr_remapping(int eim) { struct dmar_drhd_unit *drhd; @@ -606,6 +711,35 @@ error: return -1; } +static void ir_parse_one_ioapic_scope(struct acpi_dmar_device_scope *scope, + struct intel_iommu *iommu) +{ + struct acpi_dmar_pci_path *path; + u8 bus; + int count; + + bus = scope->bus; + path = (struct acpi_dmar_pci_path *)(scope + 1); + count = (scope->length - sizeof(struct acpi_dmar_device_scope)) + / sizeof(struct acpi_dmar_pci_path); + + while (--count > 0) { + /* + * Access PCI directly due to the PCI + * subsystem isn't initialized yet. + */ + bus = read_pci_config_byte(bus, path->dev, path->fn, + PCI_SECONDARY_BUS); + path++; + } + + ir_ioapic[ir_ioapic_num].bus = bus; + ir_ioapic[ir_ioapic_num].devfn = PCI_DEVFN(path->dev, path->fn); + ir_ioapic[ir_ioapic_num].iommu = iommu; + ir_ioapic[ir_ioapic_num].id = scope->enumeration_id; + ir_ioapic_num++; +} + static int ir_parse_ioapic_scope(struct acpi_dmar_header *header, struct intel_iommu *iommu) { @@ -630,9 +764,7 @@ static int ir_parse_ioapic_scope(struct acpi_dmar_header *header, " 0x%Lx\n", scope->enumeration_id, drhd->address); - ir_ioapic[ir_ioapic_num].iommu = iommu; - ir_ioapic[ir_ioapic_num].id = scope->enumeration_id; - ir_ioapic_num++; + ir_parse_one_ioapic_scope(scope, iommu); } start += scope->length; } diff --git a/drivers/pci/intr_remapping.h b/drivers/pci/intr_remapping.h index ca48f0df8ac..63a263c1841 100644 --- a/drivers/pci/intr_remapping.h +++ b/drivers/pci/intr_remapping.h @@ -3,6 +3,8 @@ struct ioapic_scope { struct intel_iommu *iommu; unsigned int id; + unsigned int bus; /* PCI bus number */ + unsigned int devfn; /* PCI devfn number */ }; #define IR_X2APIC_MODE(mode) (mode ? (1 << 11) : 0) diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c index b497daab3d4..e3a87210e94 100644 --- a/drivers/pci/iov.c +++ b/drivers/pci/iov.c @@ -5,6 +5,7 @@ * * PCI Express I/O Virtualization (IOV) support. * Single Root IOV 1.0 + * Address Translation Service 1.0 */ #include <linux/pci.h> @@ -110,7 +111,7 @@ static int virtfn_add(struct pci_dev *dev, int id, int reset) } if (reset) - pci_execute_reset_function(virtfn); + __pci_reset_function(virtfn); pci_device_add(virtfn, virtfn->bus); mutex_unlock(&iov->dev->sriov->lock); @@ -164,7 +165,7 @@ static void virtfn_remove(struct pci_dev *dev, int id, int reset) if (reset) { device_release_driver(&virtfn->dev); - pci_execute_reset_function(virtfn); + __pci_reset_function(virtfn); } sprintf(buf, "virtfn%u", id); @@ -487,13 +488,15 @@ found: iov->self = dev; pci_read_config_dword(dev, pos + PCI_SRIOV_CAP, &iov->cap); pci_read_config_byte(dev, pos + PCI_SRIOV_FUNC_LINK, &iov->link); + if (dev->pcie_type == PCI_EXP_TYPE_RC_END) + iov->link = PCI_DEVFN(PCI_SLOT(dev->devfn), iov->link); if (pdev) iov->dev = pci_dev_get(pdev); - else { + else iov->dev = dev; - mutex_init(&iov->lock); - } + + mutex_init(&iov->lock); dev->sriov = iov; dev->is_physfn = 1; @@ -513,11 +516,11 @@ static void sriov_release(struct pci_dev *dev) { BUG_ON(dev->sriov->nr_virtfn); - if (dev == dev->sriov->dev) - mutex_destroy(&dev->sriov->lock); - else + if (dev != dev->sriov->dev) pci_dev_put(dev->sriov->dev); + mutex_destroy(&dev->sriov->lock); + kfree(dev->sriov); dev->sriov = NULL; } @@ -679,3 +682,145 @@ irqreturn_t pci_sriov_migration(struct pci_dev *dev) return sriov_migration(dev) ? IRQ_HANDLED : IRQ_NONE; } EXPORT_SYMBOL_GPL(pci_sriov_migration); + +static int ats_alloc_one(struct pci_dev *dev, int ps) +{ + int pos; + u16 cap; + struct pci_ats *ats; + + pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ATS); + if (!pos) + return -ENODEV; + + ats = kzalloc(sizeof(*ats), GFP_KERNEL); + if (!ats) + return -ENOMEM; + + ats->pos = pos; + ats->stu = ps; + pci_read_config_word(dev, pos + PCI_ATS_CAP, &cap); + ats->qdep = PCI_ATS_CAP_QDEP(cap) ? PCI_ATS_CAP_QDEP(cap) : + PCI_ATS_MAX_QDEP; + dev->ats = ats; + + return 0; +} + +static void ats_free_one(struct pci_dev *dev) +{ + kfree(dev->ats); + dev->ats = NULL; +} + +/** + * pci_enable_ats - enable the ATS capability + * @dev: the PCI device + * @ps: the IOMMU page shift + * + * Returns 0 on success, or negative on failure. + */ +int pci_enable_ats(struct pci_dev *dev, int ps) +{ + int rc; + u16 ctrl; + + BUG_ON(dev->ats && dev->ats->is_enabled); + + if (ps < PCI_ATS_MIN_STU) + return -EINVAL; + + if (dev->is_physfn || dev->is_virtfn) { + struct pci_dev *pdev = dev->is_physfn ? dev : dev->physfn; + + mutex_lock(&pdev->sriov->lock); + if (pdev->ats) + rc = pdev->ats->stu == ps ? 0 : -EINVAL; + else + rc = ats_alloc_one(pdev, ps); + + if (!rc) + pdev->ats->ref_cnt++; + mutex_unlock(&pdev->sriov->lock); + if (rc) + return rc; + } + + if (!dev->is_physfn) { + rc = ats_alloc_one(dev, ps); + if (rc) + return rc; + } + + ctrl = PCI_ATS_CTRL_ENABLE; + if (!dev->is_virtfn) + ctrl |= PCI_ATS_CTRL_STU(ps - PCI_ATS_MIN_STU); + pci_write_config_word(dev, dev->ats->pos + PCI_ATS_CTRL, ctrl); + + dev->ats->is_enabled = 1; + + return 0; +} + +/** + * pci_disable_ats - disable the ATS capability + * @dev: the PCI device + */ +void pci_disable_ats(struct pci_dev *dev) +{ + u16 ctrl; + + BUG_ON(!dev->ats || !dev->ats->is_enabled); + + pci_read_config_word(dev, dev->ats->pos + PCI_ATS_CTRL, &ctrl); + ctrl &= ~PCI_ATS_CTRL_ENABLE; + pci_write_config_word(dev, dev->ats->pos + PCI_ATS_CTRL, ctrl); + + dev->ats->is_enabled = 0; + + if (dev->is_physfn || dev->is_virtfn) { + struct pci_dev *pdev = dev->is_physfn ? dev : dev->physfn; + + mutex_lock(&pdev->sriov->lock); + pdev->ats->ref_cnt--; + if (!pdev->ats->ref_cnt) + ats_free_one(pdev); + mutex_unlock(&pdev->sriov->lock); + } + + if (!dev->is_physfn) + ats_free_one(dev); +} + +/** + * pci_ats_queue_depth - query the ATS Invalidate Queue Depth + * @dev: the PCI device + * + * Returns the queue depth on success, or negative on failure. + * + * The ATS spec uses 0 in the Invalidate Queue Depth field to + * indicate that the function can accept 32 Invalidate Request. + * But here we use the `real' values (i.e. 1~32) for the Queue + * Depth; and 0 indicates the function shares the Queue with + * other functions (doesn't exclusively own a Queue). + */ +int pci_ats_queue_depth(struct pci_dev *dev) +{ + int pos; + u16 cap; + + if (dev->is_virtfn) + return 0; + + if (dev->ats) + return dev->ats->qdep; + + pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ATS); + if (!pos) + return -ENODEV; + + pci_read_config_word(dev, pos + PCI_ATS_CAP, &cap); + + return PCI_ATS_CAP_QDEP(cap) ? PCI_ATS_CAP_QDEP(cap) : + PCI_ATS_MAX_QDEP; +} diff --git a/drivers/pci/iova.c b/drivers/pci/iova.c index 2287116e982..46dd440e231 100644 --- a/drivers/pci/iova.c +++ b/drivers/pci/iova.c @@ -1,9 +1,19 @@ /* - * Copyright (c) 2006, Intel Corporation. + * Copyright © 2006-2009, Intel Corporation. * - * This file is released under the GPLv2. + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program; if not, write to the Free Software Foundation, Inc., 59 Temple + * Place - Suite 330, Boston, MA 02111-1307 USA. * - * Copyright (C) 2006-2008 Intel Corporation * Author: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com> */ @@ -123,7 +133,15 @@ move_left: /* Insert the new_iova into domain rbtree by holding writer lock */ /* Add new node and rebalance tree. */ { - struct rb_node **entry = &((prev)), *parent = NULL; + struct rb_node **entry, *parent = NULL; + + /* If we have 'prev', it's a valid place to start the + insertion. Otherwise, start from the root. */ + if (prev) + entry = &prev; + else + entry = &iovad->rbroot.rb_node; + /* Figure out where to put new node */ while (*entry) { struct iova *this = container_of(*entry, diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index 362773247fb..d986afb7032 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -75,22 +75,17 @@ void arch_teardown_msi_irqs(struct pci_dev *dev) } #endif -static void __msi_set_enable(struct pci_dev *dev, int pos, int enable) +static void msi_set_enable(struct pci_dev *dev, int pos, int enable) { u16 control; - if (pos) { - pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &control); - control &= ~PCI_MSI_FLAGS_ENABLE; - if (enable) - control |= PCI_MSI_FLAGS_ENABLE; - pci_write_config_word(dev, pos + PCI_MSI_FLAGS, control); - } -} + BUG_ON(!pos); -static void msi_set_enable(struct pci_dev *dev, int enable) -{ - __msi_set_enable(dev, pci_find_capability(dev, PCI_CAP_ID_MSI), enable); + pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &control); + control &= ~PCI_MSI_FLAGS_ENABLE; + if (enable) + control |= PCI_MSI_FLAGS_ENABLE; + pci_write_config_word(dev, pos + PCI_MSI_FLAGS, control); } static void msix_set_enable(struct pci_dev *dev, int enable) @@ -131,21 +126,24 @@ static inline __attribute_const__ u32 msi_enabled_mask(u16 control) * mask all MSI interrupts by clearing the MSI enable bit does not work * reliably as devices without an INTx disable bit will then generate a * level IRQ which will never be cleared. - * - * Returns 1 if it succeeded in masking the interrupt and 0 if the device - * doesn't support MSI masking. */ -static void msi_mask_irq(struct msi_desc *desc, u32 mask, u32 flag) +static u32 __msi_mask_irq(struct msi_desc *desc, u32 mask, u32 flag) { u32 mask_bits = desc->masked; if (!desc->msi_attrib.maskbit) - return; + return 0; mask_bits &= ~mask; mask_bits |= flag; pci_write_config_dword(desc->dev, desc->mask_pos, mask_bits); - desc->masked = mask_bits; + + return mask_bits; +} + +static void msi_mask_irq(struct msi_desc *desc, u32 mask, u32 flag) +{ + desc->masked = __msi_mask_irq(desc, mask, flag); } /* @@ -155,15 +153,21 @@ static void msi_mask_irq(struct msi_desc *desc, u32 mask, u32 flag) * file. This saves a few milliseconds when initialising devices with lots * of MSI-X interrupts. */ -static void msix_mask_irq(struct msi_desc *desc, u32 flag) +static u32 __msix_mask_irq(struct msi_desc *desc, u32 flag) { u32 mask_bits = desc->masked; unsigned offset = desc->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE + - PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET; + PCI_MSIX_ENTRY_VECTOR_CTRL; mask_bits &= ~1; mask_bits |= flag; writel(mask_bits, desc->mask_base + offset); - desc->masked = mask_bits; + + return mask_bits; +} + +static void msix_mask_irq(struct msi_desc *desc, u32 flag) +{ + desc->masked = __msix_mask_irq(desc, flag); } static void msi_set_mask_bit(unsigned irq, u32 flag) @@ -196,9 +200,9 @@ void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) void __iomem *base = entry->mask_base + entry->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE; - msg->address_lo = readl(base + PCI_MSIX_ENTRY_LOWER_ADDR_OFFSET); - msg->address_hi = readl(base + PCI_MSIX_ENTRY_UPPER_ADDR_OFFSET); - msg->data = readl(base + PCI_MSIX_ENTRY_DATA_OFFSET); + msg->address_lo = readl(base + PCI_MSIX_ENTRY_LOWER_ADDR); + msg->address_hi = readl(base + PCI_MSIX_ENTRY_UPPER_ADDR); + msg->data = readl(base + PCI_MSIX_ENTRY_DATA); } else { struct pci_dev *dev = entry->dev; int pos = entry->msi_attrib.pos; @@ -233,11 +237,9 @@ void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) base = entry->mask_base + entry->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE; - writel(msg->address_lo, - base + PCI_MSIX_ENTRY_LOWER_ADDR_OFFSET); - writel(msg->address_hi, - base + PCI_MSIX_ENTRY_UPPER_ADDR_OFFSET); - writel(msg->data, base + PCI_MSIX_ENTRY_DATA_OFFSET); + writel(msg->address_lo, base + PCI_MSIX_ENTRY_LOWER_ADDR); + writel(msg->address_hi, base + PCI_MSIX_ENTRY_UPPER_ADDR); + writel(msg->data, base + PCI_MSIX_ENTRY_DATA); } else { struct pci_dev *dev = entry->dev; int pos = entry->msi_attrib.pos; @@ -303,7 +305,7 @@ static void __pci_restore_msi_state(struct pci_dev *dev) pos = entry->msi_attrib.pos; pci_intx_for_msi(dev, 0); - msi_set_enable(dev, 0); + msi_set_enable(dev, pos, 0); write_msi_msg(dev->irq, &entry->msg); pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &control); @@ -321,22 +323,22 @@ static void __pci_restore_msix_state(struct pci_dev *dev) if (!dev->msix_enabled) return; + BUG_ON(list_empty(&dev->msi_list)); + entry = list_entry(dev->msi_list.next, struct msi_desc, list); + pos = entry->msi_attrib.pos; + pci_read_config_word(dev, pos + PCI_MSIX_FLAGS, &control); /* route the table */ pci_intx_for_msi(dev, 0); - msix_set_enable(dev, 0); + control |= PCI_MSIX_FLAGS_ENABLE | PCI_MSIX_FLAGS_MASKALL; + pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control); list_for_each_entry(entry, &dev->msi_list, list) { write_msi_msg(entry->irq, &entry->msg); msix_mask_irq(entry, entry->masked); } - BUG_ON(list_empty(&dev->msi_list)); - entry = list_entry(dev->msi_list.next, struct msi_desc, list); - pos = entry->msi_attrib.pos; - pci_read_config_word(dev, pos + PCI_MSIX_FLAGS, &control); control &= ~PCI_MSIX_FLAGS_MASKALL; - control |= PCI_MSIX_FLAGS_ENABLE; pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control); } @@ -365,9 +367,9 @@ static int msi_capability_init(struct pci_dev *dev, int nvec) u16 control; unsigned mask; - msi_set_enable(dev, 0); /* Ensure msi is disabled as I set it up */ - pos = pci_find_capability(dev, PCI_CAP_ID_MSI); + msi_set_enable(dev, pos, 0); /* Disable MSI during set up */ + pci_read_config_word(dev, msi_control_reg(pos), &control); /* MSI Entry Initialization */ entry = alloc_msi_entry(dev); @@ -381,7 +383,7 @@ static int msi_capability_init(struct pci_dev *dev, int nvec) entry->msi_attrib.default_irq = dev->irq; /* Save IOAPIC IRQ */ entry->msi_attrib.pos = pos; - entry->mask_pos = msi_mask_bits_reg(pos, entry->msi_attrib.is_64); + entry->mask_pos = msi_mask_reg(pos, entry->msi_attrib.is_64); /* All MSIs are unmasked by default, Mask them all */ if (entry->msi_attrib.maskbit) pci_read_config_dword(dev, entry->mask_pos, &entry->masked); @@ -393,13 +395,14 @@ static int msi_capability_init(struct pci_dev *dev, int nvec) /* Configure MSI capability structure */ ret = arch_setup_msi_irqs(dev, nvec, PCI_CAP_ID_MSI); if (ret) { + msi_mask_irq(entry, mask, ~mask); msi_free_irqs(dev); return ret; } /* Set MSI enabled bits */ pci_intx_for_msi(dev, 0); - msi_set_enable(dev, 1); + msi_set_enable(dev, pos, 1); dev->msi_enabled = 1; dev->irq = entry->irq; @@ -427,11 +430,14 @@ static int msix_capability_init(struct pci_dev *dev, u8 bir; void __iomem *base; - msix_set_enable(dev, 0);/* Ensure msix is disabled as I set it up */ - pos = pci_find_capability(dev, PCI_CAP_ID_MSIX); + pci_read_config_word(dev, pos + PCI_MSIX_FLAGS, &control); + + /* Ensure MSI-X is disabled while it is set up */ + control &= ~PCI_MSIX_FLAGS_ENABLE; + pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control); + /* Request & Map MSI-X table region */ - pci_read_config_word(dev, msi_control_reg(pos), &control); nr_entries = multi_msix_capable(control); pci_read_config_dword(dev, msix_table_offset_reg(pos), &table_offset); @@ -442,11 +448,16 @@ static int msix_capability_init(struct pci_dev *dev, if (base == NULL) return -ENOMEM; - /* MSI-X Table Initialization */ for (i = 0; i < nvec; i++) { entry = alloc_msi_entry(dev); - if (!entry) - break; + if (!entry) { + if (!i) + iounmap(base); + else + msi_free_irqs(dev); + /* No enough memory. Don't try again */ + return -ENOMEM; + } j = entries[i].entry; entry->msi_attrib.is_msix = 1; @@ -455,7 +466,6 @@ static int msix_capability_init(struct pci_dev *dev, entry->msi_attrib.default_irq = dev->irq; entry->msi_attrib.pos = pos; entry->mask_base = base; - msix_mask_irq(entry, 1); list_add_tail(&entry->list, &dev->msi_list); } @@ -480,22 +490,31 @@ static int msix_capability_init(struct pci_dev *dev, return ret; } + /* + * Some devices require MSI-X to be enabled before we can touch the + * MSI-X registers. We need to mask all the vectors to prevent + * interrupts coming in before they're fully set up. + */ + control |= PCI_MSIX_FLAGS_MASKALL | PCI_MSIX_FLAGS_ENABLE; + pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control); + i = 0; list_for_each_entry(entry, &dev->msi_list, list) { entries[i].vector = entry->irq; set_irq_msi(entry->irq, entry); + j = entries[i].entry; + entry->masked = readl(base + j * PCI_MSIX_ENTRY_SIZE + + PCI_MSIX_ENTRY_VECTOR_CTRL); + msix_mask_irq(entry, 1); i++; } - /* Set MSI-X enabled bits */ + + /* Set MSI-X enabled bits and unmask the function */ pci_intx_for_msi(dev, 0); - msix_set_enable(dev, 1); dev->msix_enabled = 1; - list_for_each_entry(entry, &dev->msi_list, list) { - int vector = entry->msi_attrib.entry_nr; - entry->masked = readl(base + vector * PCI_MSIX_ENTRY_SIZE + - PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET); - } + control &= ~PCI_MSIX_FLAGS_MASKALL; + pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control); return 0; } @@ -596,19 +615,24 @@ void pci_msi_shutdown(struct pci_dev *dev) struct msi_desc *desc; u32 mask; u16 ctrl; + unsigned pos; if (!pci_msi_enable || !dev || !dev->msi_enabled) return; - msi_set_enable(dev, 0); + BUG_ON(list_empty(&dev->msi_list)); + desc = list_first_entry(&dev->msi_list, struct msi_desc, list); + pos = desc->msi_attrib.pos; + + msi_set_enable(dev, pos, 0); pci_intx_for_msi(dev, 1); dev->msi_enabled = 0; - BUG_ON(list_empty(&dev->msi_list)); - desc = list_first_entry(&dev->msi_list, struct msi_desc, list); - pci_read_config_word(dev, desc->msi_attrib.pos + PCI_MSI_FLAGS, &ctrl); + /* Return the device with MSI unmasked as initial states */ + pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &ctrl); mask = msi_capable_mask(ctrl); - msi_mask_irq(desc, mask, ~mask); + /* Keep cached state to be restored */ + __msi_mask_irq(desc, mask, ~mask); /* Restore dev->irq to its default pin-assertion irq */ dev->irq = desc->msi_attrib.default_irq; @@ -648,10 +672,6 @@ static int msi_free_irqs(struct pci_dev* dev) list_for_each_entry_safe(entry, tmp, &dev->msi_list, list) { if (entry->msi_attrib.is_msix) { - writel(1, entry->mask_base + entry->msi_attrib.entry_nr - * PCI_MSIX_ENTRY_SIZE - + PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET); - if (list_is_last(&entry->list, &dev->msi_list)) iounmap(entry->mask_base); } @@ -691,8 +711,8 @@ int pci_msix_table_size(struct pci_dev *dev) * indicates the successful configuration of MSI-X capability structure * with new allocated MSI-X irqs. A return of < 0 indicates a failure. * Or a return of > 0 indicates that driver request is exceeding the number - * of irqs available. Driver should use the returned value to re-send - * its request. + * of irqs or MSI-X vectors available. Driver should use the returned value to + * re-send its request. **/ int pci_enable_msix(struct pci_dev* dev, struct msix_entry *entries, int nvec) { @@ -708,7 +728,7 @@ int pci_enable_msix(struct pci_dev* dev, struct msix_entry *entries, int nvec) nr_entries = pci_msix_table_size(dev); if (nvec > nr_entries) - return -EINVAL; + return nr_entries; /* Check for any invalid entries */ for (i = 0; i < nvec; i++) { @@ -739,9 +759,17 @@ static void msix_free_all_irqs(struct pci_dev *dev) void pci_msix_shutdown(struct pci_dev* dev) { + struct msi_desc *entry; + if (!pci_msi_enable || !dev || !dev->msix_enabled) return; + /* Return the device with MSI-X masked as initial states */ + list_for_each_entry(entry, &dev->msi_list, list) { + /* Keep cached states to be restored */ + __msix_mask_irq(entry, 1); + } + msix_set_enable(dev, 0); pci_intx_for_msi(dev, 1); dev->msix_enabled = 0; diff --git a/drivers/pci/msi.h b/drivers/pci/msi.h index 71f4df2ef65..de27c1cb5a2 100644 --- a/drivers/pci/msi.h +++ b/drivers/pci/msi.h @@ -6,31 +6,25 @@ #ifndef MSI_H #define MSI_H -#define PCI_MSIX_ENTRY_SIZE 16 -#define PCI_MSIX_ENTRY_LOWER_ADDR_OFFSET 0 -#define PCI_MSIX_ENTRY_UPPER_ADDR_OFFSET 4 -#define PCI_MSIX_ENTRY_DATA_OFFSET 8 -#define PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET 12 +#define PCI_MSIX_ENTRY_SIZE 16 +#define PCI_MSIX_ENTRY_LOWER_ADDR 0 +#define PCI_MSIX_ENTRY_UPPER_ADDR 4 +#define PCI_MSIX_ENTRY_DATA 8 +#define PCI_MSIX_ENTRY_VECTOR_CTRL 12 #define msi_control_reg(base) (base + PCI_MSI_FLAGS) #define msi_lower_address_reg(base) (base + PCI_MSI_ADDRESS_LO) #define msi_upper_address_reg(base) (base + PCI_MSI_ADDRESS_HI) #define msi_data_reg(base, is64bit) \ - ( (is64bit == 1) ? base+PCI_MSI_DATA_64 : base+PCI_MSI_DATA_32 ) -#define msi_mask_bits_reg(base, is64bit) \ - ( (is64bit == 1) ? base+PCI_MSI_MASK_BIT : base+PCI_MSI_MASK_BIT-4) -#define msi_disable(control) control &= ~PCI_MSI_FLAGS_ENABLE + (base + ((is64bit == 1) ? PCI_MSI_DATA_64 : PCI_MSI_DATA_32)) +#define msi_mask_reg(base, is64bit) \ + (base + ((is64bit == 1) ? PCI_MSI_MASK_64 : PCI_MSI_MASK_32)) #define is_64bit_address(control) (!!(control & PCI_MSI_FLAGS_64BIT)) #define is_mask_bit_support(control) (!!(control & PCI_MSI_FLAGS_MASKBIT)) #define msix_table_offset_reg(base) (base + 0x04) #define msix_pba_offset_reg(base) (base + 0x08) -#define msix_enable(control) control |= PCI_MSIX_FLAGS_ENABLE -#define msix_disable(control) control &= ~PCI_MSIX_FLAGS_ENABLE #define msix_table_size(control) ((control & PCI_MSIX_FLAGS_QSIZE)+1) -#define multi_msix_capable msix_table_size -#define msix_unmask(address) (address & ~PCI_MSIX_FLAGS_BITMASK) -#define msix_mask(address) (address | PCI_MSIX_FLAGS_BITMASK) -#define msix_is_pending(address) (address & PCI_MSIX_FLAGS_PENDMASK) +#define multi_msix_capable(control) msix_table_size((control)) #endif /* MSI_H */ diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index d76c4c85367..a7eb7277b10 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -508,7 +508,7 @@ static int pci_restore_standard_config(struct pci_dev *pci_dev) return error; } - return pci_dev->state_saved ? pci_restore_state(pci_dev) : 0; + return pci_restore_state(pci_dev); } static void pci_pm_default_resume_noirq(struct pci_dev *pci_dev) @@ -575,7 +575,7 @@ static void pci_pm_complete(struct device *dev) static int pci_pm_suspend(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_suspend(dev, PMSG_SUSPEND); @@ -613,7 +613,7 @@ static int pci_pm_suspend(struct device *dev) static int pci_pm_suspend_noirq(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_suspend_late(dev, PMSG_SUSPEND); @@ -672,7 +672,7 @@ static int pci_pm_resume_noirq(struct device *dev) static int pci_pm_resume(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; int error = 0; /* @@ -711,7 +711,7 @@ static int pci_pm_resume(struct device *dev) static int pci_pm_freeze(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_suspend(dev, PMSG_FREEZE); @@ -780,7 +780,7 @@ static int pci_pm_thaw_noirq(struct device *dev) static int pci_pm_thaw(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; int error = 0; if (pci_has_legacy_pm_support(pci_dev)) @@ -799,7 +799,7 @@ static int pci_pm_thaw(struct device *dev) static int pci_pm_poweroff(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_suspend(dev, PMSG_HIBERNATE); @@ -872,7 +872,7 @@ static int pci_pm_restore_noirq(struct device *dev) static int pci_pm_restore(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); - struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; + const struct dev_pm_ops *pm = dev->driver ? dev->driver->pm : NULL; int error = 0; /* @@ -910,7 +910,7 @@ static int pci_pm_restore(struct device *dev) #endif /* !CONFIG_HIBERNATION */ -struct dev_pm_ops pci_dev_pm_ops = { +const struct dev_pm_ops pci_dev_pm_ops = { .prepare = pci_pm_prepare, .complete = pci_pm_complete, .suspend = pci_pm_suspend, diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 1a91bf9687a..7b70312181d 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -24,6 +24,11 @@ #include <asm/setup.h> #include "pci.h" +const char *pci_power_names[] = { + "error", "D0", "D1", "D2", "D3hot", "D3cold", "unknown", +}; +EXPORT_SYMBOL_GPL(pci_power_names); + unsigned int pci_pm_d3_delay = PCI_PM_D3_WAIT; #ifdef CONFIG_PCI_DOMAINS @@ -480,6 +485,8 @@ static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state) pmcsr &= ~PCI_PM_CTRL_STATE_MASK; pmcsr |= state; break; + case PCI_D3hot: + case PCI_D3cold: case PCI_UNKNOWN: /* Boot-up */ if ((pmcsr & PCI_PM_CTRL_STATE_MASK) == PCI_D3hot && !(pmcsr & PCI_PM_CTRL_NO_SOFT_RESET)) @@ -839,6 +846,8 @@ pci_restore_state(struct pci_dev *dev) int i; u32 val; + if (!dev->state_saved) + return 0; /* PCI Express register must be restored first */ pci_restore_pcie_state(dev); @@ -1203,7 +1212,7 @@ void pci_pme_active(struct pci_dev *dev, bool enable) * Error code depending on the platform is returned if both the platform and * the native mechanism fail to enable the generation of wake-up events */ -int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable) +int pci_enable_wake(struct pci_dev *dev, pci_power_t state, bool enable) { int error = 0; bool pme_done = false; @@ -1282,15 +1291,14 @@ pci_power_t pci_target_state(struct pci_dev *dev) default: target_state = state; } + } else if (!dev->pm_cap) { + target_state = PCI_D0; } else if (device_may_wakeup(&dev->dev)) { /* * Find the deepest state from which the device can generate * wake-up events, make it the target state and enable device * to generate PME#. */ - if (!dev->pm_cap) - return PCI_POWER_ERROR; - if (dev->pme_support) { while (target_state && !(dev->pme_support & (1 << target_state))) @@ -1511,11 +1519,20 @@ void pci_enable_ari(struct pci_dev *dev) * * Perform INTx swizzling for a device behind one level of bridge. This is * required by section 9.1 of the PCI-to-PCI bridge specification for devices - * behind bridges on add-in cards. + * behind bridges on add-in cards. For devices with ARI enabled, the slot + * number is always 0 (see the Implementation Note in section 2.2.8.1 of + * the PCI Express Base Specification, Revision 2.1) */ u8 pci_swizzle_interrupt_pin(struct pci_dev *dev, u8 pin) { - return (((pin - 1) + PCI_SLOT(dev->devfn)) % 4) + 1; + int slot; + + if (pci_ari_enabled(dev->bus)) + slot = 0; + else + slot = PCI_SLOT(dev->devfn); + + return (((pin - 1) + slot) % 4) + 1; } int @@ -1527,7 +1544,7 @@ pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge) if (!pin) return -1; - while (dev->bus->parent) { + while (!pci_is_root_bus(dev->bus)) { pin = pci_swizzle_interrupt_pin(dev, pin); dev = dev->bus->self; } @@ -1547,7 +1564,7 @@ u8 pci_common_swizzle(struct pci_dev *dev, u8 *pinp) { u8 pin = *pinp; - while (dev->bus->parent) { + while (!pci_is_root_bus(dev->bus)) { pin = pci_swizzle_interrupt_pin(dev, pin); dev = dev->bus->self; } @@ -2053,111 +2070,177 @@ int pci_set_dma_seg_boundary(struct pci_dev *dev, unsigned long mask) EXPORT_SYMBOL(pci_set_dma_seg_boundary); #endif -static int __pcie_flr(struct pci_dev *dev, int probe) +static int pcie_flr(struct pci_dev *dev, int probe) { - u16 status; + int i; + int pos; u32 cap; - int exppos = pci_find_capability(dev, PCI_CAP_ID_EXP); + u16 status; - if (!exppos) + pos = pci_find_capability(dev, PCI_CAP_ID_EXP); + if (!pos) return -ENOTTY; - pci_read_config_dword(dev, exppos + PCI_EXP_DEVCAP, &cap); + + pci_read_config_dword(dev, pos + PCI_EXP_DEVCAP, &cap); if (!(cap & PCI_EXP_DEVCAP_FLR)) return -ENOTTY; if (probe) return 0; - pci_block_user_cfg_access(dev); - /* Wait for Transaction Pending bit clean */ - pci_read_config_word(dev, exppos + PCI_EXP_DEVSTA, &status); - if (!(status & PCI_EXP_DEVSTA_TRPND)) - goto transaction_done; + for (i = 0; i < 4; i++) { + if (i) + msleep((1 << (i - 1)) * 100); - msleep(100); - pci_read_config_word(dev, exppos + PCI_EXP_DEVSTA, &status); - if (!(status & PCI_EXP_DEVSTA_TRPND)) - goto transaction_done; - - dev_info(&dev->dev, "Busy after 100ms while trying to reset; " - "sleeping for 1 second\n"); - ssleep(1); - pci_read_config_word(dev, exppos + PCI_EXP_DEVSTA, &status); - if (status & PCI_EXP_DEVSTA_TRPND) - dev_info(&dev->dev, "Still busy after 1s; " - "proceeding with reset anyway\n"); - -transaction_done: - pci_write_config_word(dev, exppos + PCI_EXP_DEVCTL, + pci_read_config_word(dev, pos + PCI_EXP_DEVSTA, &status); + if (!(status & PCI_EXP_DEVSTA_TRPND)) + goto clear; + } + + dev_err(&dev->dev, "transaction is not cleared; " + "proceeding with reset anyway\n"); + +clear: + pci_write_config_word(dev, pos + PCI_EXP_DEVCTL, PCI_EXP_DEVCTL_BCR_FLR); - mdelay(100); + msleep(100); - pci_unblock_user_cfg_access(dev); return 0; } -static int __pci_af_flr(struct pci_dev *dev, int probe) +static int pci_af_flr(struct pci_dev *dev, int probe) { - int cappos = pci_find_capability(dev, PCI_CAP_ID_AF); - u8 status; + int i; + int pos; u8 cap; + u8 status; - if (!cappos) + pos = pci_find_capability(dev, PCI_CAP_ID_AF); + if (!pos) return -ENOTTY; - pci_read_config_byte(dev, cappos + PCI_AF_CAP, &cap); + + pci_read_config_byte(dev, pos + PCI_AF_CAP, &cap); if (!(cap & PCI_AF_CAP_TP) || !(cap & PCI_AF_CAP_FLR)) return -ENOTTY; if (probe) return 0; - pci_block_user_cfg_access(dev); - /* Wait for Transaction Pending bit clean */ - pci_read_config_byte(dev, cappos + PCI_AF_STATUS, &status); - if (!(status & PCI_AF_STATUS_TP)) - goto transaction_done; + for (i = 0; i < 4; i++) { + if (i) + msleep((1 << (i - 1)) * 100); + + pci_read_config_byte(dev, pos + PCI_AF_STATUS, &status); + if (!(status & PCI_AF_STATUS_TP)) + goto clear; + } + dev_err(&dev->dev, "transaction is not cleared; " + "proceeding with reset anyway\n"); + +clear: + pci_write_config_byte(dev, pos + PCI_AF_CTRL, PCI_AF_CTRL_FLR); msleep(100); - pci_read_config_byte(dev, cappos + PCI_AF_STATUS, &status); - if (!(status & PCI_AF_STATUS_TP)) - goto transaction_done; - - dev_info(&dev->dev, "Busy after 100ms while trying to" - " reset; sleeping for 1 second\n"); - ssleep(1); - pci_read_config_byte(dev, cappos + PCI_AF_STATUS, &status); - if (status & PCI_AF_STATUS_TP) - dev_info(&dev->dev, "Still busy after 1s; " - "proceeding with reset anyway\n"); - -transaction_done: - pci_write_config_byte(dev, cappos + PCI_AF_CTRL, PCI_AF_CTRL_FLR); - mdelay(100); - - pci_unblock_user_cfg_access(dev); + return 0; } -static int __pci_reset_function(struct pci_dev *pdev, int probe) +static int pci_pm_reset(struct pci_dev *dev, int probe) { - int res; + u16 csr; + + if (!dev->pm_cap) + return -ENOTTY; - res = __pcie_flr(pdev, probe); - if (res != -ENOTTY) - return res; + pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &csr); + if (csr & PCI_PM_CTRL_NO_SOFT_RESET) + return -ENOTTY; - res = __pci_af_flr(pdev, probe); - if (res != -ENOTTY) - return res; + if (probe) + return 0; + + if (dev->current_state != PCI_D0) + return -EINVAL; - return res; + csr &= ~PCI_PM_CTRL_STATE_MASK; + csr |= PCI_D3hot; + pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, csr); + msleep(pci_pm_d3_delay); + + csr &= ~PCI_PM_CTRL_STATE_MASK; + csr |= PCI_D0; + pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, csr); + msleep(pci_pm_d3_delay); + + return 0; +} + +static int pci_parent_bus_reset(struct pci_dev *dev, int probe) +{ + u16 ctrl; + struct pci_dev *pdev; + + if (pci_is_root_bus(dev->bus) || dev->subordinate || !dev->bus->self) + return -ENOTTY; + + list_for_each_entry(pdev, &dev->bus->devices, bus_list) + if (pdev != dev) + return -ENOTTY; + + if (probe) + return 0; + + pci_read_config_word(dev->bus->self, PCI_BRIDGE_CONTROL, &ctrl); + ctrl |= PCI_BRIDGE_CTL_BUS_RESET; + pci_write_config_word(dev->bus->self, PCI_BRIDGE_CONTROL, ctrl); + msleep(100); + + ctrl &= ~PCI_BRIDGE_CTL_BUS_RESET; + pci_write_config_word(dev->bus->self, PCI_BRIDGE_CONTROL, ctrl); + msleep(100); + + return 0; +} + +static int pci_dev_reset(struct pci_dev *dev, int probe) +{ + int rc; + + might_sleep(); + + if (!probe) { + pci_block_user_cfg_access(dev); + /* block PM suspend, driver probe, etc. */ + down(&dev->dev.sem); + } + + rc = pcie_flr(dev, probe); + if (rc != -ENOTTY) + goto done; + + rc = pci_af_flr(dev, probe); + if (rc != -ENOTTY) + goto done; + + rc = pci_pm_reset(dev, probe); + if (rc != -ENOTTY) + goto done; + + rc = pci_parent_bus_reset(dev, probe); +done: + if (!probe) { + up(&dev->dev.sem); + pci_unblock_user_cfg_access(dev); + } + + return rc; } /** - * pci_execute_reset_function() - Reset a PCI device function - * @dev: Device function to reset + * __pci_reset_function - reset a PCI device function + * @dev: PCI device to reset * * Some devices allow an individual function to be reset without affecting * other functions in the same device. The PCI device must be responsive @@ -2169,18 +2252,18 @@ static int __pci_reset_function(struct pci_dev *pdev, int probe) * device including MSI, bus mastering, BARs, decoding IO and memory spaces, * etc. * - * Returns 0 if the device function was successfully reset or -ENOTTY if the + * Returns 0 if the device function was successfully reset or negative if the * device doesn't support resetting a single function. */ -int pci_execute_reset_function(struct pci_dev *dev) +int __pci_reset_function(struct pci_dev *dev) { - return __pci_reset_function(dev, 0); + return pci_dev_reset(dev, 0); } -EXPORT_SYMBOL_GPL(pci_execute_reset_function); +EXPORT_SYMBOL_GPL(__pci_reset_function); /** - * pci_reset_function() - quiesce and reset a PCI device function - * @dev: Device function to reset + * pci_reset_function - quiesce and reset a PCI device function + * @dev: PCI device to reset * * Some devices allow an individual function to be reset without affecting * other functions in the same device. The PCI device must be responsive @@ -2188,32 +2271,33 @@ EXPORT_SYMBOL_GPL(pci_execute_reset_function); * * This function does not just reset the PCI portion of a device, but * clears all the state associated with the device. This function differs - * from pci_execute_reset_function in that it saves and restores device state + * from __pci_reset_function in that it saves and restores device state * over the reset. * - * Returns 0 if the device function was successfully reset or -ENOTTY if the + * Returns 0 if the device function was successfully reset or negative if the * device doesn't support resetting a single function. */ int pci_reset_function(struct pci_dev *dev) { - int r = __pci_reset_function(dev, 1); + int rc; - if (r < 0) - return r; + rc = pci_dev_reset(dev, 1); + if (rc) + return rc; - if (!dev->msi_enabled && !dev->msix_enabled && dev->irq != 0) - disable_irq(dev->irq); pci_save_state(dev); + /* + * both INTx and MSI are disabled after the Interrupt Disable bit + * is set and the Bus Master bit is cleared. + */ pci_write_config_word(dev, PCI_COMMAND, PCI_COMMAND_INTX_DISABLE); - r = pci_execute_reset_function(dev); + rc = pci_dev_reset(dev, 0); pci_restore_state(dev); - if (!dev->msi_enabled && !dev->msix_enabled && dev->irq != 0) - enable_irq(dev->irq); - return r; + return rc; } EXPORT_SYMBOL_GPL(pci_reset_function); @@ -2586,6 +2670,8 @@ static int __init pci_setup(char *str) } else if (!strncmp(str, "resource_alignment=", 19)) { pci_set_resource_alignment_param(str + 19, strlen(str + 19)); + } else if (!strncmp(str, "ecrc=", 5)) { + pcie_ecrc_get_policy(str + 5); } else { printk(KERN_ERR "PCI: Unknown option `%s'\n", str); diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index d03f6b99f29..f73bcbedf37 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -229,6 +229,15 @@ struct pci_sriov { u8 __iomem *mstate; /* VF Migration State Array */ }; +/* Address Translation Service */ +struct pci_ats { + int pos; /* capability position */ + int stu; /* Smallest Translation Unit */ + int qdep; /* Invalidate Queue Depth */ + int ref_cnt; /* Physical Function reference count */ + int is_enabled:1; /* Enable bit is set */ +}; + #ifdef CONFIG_PCI_IOV extern int pci_iov_init(struct pci_dev *dev); extern void pci_iov_release(struct pci_dev *dev); @@ -236,6 +245,20 @@ extern int pci_iov_resource_bar(struct pci_dev *dev, int resno, enum pci_bar_type *type); extern void pci_restore_iov_state(struct pci_dev *dev); extern int pci_iov_bus_range(struct pci_bus *bus); + +extern int pci_enable_ats(struct pci_dev *dev, int ps); +extern void pci_disable_ats(struct pci_dev *dev); +extern int pci_ats_queue_depth(struct pci_dev *dev); +/** + * pci_ats_enabled - query the ATS status + * @dev: the PCI device + * + * Returns 1 if ATS capability is enabled, or 0 if not. + */ +static inline int pci_ats_enabled(struct pci_dev *dev) +{ + return dev->ats && dev->ats->is_enabled; +} #else static inline int pci_iov_init(struct pci_dev *dev) { @@ -257,6 +280,22 @@ static inline int pci_iov_bus_range(struct pci_bus *bus) { return 0; } + +static inline int pci_enable_ats(struct pci_dev *dev, int ps) +{ + return -ENODEV; +} +static inline void pci_disable_ats(struct pci_dev *dev) +{ +} +static inline int pci_ats_queue_depth(struct pci_dev *dev) +{ + return -ENODEV; +} +static inline int pci_ats_enabled(struct pci_dev *dev) +{ + return 0; +} #endif /* CONFIG_PCI_IOV */ #endif /* DRIVERS_PCI_H */ diff --git a/drivers/pci/pcie/aer/Kconfig b/drivers/pci/pcie/aer/Kconfig index c3bde588aa1..50e94e02378 100644 --- a/drivers/pci/pcie/aer/Kconfig +++ b/drivers/pci/pcie/aer/Kconfig @@ -10,3 +10,18 @@ config PCIEAER This enables PCI Express Root Port Advanced Error Reporting (AER) driver support. Error reporting messages sent to Root Port will be handled by PCI Express AER driver. + + +# +# PCI Express ECRC +# +config PCIE_ECRC + bool "PCI Express ECRC settings control" + depends on PCIEAER + help + Used to override firmware/bios settings for PCI Express ECRC + (transaction layer end-to-end CRC checking). + + When in doubt, say N. + +source "drivers/pci/pcie/aer/Kconfig.debug" diff --git a/drivers/pci/pcie/aer/Kconfig.debug b/drivers/pci/pcie/aer/Kconfig.debug new file mode 100644 index 00000000000..b8c925c1f6a --- /dev/null +++ b/drivers/pci/pcie/aer/Kconfig.debug @@ -0,0 +1,18 @@ +# +# PCI Express Root Port Device AER Debug Configuration +# + +config PCIEAER_INJECT + tristate "PCIE AER error injector support" + depends on PCIEAER + default n + help + This enables PCI Express Root Port Advanced Error Reporting + (AER) software error injector. + + Debuging PCIE AER code is quite difficult because it is hard + to trigger various real hardware errors. Software based + error injection can fake almost all kinds of errors with the + help of a user space helper tool aer-inject, which can be + gotten from: + http://www.kernel.org/pub/linux/utils/pci/aer-inject/ diff --git a/drivers/pci/pcie/aer/Makefile b/drivers/pci/pcie/aer/Makefile index 8da3bd8455a..2cba67510dc 100644 --- a/drivers/pci/pcie/aer/Makefile +++ b/drivers/pci/pcie/aer/Makefile @@ -4,6 +4,9 @@ obj-$(CONFIG_PCIEAER) += aerdriver.o +obj-$(CONFIG_PCIE_ECRC) += ecrc.o + aerdriver-objs := aerdrv_errprint.o aerdrv_core.o aerdrv.o aerdriver-$(CONFIG_ACPI) += aerdrv_acpi.o +obj-$(CONFIG_PCIEAER_INJECT) += aer_inject.o diff --git a/drivers/pci/pcie/aer/aer_inject.c b/drivers/pci/pcie/aer/aer_inject.c new file mode 100644 index 00000000000..d92ae21a59d --- /dev/null +++ b/drivers/pci/pcie/aer/aer_inject.c @@ -0,0 +1,473 @@ +/* + * PCIE AER software error injection support. + * + * Debuging PCIE AER code is quite difficult because it is hard to + * trigger various real hardware errors. Software based error + * injection can fake almost all kinds of errors with the help of a + * user space helper tool aer-inject, which can be gotten from: + * http://www.kernel.org/pub/linux/utils/pci/aer-inject/ + * + * Copyright 2009 Intel Corporation. + * Huang Ying <ying.huang@intel.com> + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; version 2 + * of the License. + * + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/miscdevice.h> +#include <linux/pci.h> +#include <linux/fs.h> +#include <asm/uaccess.h> +#include "aerdrv.h" + +struct aer_error_inj +{ + u8 bus; + u8 dev; + u8 fn; + u32 uncor_status; + u32 cor_status; + u32 header_log0; + u32 header_log1; + u32 header_log2; + u32 header_log3; +}; + +struct aer_error +{ + struct list_head list; + unsigned int bus; + unsigned int devfn; + int pos_cap_err; + + u32 uncor_status; + u32 cor_status; + u32 header_log0; + u32 header_log1; + u32 header_log2; + u32 header_log3; + u32 root_status; + u32 source_id; +}; + +struct pci_bus_ops +{ + struct list_head list; + struct pci_bus *bus; + struct pci_ops *ops; +}; + +static LIST_HEAD(einjected); + +static LIST_HEAD(pci_bus_ops_list); + +/* Protect einjected and pci_bus_ops_list */ +static DEFINE_SPINLOCK(inject_lock); + +static void aer_error_init(struct aer_error *err, unsigned int bus, + unsigned int devfn, int pos_cap_err) +{ + INIT_LIST_HEAD(&err->list); + err->bus = bus; + err->devfn = devfn; + err->pos_cap_err = pos_cap_err; +} + +/* inject_lock must be held before calling */ +static struct aer_error *__find_aer_error(unsigned int bus, unsigned int devfn) +{ + struct aer_error *err; + + list_for_each_entry(err, &einjected, list) { + if (bus == err->bus && devfn == err->devfn) + return err; + } + return NULL; +} + +/* inject_lock must be held before calling */ +static struct aer_error *__find_aer_error_by_dev(struct pci_dev *dev) +{ + return __find_aer_error(dev->bus->number, dev->devfn); +} + +/* inject_lock must be held before calling */ +static struct pci_ops *__find_pci_bus_ops(struct pci_bus *bus) +{ + struct pci_bus_ops *bus_ops; + + list_for_each_entry(bus_ops, &pci_bus_ops_list, list) { + if (bus_ops->bus == bus) + return bus_ops->ops; + } + return NULL; +} + +static struct pci_bus_ops *pci_bus_ops_pop(void) +{ + unsigned long flags; + struct pci_bus_ops *bus_ops = NULL; + + spin_lock_irqsave(&inject_lock, flags); + if (list_empty(&pci_bus_ops_list)) + bus_ops = NULL; + else { + struct list_head *lh = pci_bus_ops_list.next; + list_del(lh); + bus_ops = list_entry(lh, struct pci_bus_ops, list); + } + spin_unlock_irqrestore(&inject_lock, flags); + return bus_ops; +} + +static u32 *find_pci_config_dword(struct aer_error *err, int where, + int *prw1cs) +{ + int rw1cs = 0; + u32 *target = NULL; + + if (err->pos_cap_err == -1) + return NULL; + + switch (where - err->pos_cap_err) { + case PCI_ERR_UNCOR_STATUS: + target = &err->uncor_status; + rw1cs = 1; + break; + case PCI_ERR_COR_STATUS: + target = &err->cor_status; + rw1cs = 1; + break; + case PCI_ERR_HEADER_LOG: + target = &err->header_log0; + break; + case PCI_ERR_HEADER_LOG+4: + target = &err->header_log1; + break; + case PCI_ERR_HEADER_LOG+8: + target = &err->header_log2; + break; + case PCI_ERR_HEADER_LOG+12: + target = &err->header_log3; + break; + case PCI_ERR_ROOT_STATUS: + target = &err->root_status; + rw1cs = 1; + break; + case PCI_ERR_ROOT_COR_SRC: + target = &err->source_id; + break; + } + if (prw1cs) + *prw1cs = rw1cs; + return target; +} + +static int pci_read_aer(struct pci_bus *bus, unsigned int devfn, int where, + int size, u32 *val) +{ + u32 *sim; + struct aer_error *err; + unsigned long flags; + struct pci_ops *ops; + + spin_lock_irqsave(&inject_lock, flags); + if (size != sizeof(u32)) + goto out; + err = __find_aer_error(bus->number, devfn); + if (!err) + goto out; + + sim = find_pci_config_dword(err, where, NULL); + if (sim) { + *val = *sim; + spin_unlock_irqrestore(&inject_lock, flags); + return 0; + } +out: + ops = __find_pci_bus_ops(bus); + spin_unlock_irqrestore(&inject_lock, flags); + return ops->read(bus, devfn, where, size, val); +} + +int pci_write_aer(struct pci_bus *bus, unsigned int devfn, int where, int size, + u32 val) +{ + u32 *sim; + struct aer_error *err; + unsigned long flags; + int rw1cs; + struct pci_ops *ops; + + spin_lock_irqsave(&inject_lock, flags); + if (size != sizeof(u32)) + goto out; + err = __find_aer_error(bus->number, devfn); + if (!err) + goto out; + + sim = find_pci_config_dword(err, where, &rw1cs); + if (sim) { + if (rw1cs) + *sim ^= val; + else + *sim = val; + spin_unlock_irqrestore(&inject_lock, flags); + return 0; + } +out: + ops = __find_pci_bus_ops(bus); + spin_unlock_irqrestore(&inject_lock, flags); + return ops->write(bus, devfn, where, size, val); +} + +static struct pci_ops pci_ops_aer = { + .read = pci_read_aer, + .write = pci_write_aer, +}; + +static void pci_bus_ops_init(struct pci_bus_ops *bus_ops, + struct pci_bus *bus, + struct pci_ops *ops) +{ + INIT_LIST_HEAD(&bus_ops->list); + bus_ops->bus = bus; + bus_ops->ops = ops; +} + +static int pci_bus_set_aer_ops(struct pci_bus *bus) +{ + struct pci_ops *ops; + struct pci_bus_ops *bus_ops; + unsigned long flags; + + bus_ops = kmalloc(sizeof(*bus_ops), GFP_KERNEL); + if (!bus_ops) + return -ENOMEM; + ops = pci_bus_set_ops(bus, &pci_ops_aer); + spin_lock_irqsave(&inject_lock, flags); + if (ops == &pci_ops_aer) + goto out; + pci_bus_ops_init(bus_ops, bus, ops); + list_add(&bus_ops->list, &pci_bus_ops_list); + bus_ops = NULL; +out: + spin_unlock_irqrestore(&inject_lock, flags); + if (bus_ops) + kfree(bus_ops); + return 0; +} + +static struct pci_dev *pcie_find_root_port(struct pci_dev *dev) +{ + while (1) { + if (!dev->is_pcie) + break; + if (dev->pcie_type == PCI_EXP_TYPE_ROOT_PORT) + return dev; + if (!dev->bus->self) + break; + dev = dev->bus->self; + } + return NULL; +} + +static int find_aer_device_iter(struct device *device, void *data) +{ + struct pcie_device **result = data; + struct pcie_device *pcie_dev; + + if (device->bus == &pcie_port_bus_type) { + pcie_dev = to_pcie_device(device); + if (pcie_dev->service & PCIE_PORT_SERVICE_AER) { + *result = pcie_dev; + return 1; + } + } + return 0; +} + +static int find_aer_device(struct pci_dev *dev, struct pcie_device **result) +{ + return device_for_each_child(&dev->dev, result, find_aer_device_iter); +} + +static int aer_inject(struct aer_error_inj *einj) +{ + struct aer_error *err, *rperr; + struct aer_error *err_alloc = NULL, *rperr_alloc = NULL; + struct pci_dev *dev, *rpdev; + struct pcie_device *edev; + unsigned long flags; + unsigned int devfn = PCI_DEVFN(einj->dev, einj->fn); + int pos_cap_err, rp_pos_cap_err; + u32 sever; + int ret = 0; + + dev = pci_get_bus_and_slot(einj->bus, devfn); + if (!dev) + return -EINVAL; + rpdev = pcie_find_root_port(dev); + if (!rpdev) { + ret = -EINVAL; + goto out_put; + } + + pos_cap_err = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR); + if (!pos_cap_err) { + ret = -EIO; + goto out_put; + } + pci_read_config_dword(dev, pos_cap_err + PCI_ERR_UNCOR_SEVER, &sever); + + rp_pos_cap_err = pci_find_ext_capability(rpdev, PCI_EXT_CAP_ID_ERR); + if (!rp_pos_cap_err) { + ret = -EIO; + goto out_put; + } + + err_alloc = kzalloc(sizeof(struct aer_error), GFP_KERNEL); + if (!err_alloc) { + ret = -ENOMEM; + goto out_put; + } + rperr_alloc = kzalloc(sizeof(struct aer_error), GFP_KERNEL); + if (!rperr_alloc) { + ret = -ENOMEM; + goto out_put; + } + + spin_lock_irqsave(&inject_lock, flags); + + err = __find_aer_error_by_dev(dev); + if (!err) { + err = err_alloc; + err_alloc = NULL; + aer_error_init(err, einj->bus, devfn, pos_cap_err); + list_add(&err->list, &einjected); + } + err->uncor_status |= einj->uncor_status; + err->cor_status |= einj->cor_status; + err->header_log0 = einj->header_log0; + err->header_log1 = einj->header_log1; + err->header_log2 = einj->header_log2; + err->header_log3 = einj->header_log3; + + rperr = __find_aer_error_by_dev(rpdev); + if (!rperr) { + rperr = rperr_alloc; + rperr_alloc = NULL; + aer_error_init(rperr, rpdev->bus->number, rpdev->devfn, + rp_pos_cap_err); + list_add(&rperr->list, &einjected); + } + if (einj->cor_status) { + if (rperr->root_status & PCI_ERR_ROOT_COR_RCV) + rperr->root_status |= PCI_ERR_ROOT_MULTI_COR_RCV; + else + rperr->root_status |= PCI_ERR_ROOT_COR_RCV; + rperr->source_id &= 0xffff0000; + rperr->source_id |= (einj->bus << 8) | devfn; + } + if (einj->uncor_status) { + if (rperr->root_status & PCI_ERR_ROOT_UNCOR_RCV) + rperr->root_status |= PCI_ERR_ROOT_MULTI_UNCOR_RCV; + if (sever & einj->uncor_status) { + rperr->root_status |= PCI_ERR_ROOT_FATAL_RCV; + if (!(rperr->root_status & PCI_ERR_ROOT_UNCOR_RCV)) + rperr->root_status |= PCI_ERR_ROOT_FIRST_FATAL; + } else + rperr->root_status |= PCI_ERR_ROOT_NONFATAL_RCV; + rperr->root_status |= PCI_ERR_ROOT_UNCOR_RCV; + rperr->source_id &= 0x0000ffff; + rperr->source_id |= ((einj->bus << 8) | devfn) << 16; + } + spin_unlock_irqrestore(&inject_lock, flags); + + ret = pci_bus_set_aer_ops(dev->bus); + if (ret) + goto out_put; + ret = pci_bus_set_aer_ops(rpdev->bus); + if (ret) + goto out_put; + + if (find_aer_device(rpdev, &edev)) + aer_irq(-1, edev); + else + ret = -EINVAL; +out_put: + if (err_alloc) + kfree(err_alloc); + if (rperr_alloc) + kfree(rperr_alloc); + pci_dev_put(dev); + return ret; +} + +static ssize_t aer_inject_write(struct file *filp, const char __user *ubuf, + size_t usize, loff_t *off) +{ + struct aer_error_inj einj; + int ret; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (usize != sizeof(struct aer_error_inj)) + return -EINVAL; + + if (copy_from_user(&einj, ubuf, usize)) + return -EFAULT; + + ret = aer_inject(&einj); + return ret ? ret : usize; +} + +static const struct file_operations aer_inject_fops = { + .write = aer_inject_write, + .owner = THIS_MODULE, +}; + +static struct miscdevice aer_inject_device = { + .minor = MISC_DYNAMIC_MINOR, + .name = "aer_inject", + .fops = &aer_inject_fops, +}; + +static int __init aer_inject_init(void) +{ + return misc_register(&aer_inject_device); +} + +static void __exit aer_inject_exit(void) +{ + struct aer_error *err, *err_next; + unsigned long flags; + struct pci_bus_ops *bus_ops; + + misc_deregister(&aer_inject_device); + + while ((bus_ops = pci_bus_ops_pop())) { + pci_bus_set_ops(bus_ops->bus, bus_ops->ops); + kfree(bus_ops); + } + + spin_lock_irqsave(&inject_lock, flags); + list_for_each_entry_safe(err, err_next, + &pci_bus_ops_list, list) { + list_del(&err->list); + kfree(err); + } + spin_unlock_irqrestore(&inject_lock, flags); +} + +module_init(aer_inject_init); +module_exit(aer_inject_exit); + +MODULE_DESCRIPTION("PCIE AER software error injector"); +MODULE_LICENSE("GPL"); diff --git a/drivers/pci/pcie/aer/aerdrv.c b/drivers/pci/pcie/aer/aerdrv.c index 32ade5af927..4770f13b3ca 100644 --- a/drivers/pci/pcie/aer/aerdrv.c +++ b/drivers/pci/pcie/aer/aerdrv.c @@ -77,7 +77,7 @@ void pci_no_aer(void) * * Invoked when Root Port detects AER messages. **/ -static irqreturn_t aer_irq(int irq, void *context) +irqreturn_t aer_irq(int irq, void *context) { unsigned int status, id; struct pcie_device *pdev = (struct pcie_device *)context; @@ -126,6 +126,7 @@ static irqreturn_t aer_irq(int irq, void *context) return IRQ_HANDLED; } +EXPORT_SYMBOL_GPL(aer_irq); /** * aer_alloc_rpc - allocate Root Port data structure diff --git a/drivers/pci/pcie/aer/aerdrv.h b/drivers/pci/pcie/aer/aerdrv.h index aa14482a477..bbd7428ca2d 100644 --- a/drivers/pci/pcie/aer/aerdrv.h +++ b/drivers/pci/pcie/aer/aerdrv.h @@ -11,6 +11,7 @@ #include <linux/workqueue.h> #include <linux/pcieport_if.h> #include <linux/aer.h> +#include <linux/interrupt.h> #define AER_NONFATAL 0 #define AER_FATAL 1 @@ -56,7 +57,11 @@ struct header_log_regs { unsigned int dw3; }; +#define AER_MAX_MULTI_ERR_DEVICES 5 /* Not likely to have more */ struct aer_err_info { + struct pci_dev *dev[AER_MAX_MULTI_ERR_DEVICES]; + int error_dev_num; + u16 id; int severity; /* 0:NONFATAL | 1:FATAL | 2:COR */ int flags; unsigned int status; /* COR/UNCOR Error Status */ @@ -120,6 +125,7 @@ extern void aer_delete_rootport(struct aer_rpc *rpc); extern int aer_init(struct pcie_device *dev); extern void aer_isr(struct work_struct *work); extern void aer_print_error(struct pci_dev *dev, struct aer_err_info *info); +extern irqreturn_t aer_irq(int irq, void *context); #ifdef CONFIG_ACPI extern int aer_osc_setup(struct pcie_device *pciedev); diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c index 307452f3003..3d8872704a5 100644 --- a/drivers/pci/pcie/aer/aerdrv_core.c +++ b/drivers/pci/pcie/aer/aerdrv_core.c @@ -26,7 +26,9 @@ #include "aerdrv.h" static int forceload; +static int nosourceid; module_param(forceload, bool, 0); +module_param(nosourceid, bool, 0); int pci_enable_pcie_error_reporting(struct pci_dev *dev) { @@ -109,19 +111,23 @@ int pci_cleanup_aer_correct_error_status(struct pci_dev *dev) #endif /* 0 */ -static void set_device_error_reporting(struct pci_dev *dev, void *data) +static int set_device_error_reporting(struct pci_dev *dev, void *data) { bool enable = *((bool *)data); - if (dev->pcie_type != PCIE_RC_PORT && - dev->pcie_type != PCIE_SW_UPSTREAM_PORT && - dev->pcie_type != PCIE_SW_DOWNSTREAM_PORT) - return; + if (dev->pcie_type == PCIE_RC_PORT || + dev->pcie_type == PCIE_SW_UPSTREAM_PORT || + dev->pcie_type == PCIE_SW_DOWNSTREAM_PORT) { + if (enable) + pci_enable_pcie_error_reporting(dev); + else + pci_disable_pcie_error_reporting(dev); + } if (enable) - pci_enable_pcie_error_reporting(dev); - else - pci_disable_pcie_error_reporting(dev); + pcie_set_ecrc_checking(dev); + + return 0; } /** @@ -139,73 +145,148 @@ static void set_downstream_devices_error_reporting(struct pci_dev *dev, pci_walk_bus(dev->subordinate, set_device_error_reporting, &enable); } -static int find_device_iter(struct device *device, void *data) +static inline int compare_device_id(struct pci_dev *dev, + struct aer_err_info *e_info) { - struct pci_dev *dev; - u16 id = *(unsigned long *)data; - u8 secondary, subordinate, d_bus = id >> 8; + if (e_info->id == ((dev->bus->number << 8) | dev->devfn)) { + /* + * Device ID match + */ + return 1; + } - if (device->bus == &pci_bus_type) { - dev = to_pci_dev(device); - if (id == ((dev->bus->number << 8) | dev->devfn)) { - /* - * Device ID match - */ - *(unsigned long*)data = (unsigned long)device; - return 1; - } + return 0; +} + +static int add_error_device(struct aer_err_info *e_info, struct pci_dev *dev) +{ + if (e_info->error_dev_num < AER_MAX_MULTI_ERR_DEVICES) { + e_info->dev[e_info->error_dev_num] = dev; + e_info->error_dev_num++; + return 1; + } else + return 0; +} + + +#define PCI_BUS(x) (((x) >> 8) & 0xff) + +static int find_device_iter(struct pci_dev *dev, void *data) +{ + int pos; + u32 status; + u32 mask; + u16 reg16; + int result; + struct aer_err_info *e_info = (struct aer_err_info *)data; + + /* + * When bus id is equal to 0, it might be a bad id + * reported by root port. + */ + if (!nosourceid && (PCI_BUS(e_info->id) != 0)) { + result = compare_device_id(dev, e_info); + if (result) + add_error_device(e_info, dev); /* - * If device is P2P, check if it is an upstream? + * If there is no multiple error, we stop + * or continue based on the id comparing. */ - if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE) { - pci_read_config_byte(dev, PCI_SECONDARY_BUS, - &secondary); - pci_read_config_byte(dev, PCI_SUBORDINATE_BUS, - &subordinate); - if (d_bus >= secondary && d_bus <= subordinate) { - *(unsigned long*)data = (unsigned long)device; - return 1; - } + if (!(e_info->flags & AER_MULTI_ERROR_VALID_FLAG)) + return result; + + /* + * If there are multiple errors and id does match, + * We need continue to search other devices under + * the root port. Return 0 means that. + */ + if (result) + return 0; + } + + /* + * When either + * 1) nosourceid==y; + * 2) bus id is equal to 0. Some ports might lose the bus + * id of error source id; + * 3) There are multiple errors and prior id comparing fails; + * We check AER status registers to find the initial reporter. + */ + if (atomic_read(&dev->enable_cnt) == 0) + return 0; + pos = pci_find_capability(dev, PCI_CAP_ID_EXP); + if (!pos) + return 0; + /* Check if AER is enabled */ + pci_read_config_word(dev, pos+PCI_EXP_DEVCTL, ®16); + if (!(reg16 & ( + PCI_EXP_DEVCTL_CERE | + PCI_EXP_DEVCTL_NFERE | + PCI_EXP_DEVCTL_FERE | + PCI_EXP_DEVCTL_URRE))) + return 0; + pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR); + if (!pos) + return 0; + + status = 0; + mask = 0; + if (e_info->severity == AER_CORRECTABLE) { + pci_read_config_dword(dev, + pos + PCI_ERR_COR_STATUS, + &status); + pci_read_config_dword(dev, + pos + PCI_ERR_COR_MASK, + &mask); + if (status & ERR_CORRECTABLE_ERROR_MASK & ~mask) { + add_error_device(e_info, dev); + goto added; + } + } else { + pci_read_config_dword(dev, + pos + PCI_ERR_UNCOR_STATUS, + &status); + pci_read_config_dword(dev, + pos + PCI_ERR_UNCOR_MASK, + &mask); + if (status & ERR_UNCORRECTABLE_ERROR_MASK & ~mask) { + add_error_device(e_info, dev); + goto added; } } return 0; + +added: + if (e_info->flags & AER_MULTI_ERROR_VALID_FLAG) + return 0; + else + return 1; } /** * find_source_device - search through device hierarchy for source device * @parent: pointer to Root Port pci_dev data structure - * @id: device ID of agent who sends an error message to this Root Port + * @err_info: including detailed error information such like id * * Invoked when error is detected at the Root Port. */ -static struct device* find_source_device(struct pci_dev *parent, u16 id) +static void find_source_device(struct pci_dev *parent, + struct aer_err_info *e_info) { struct pci_dev *dev = parent; - struct device *device; - unsigned long device_addr; - int status; + int result; /* Is Root Port an agent that sends error message? */ - if (id == ((dev->bus->number << 8) | dev->devfn)) - return &dev->dev; - - do { - device_addr = id; - if ((status = device_for_each_child(&dev->dev, - &device_addr, find_device_iter))) { - device = (struct device*)device_addr; - dev = to_pci_dev(device); - if (id == ((dev->bus->number << 8) | dev->devfn)) - return device; - } - }while (status); + result = find_device_iter(dev, e_info); + if (result) + return; - return NULL; + pci_walk_bus(parent->subordinate, find_device_iter, e_info); } -static void report_error_detected(struct pci_dev *dev, void *data) +static int report_error_detected(struct pci_dev *dev, void *data) { pci_ers_result_t vote; struct pci_error_handlers *err_handler; @@ -230,16 +311,16 @@ static void report_error_detected(struct pci_dev *dev, void *data) dev->driver ? "no AER-aware driver" : "no driver"); } - return; + return 0; } err_handler = dev->driver->err_handler; vote = err_handler->error_detected(dev, result_data->state); result_data->result = merge_result(result_data->result, vote); - return; + return 0; } -static void report_mmio_enabled(struct pci_dev *dev, void *data) +static int report_mmio_enabled(struct pci_dev *dev, void *data) { pci_ers_result_t vote; struct pci_error_handlers *err_handler; @@ -249,15 +330,15 @@ static void report_mmio_enabled(struct pci_dev *dev, void *data) if (!dev->driver || !dev->driver->err_handler || !dev->driver->err_handler->mmio_enabled) - return; + return 0; err_handler = dev->driver->err_handler; vote = err_handler->mmio_enabled(dev); result_data->result = merge_result(result_data->result, vote); - return; + return 0; } -static void report_slot_reset(struct pci_dev *dev, void *data) +static int report_slot_reset(struct pci_dev *dev, void *data) { pci_ers_result_t vote; struct pci_error_handlers *err_handler; @@ -267,15 +348,15 @@ static void report_slot_reset(struct pci_dev *dev, void *data) if (!dev->driver || !dev->driver->err_handler || !dev->driver->err_handler->slot_reset) - return; + return 0; err_handler = dev->driver->err_handler; vote = err_handler->slot_reset(dev); result_data->result = merge_result(result_data->result, vote); - return; + return 0; } -static void report_resume(struct pci_dev *dev, void *data) +static int report_resume(struct pci_dev *dev, void *data) { struct pci_error_handlers *err_handler; @@ -284,11 +365,11 @@ static void report_resume(struct pci_dev *dev, void *data) if (!dev->driver || !dev->driver->err_handler || !dev->driver->err_handler->resume) - return; + return 0; err_handler = dev->driver->err_handler; err_handler->resume(dev); - return; + return 0; } /** @@ -305,7 +386,7 @@ static void report_resume(struct pci_dev *dev, void *data) static pci_ers_result_t broadcast_error_message(struct pci_dev *dev, enum pci_channel_state state, char *error_mesg, - void (*cb)(struct pci_dev *, void *)) + int (*cb)(struct pci_dev *, void *)) { struct aer_broadcast_data result_data; @@ -497,12 +578,12 @@ static pci_ers_result_t do_recovery(struct pcie_device *aerdev, */ static void handle_error_source(struct pcie_device * aerdev, struct pci_dev *dev, - struct aer_err_info info) + struct aer_err_info *info) { pci_ers_result_t status = 0; int pos; - if (info.severity == AER_CORRECTABLE) { + if (info->severity == AER_CORRECTABLE) { /* * Correctable error does not need software intevention. * No need to go through error recovery process. @@ -510,9 +591,9 @@ static void handle_error_source(struct pcie_device * aerdev, pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR); if (pos) pci_write_config_dword(dev, pos + PCI_ERR_COR_STATUS, - info.status); + info->status); } else { - status = do_recovery(aerdev, dev, info.severity); + status = do_recovery(aerdev, dev, info->severity); if (status == PCI_ERS_RESULT_RECOVERED) { dev_printk(KERN_DEBUG, &dev->dev, "AER driver " "successfully recovered\n"); @@ -661,6 +742,28 @@ static int get_device_error_info(struct pci_dev *dev, struct aer_err_info *info) return AER_SUCCESS; } +static inline void aer_process_err_devices(struct pcie_device *p_device, + struct aer_err_info *e_info) +{ + int i; + + if (!e_info->dev[0]) { + dev_printk(KERN_DEBUG, &p_device->port->dev, + "can't find device of ID%04x\n", + e_info->id); + } + + for (i = 0; i < e_info->error_dev_num && e_info->dev[i]; i++) { + if (get_device_error_info(e_info->dev[i], e_info) == + AER_SUCCESS) { + aer_print_error(e_info->dev[i], e_info); + handle_error_source(p_device, + e_info->dev[i], + e_info); + } + } +} + /** * aer_isr_one_error - consume an error detected by root port * @p_device: pointer to error root port service device @@ -669,10 +772,16 @@ static int get_device_error_info(struct pci_dev *dev, struct aer_err_info *info) static void aer_isr_one_error(struct pcie_device *p_device, struct aer_err_source *e_src) { - struct device *s_device; - struct aer_err_info e_info = {0, 0, 0,}; + struct aer_err_info *e_info; int i; - u16 id; + + /* struct aer_err_info might be big, so we allocate it with slab */ + e_info = kmalloc(sizeof(struct aer_err_info), GFP_KERNEL); + if (e_info == NULL) { + dev_printk(KERN_DEBUG, &p_device->port->dev, + "Can't allocate mem when processing AER errors\n"); + return; + } /* * There is a possibility that both correctable error and @@ -684,31 +793,26 @@ static void aer_isr_one_error(struct pcie_device *p_device, if (!(e_src->status & i)) continue; + memset(e_info, 0, sizeof(struct aer_err_info)); + /* Init comprehensive error information */ if (i & PCI_ERR_ROOT_COR_RCV) { - id = ERR_COR_ID(e_src->id); - e_info.severity = AER_CORRECTABLE; + e_info->id = ERR_COR_ID(e_src->id); + e_info->severity = AER_CORRECTABLE; } else { - id = ERR_UNCOR_ID(e_src->id); - e_info.severity = ((e_src->status >> 6) & 1); + e_info->id = ERR_UNCOR_ID(e_src->id); + e_info->severity = ((e_src->status >> 6) & 1); } if (e_src->status & (PCI_ERR_ROOT_MULTI_COR_RCV | PCI_ERR_ROOT_MULTI_UNCOR_RCV)) - e_info.flags |= AER_MULTI_ERROR_VALID_FLAG; - if (!(s_device = find_source_device(p_device->port, id))) { - printk(KERN_DEBUG "%s->can't find device of ID%04x\n", - __func__, id); - continue; - } - if (get_device_error_info(to_pci_dev(s_device), &e_info) == - AER_SUCCESS) { - aer_print_error(to_pci_dev(s_device), &e_info); - handle_error_source(p_device, - to_pci_dev(s_device), - e_info); - } + e_info->flags |= AER_MULTI_ERROR_VALID_FLAG; + + find_source_device(p_device->port, e_info); + aer_process_err_devices(p_device, e_info); } + + kfree(e_info); } /** diff --git a/drivers/pci/pcie/aer/ecrc.c b/drivers/pci/pcie/aer/ecrc.c new file mode 100644 index 00000000000..a928d8ab6bd --- /dev/null +++ b/drivers/pci/pcie/aer/ecrc.c @@ -0,0 +1,131 @@ +/* + * Enables/disables PCIe ECRC checking. + * + * (C) Copyright 2009 Hewlett-Packard Development Company, L.P. + * Andrew Patterson <andrew.patterson@hp.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + * 02111-1307, USA. + * + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/pci.h> +#include <linux/pci_regs.h> +#include <linux/errno.h> +#include "../../pci.h" + +#define ECRC_POLICY_DEFAULT 0 /* ECRC set by BIOS */ +#define ECRC_POLICY_OFF 1 /* ECRC off for performance */ +#define ECRC_POLICY_ON 2 /* ECRC on for data integrity */ + +static int ecrc_policy = ECRC_POLICY_DEFAULT; + +static const char *ecrc_policy_str[] = { + [ECRC_POLICY_DEFAULT] = "bios", + [ECRC_POLICY_OFF] = "off", + [ECRC_POLICY_ON] = "on" +}; + +/** + * enable_ercr_checking - enable PCIe ECRC checking for a device + * @dev: the PCI device + * + * Returns 0 on success, or negative on failure. + */ +static int enable_ecrc_checking(struct pci_dev *dev) +{ + int pos; + u32 reg32; + + if (!dev->is_pcie) + return -ENODEV; + + pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR); + if (!pos) + return -ENODEV; + + pci_read_config_dword(dev, pos + PCI_ERR_CAP, ®32); + if (reg32 & PCI_ERR_CAP_ECRC_GENC) + reg32 |= PCI_ERR_CAP_ECRC_GENE; + if (reg32 & PCI_ERR_CAP_ECRC_CHKC) + reg32 |= PCI_ERR_CAP_ECRC_CHKE; + pci_write_config_dword(dev, pos + PCI_ERR_CAP, reg32); + + return 0; +} + +/** + * disable_ercr_checking - disables PCIe ECRC checking for a device + * @dev: the PCI device + * + * Returns 0 on success, or negative on failure. + */ +static int disable_ecrc_checking(struct pci_dev *dev) +{ + int pos; + u32 reg32; + + if (!dev->is_pcie) + return -ENODEV; + + pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR); + if (!pos) + return -ENODEV; + + pci_read_config_dword(dev, pos + PCI_ERR_CAP, ®32); + reg32 &= ~(PCI_ERR_CAP_ECRC_GENE | PCI_ERR_CAP_ECRC_CHKE); + pci_write_config_dword(dev, pos + PCI_ERR_CAP, reg32); + + return 0; +} + +/** + * pcie_set_ecrc_checking - set/unset PCIe ECRC checking for a device based on global policy + * @dev: the PCI device + */ +void pcie_set_ecrc_checking(struct pci_dev *dev) +{ + switch (ecrc_policy) { + case ECRC_POLICY_DEFAULT: + return; + case ECRC_POLICY_OFF: + disable_ecrc_checking(dev); + break; + case ECRC_POLICY_ON: + enable_ecrc_checking(dev); + break; + default: + return; + } +} + +/** + * pcie_ecrc_get_policy - parse kernel command-line ecrc option + */ +void pcie_ecrc_get_policy(char *str) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(ecrc_policy_str); i++) + if (!strncmp(str, ecrc_policy_str[i], + strlen(ecrc_policy_str[i]))) + break; + if (i >= ARRAY_SIZE(ecrc_policy_str)) + return; + + ecrc_policy = i; +} diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c index b0367f168af..3d27c97e048 100644 --- a/drivers/pci/pcie/aspm.c +++ b/drivers/pci/pcie/aspm.c @@ -26,40 +26,36 @@ #endif #define MODULE_PARAM_PREFIX "pcie_aspm." -struct endpoint_state { - unsigned int l0s_acceptable_latency; - unsigned int l1_acceptable_latency; +struct aspm_latency { + u32 l0s; /* L0s latency (nsec) */ + u32 l1; /* L1 latency (nsec) */ }; struct pcie_link_state { - struct list_head sibiling; - struct pci_dev *pdev; - bool downstream_has_switch; - - struct pcie_link_state *parent; - struct list_head children; - struct list_head link; + struct pci_dev *pdev; /* Upstream component of the Link */ + struct pcie_link_state *root; /* pointer to the root port link */ + struct pcie_link_state *parent; /* pointer to the parent Link state */ + struct list_head sibling; /* node in link_list */ + struct list_head children; /* list of child link states */ + struct list_head link; /* node in parent's children list */ /* ASPM state */ - unsigned int support_state; - unsigned int enabled_state; - unsigned int bios_aspm_state; - /* upstream component */ - unsigned int l0s_upper_latency; - unsigned int l1_upper_latency; - /* downstream component */ - unsigned int l0s_down_latency; - unsigned int l1_down_latency; - /* Clock PM state*/ - unsigned int clk_pm_capable; - unsigned int clk_pm_enabled; - unsigned int bios_clk_state; + u32 aspm_support:2; /* Supported ASPM state */ + u32 aspm_enabled:2; /* Enabled ASPM state */ + u32 aspm_default:2; /* Default ASPM state by BIOS */ + + /* Clock PM state */ + u32 clkpm_capable:1; /* Clock PM capable? */ + u32 clkpm_enabled:1; /* Current Clock PM state */ + u32 clkpm_default:1; /* Default Clock PM state by BIOS */ + /* Latencies */ + struct aspm_latency latency; /* Exit latency */ /* - * A pcie downstream port only has one slot under it, so at most there - * are 8 functions + * Endpoint acceptable latencies. A pcie downstream port only + * has one slot under it, so at most there are 8 functions. */ - struct endpoint_state endpoints[8]; + struct aspm_latency acceptable[8]; }; static int aspm_disabled, aspm_force; @@ -78,27 +74,23 @@ static const char *policy_str[] = { #define LINK_RETRAIN_TIMEOUT HZ -static int policy_to_aspm_state(struct pci_dev *pdev) +static int policy_to_aspm_state(struct pcie_link_state *link) { - struct pcie_link_state *link_state = pdev->link_state; - switch (aspm_policy) { case POLICY_PERFORMANCE: /* Disable ASPM and Clock PM */ return 0; case POLICY_POWERSAVE: /* Enable ASPM L0s/L1 */ - return PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1; + return PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1; case POLICY_DEFAULT: - return link_state->bios_aspm_state; + return link->aspm_default; } return 0; } -static int policy_to_clkpm_state(struct pci_dev *pdev) +static int policy_to_clkpm_state(struct pcie_link_state *link) { - struct pcie_link_state *link_state = pdev->link_state; - switch (aspm_policy) { case POLICY_PERFORMANCE: /* Disable ASPM and Clock PM */ @@ -107,73 +99,78 @@ static int policy_to_clkpm_state(struct pci_dev *pdev) /* Disable Clock PM */ return 1; case POLICY_DEFAULT: - return link_state->bios_clk_state; + return link->clkpm_default; } return 0; } -static void pcie_set_clock_pm(struct pci_dev *pdev, int enable) +static void pcie_set_clkpm_nocheck(struct pcie_link_state *link, int enable) { - struct pci_dev *child_dev; int pos; u16 reg16; - struct pcie_link_state *link_state = pdev->link_state; + struct pci_dev *child; + struct pci_bus *linkbus = link->pdev->subordinate; - list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) { - pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP); + list_for_each_entry(child, &linkbus->devices, bus_list) { + pos = pci_find_capability(child, PCI_CAP_ID_EXP); if (!pos) return; - pci_read_config_word(child_dev, pos + PCI_EXP_LNKCTL, ®16); + pci_read_config_word(child, pos + PCI_EXP_LNKCTL, ®16); if (enable) reg16 |= PCI_EXP_LNKCTL_CLKREQ_EN; else reg16 &= ~PCI_EXP_LNKCTL_CLKREQ_EN; - pci_write_config_word(child_dev, pos + PCI_EXP_LNKCTL, reg16); + pci_write_config_word(child, pos + PCI_EXP_LNKCTL, reg16); } - link_state->clk_pm_enabled = !!enable; + link->clkpm_enabled = !!enable; } -static void pcie_check_clock_pm(struct pci_dev *pdev, int blacklist) +static void pcie_set_clkpm(struct pcie_link_state *link, int enable) { - int pos; + /* Don't enable Clock PM if the link is not Clock PM capable */ + if (!link->clkpm_capable && enable) + return; + /* Need nothing if the specified equals to current state */ + if (link->clkpm_enabled == enable) + return; + pcie_set_clkpm_nocheck(link, enable); +} + +static void pcie_clkpm_cap_init(struct pcie_link_state *link, int blacklist) +{ + int pos, capable = 1, enabled = 1; u32 reg32; u16 reg16; - int capable = 1, enabled = 1; - struct pci_dev *child_dev; - struct pcie_link_state *link_state = pdev->link_state; + struct pci_dev *child; + struct pci_bus *linkbus = link->pdev->subordinate; /* All functions should have the same cap and state, take the worst */ - list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) { - pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP); + list_for_each_entry(child, &linkbus->devices, bus_list) { + pos = pci_find_capability(child, PCI_CAP_ID_EXP); if (!pos) return; - pci_read_config_dword(child_dev, pos + PCI_EXP_LNKCAP, ®32); + pci_read_config_dword(child, pos + PCI_EXP_LNKCAP, ®32); if (!(reg32 & PCI_EXP_LNKCAP_CLKPM)) { capable = 0; enabled = 0; break; } - pci_read_config_word(child_dev, pos + PCI_EXP_LNKCTL, ®16); + pci_read_config_word(child, pos + PCI_EXP_LNKCTL, ®16); if (!(reg16 & PCI_EXP_LNKCTL_CLKREQ_EN)) enabled = 0; } - link_state->clk_pm_enabled = enabled; - link_state->bios_clk_state = enabled; - if (!blacklist) { - link_state->clk_pm_capable = capable; - pcie_set_clock_pm(pdev, policy_to_clkpm_state(pdev)); - } else { - link_state->clk_pm_capable = 0; - pcie_set_clock_pm(pdev, 0); - } + link->clkpm_enabled = enabled; + link->clkpm_default = enabled; + link->clkpm_capable = (blacklist) ? 0 : capable; } -static bool pcie_aspm_downstream_has_switch(struct pci_dev *pdev) +static bool pcie_aspm_downstream_has_switch(struct pcie_link_state *link) { - struct pci_dev *child_dev; + struct pci_dev *child; + struct pci_bus *linkbus = link->pdev->subordinate; - list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) { - if (child_dev->pcie_type == PCI_EXP_TYPE_UPSTREAM) + list_for_each_entry(child, &linkbus->devices, bus_list) { + if (child->pcie_type == PCI_EXP_TYPE_UPSTREAM) return true; } return false; @@ -184,289 +181,263 @@ static bool pcie_aspm_downstream_has_switch(struct pci_dev *pdev) * could use common clock. If they are, configure them to use the * common clock. That will reduce the ASPM state exit latency. */ -static void pcie_aspm_configure_common_clock(struct pci_dev *pdev) +static void pcie_aspm_configure_common_clock(struct pcie_link_state *link) { - int pos, child_pos, i = 0; - u16 reg16 = 0; - struct pci_dev *child_dev; - int same_clock = 1; + int ppos, cpos, same_clock = 1; + u16 reg16, parent_reg, child_reg[8]; unsigned long start_jiffies; - u16 child_regs[8], parent_reg; + struct pci_dev *child, *parent = link->pdev; + struct pci_bus *linkbus = parent->subordinate; /* - * all functions of a slot should have the same Slot Clock + * All functions of a slot should have the same Slot Clock * Configuration, so just check one function - * */ - child_dev = list_entry(pdev->subordinate->devices.next, struct pci_dev, - bus_list); - BUG_ON(!child_dev->is_pcie); + */ + child = list_entry(linkbus->devices.next, struct pci_dev, bus_list); + BUG_ON(!child->is_pcie); /* Check downstream component if bit Slot Clock Configuration is 1 */ - child_pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP); - pci_read_config_word(child_dev, child_pos + PCI_EXP_LNKSTA, ®16); + cpos = pci_find_capability(child, PCI_CAP_ID_EXP); + pci_read_config_word(child, cpos + PCI_EXP_LNKSTA, ®16); if (!(reg16 & PCI_EXP_LNKSTA_SLC)) same_clock = 0; /* Check upstream component if bit Slot Clock Configuration is 1 */ - pos = pci_find_capability(pdev, PCI_CAP_ID_EXP); - pci_read_config_word(pdev, pos + PCI_EXP_LNKSTA, ®16); + ppos = pci_find_capability(parent, PCI_CAP_ID_EXP); + pci_read_config_word(parent, ppos + PCI_EXP_LNKSTA, ®16); if (!(reg16 & PCI_EXP_LNKSTA_SLC)) same_clock = 0; /* Configure downstream component, all functions */ - list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) { - child_pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP); - pci_read_config_word(child_dev, child_pos + PCI_EXP_LNKCTL, - ®16); - child_regs[i] = reg16; + list_for_each_entry(child, &linkbus->devices, bus_list) { + cpos = pci_find_capability(child, PCI_CAP_ID_EXP); + pci_read_config_word(child, cpos + PCI_EXP_LNKCTL, ®16); + child_reg[PCI_FUNC(child->devfn)] = reg16; if (same_clock) reg16 |= PCI_EXP_LNKCTL_CCC; else reg16 &= ~PCI_EXP_LNKCTL_CCC; - pci_write_config_word(child_dev, child_pos + PCI_EXP_LNKCTL, - reg16); - i++; + pci_write_config_word(child, cpos + PCI_EXP_LNKCTL, reg16); } /* Configure upstream component */ - pci_read_config_word(pdev, pos + PCI_EXP_LNKCTL, ®16); + pci_read_config_word(parent, ppos + PCI_EXP_LNKCTL, ®16); parent_reg = reg16; if (same_clock) reg16 |= PCI_EXP_LNKCTL_CCC; else reg16 &= ~PCI_EXP_LNKCTL_CCC; - pci_write_config_word(pdev, pos + PCI_EXP_LNKCTL, reg16); + pci_write_config_word(parent, ppos + PCI_EXP_LNKCTL, reg16); - /* retrain link */ + /* Retrain link */ reg16 |= PCI_EXP_LNKCTL_RL; - pci_write_config_word(pdev, pos + PCI_EXP_LNKCTL, reg16); + pci_write_config_word(parent, ppos + PCI_EXP_LNKCTL, reg16); - /* Wait for link training end */ - /* break out after waiting for timeout */ + /* Wait for link training end. Break out after waiting for timeout */ start_jiffies = jiffies; for (;;) { - pci_read_config_word(pdev, pos + PCI_EXP_LNKSTA, ®16); + pci_read_config_word(parent, ppos + PCI_EXP_LNKSTA, ®16); if (!(reg16 & PCI_EXP_LNKSTA_LT)) break; if (time_after(jiffies, start_jiffies + LINK_RETRAIN_TIMEOUT)) break; msleep(1); } - /* training failed -> recover */ - if (reg16 & PCI_EXP_LNKSTA_LT) { - dev_printk (KERN_ERR, &pdev->dev, "ASPM: Could not configure" - " common clock\n"); - i = 0; - list_for_each_entry(child_dev, &pdev->subordinate->devices, - bus_list) { - child_pos = pci_find_capability(child_dev, - PCI_CAP_ID_EXP); - pci_write_config_word(child_dev, - child_pos + PCI_EXP_LNKCTL, - child_regs[i]); - i++; - } - pci_write_config_word(pdev, pos + PCI_EXP_LNKCTL, parent_reg); + if (!(reg16 & PCI_EXP_LNKSTA_LT)) + return; + + /* Training failed. Restore common clock configurations */ + dev_printk(KERN_ERR, &parent->dev, + "ASPM: Could not configure common clock\n"); + list_for_each_entry(child, &linkbus->devices, bus_list) { + cpos = pci_find_capability(child, PCI_CAP_ID_EXP); + pci_write_config_word(child, cpos + PCI_EXP_LNKCTL, + child_reg[PCI_FUNC(child->devfn)]); } + pci_write_config_word(parent, ppos + PCI_EXP_LNKCTL, parent_reg); } -/* - * calc_L0S_latency: Convert L0s latency encoding to ns - */ -static unsigned int calc_L0S_latency(unsigned int latency_encoding, int ac) +/* Convert L0s latency encoding to ns */ +static u32 calc_l0s_latency(u32 encoding) { - unsigned int ns = 64; + if (encoding == 0x7) + return (5 * 1000); /* > 4us */ + return (64 << encoding); +} - if (latency_encoding == 0x7) { - if (ac) - ns = -1U; - else - ns = 5*1000; /* > 4us */ - } else - ns *= (1 << latency_encoding); - return ns; +/* Convert L0s acceptable latency encoding to ns */ +static u32 calc_l0s_acceptable(u32 encoding) +{ + if (encoding == 0x7) + return -1U; + return (64 << encoding); } -/* - * calc_L1_latency: Convert L1 latency encoding to ns - */ -static unsigned int calc_L1_latency(unsigned int latency_encoding, int ac) +/* Convert L1 latency encoding to ns */ +static u32 calc_l1_latency(u32 encoding) { - unsigned int ns = 1000; + if (encoding == 0x7) + return (65 * 1000); /* > 64us */ + return (1000 << encoding); +} - if (latency_encoding == 0x7) { - if (ac) - ns = -1U; - else - ns = 65*1000; /* > 64us */ - } else - ns *= (1 << latency_encoding); - return ns; +/* Convert L1 acceptable latency encoding to ns */ +static u32 calc_l1_acceptable(u32 encoding) +{ + if (encoding == 0x7) + return -1U; + return (1000 << encoding); } static void pcie_aspm_get_cap_device(struct pci_dev *pdev, u32 *state, - unsigned int *l0s, unsigned int *l1, unsigned int *enabled) + u32 *l0s, u32 *l1, u32 *enabled) { int pos; u16 reg16; - u32 reg32; - unsigned int latency; + u32 reg32, encoding; + *l0s = *l1 = *enabled = 0; pos = pci_find_capability(pdev, PCI_CAP_ID_EXP); pci_read_config_dword(pdev, pos + PCI_EXP_LNKCAP, ®32); *state = (reg32 & PCI_EXP_LNKCAP_ASPMS) >> 10; if (*state != PCIE_LINK_STATE_L0S && - *state != (PCIE_LINK_STATE_L1|PCIE_LINK_STATE_L0S)) + *state != (PCIE_LINK_STATE_L1 | PCIE_LINK_STATE_L0S)) *state = 0; if (*state == 0) return; - latency = (reg32 & PCI_EXP_LNKCAP_L0SEL) >> 12; - *l0s = calc_L0S_latency(latency, 0); + encoding = (reg32 & PCI_EXP_LNKCAP_L0SEL) >> 12; + *l0s = calc_l0s_latency(encoding); if (*state & PCIE_LINK_STATE_L1) { - latency = (reg32 & PCI_EXP_LNKCAP_L1EL) >> 15; - *l1 = calc_L1_latency(latency, 0); + encoding = (reg32 & PCI_EXP_LNKCAP_L1EL) >> 15; + *l1 = calc_l1_latency(encoding); } pci_read_config_word(pdev, pos + PCI_EXP_LNKCTL, ®16); - *enabled = reg16 & (PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1); + *enabled = reg16 & (PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1); } -static void pcie_aspm_cap_init(struct pci_dev *pdev) +static void pcie_aspm_cap_init(struct pcie_link_state *link, int blacklist) { - struct pci_dev *child_dev; - u32 state, tmp; - struct pcie_link_state *link_state = pdev->link_state; + u32 support, l0s, l1, enabled; + struct pci_dev *child, *parent = link->pdev; + struct pci_bus *linkbus = parent->subordinate; + + if (blacklist) { + /* Set support state to 0, so we will disable ASPM later */ + link->aspm_support = 0; + link->aspm_default = 0; + link->aspm_enabled = PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1; + return; + } + + /* Configure common clock before checking latencies */ + pcie_aspm_configure_common_clock(link); /* upstream component states */ - pcie_aspm_get_cap_device(pdev, &link_state->support_state, - &link_state->l0s_upper_latency, - &link_state->l1_upper_latency, - &link_state->enabled_state); + pcie_aspm_get_cap_device(parent, &support, &l0s, &l1, &enabled); + link->aspm_support = support; + link->latency.l0s = l0s; + link->latency.l1 = l1; + link->aspm_enabled = enabled; + /* downstream component states, all functions have the same setting */ - child_dev = list_entry(pdev->subordinate->devices.next, struct pci_dev, - bus_list); - pcie_aspm_get_cap_device(child_dev, &state, - &link_state->l0s_down_latency, - &link_state->l1_down_latency, - &tmp); - link_state->support_state &= state; - if (!link_state->support_state) + child = list_entry(linkbus->devices.next, struct pci_dev, bus_list); + pcie_aspm_get_cap_device(child, &support, &l0s, &l1, &enabled); + link->aspm_support &= support; + link->latency.l0s = max_t(u32, link->latency.l0s, l0s); + link->latency.l1 = max_t(u32, link->latency.l1, l1); + + if (!link->aspm_support) return; - link_state->enabled_state &= link_state->support_state; - link_state->bios_aspm_state = link_state->enabled_state; + + link->aspm_enabled &= link->aspm_support; + link->aspm_default = link->aspm_enabled; /* ENDPOINT states*/ - list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) { + list_for_each_entry(child, &linkbus->devices, bus_list) { int pos; - u32 reg32; - unsigned int latency; - struct endpoint_state *ep_state = - &link_state->endpoints[PCI_FUNC(child_dev->devfn)]; + u32 reg32, encoding; + struct aspm_latency *acceptable = + &link->acceptable[PCI_FUNC(child->devfn)]; - if (child_dev->pcie_type != PCI_EXP_TYPE_ENDPOINT && - child_dev->pcie_type != PCI_EXP_TYPE_LEG_END) + if (child->pcie_type != PCI_EXP_TYPE_ENDPOINT && + child->pcie_type != PCI_EXP_TYPE_LEG_END) continue; - pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP); - pci_read_config_dword(child_dev, pos + PCI_EXP_DEVCAP, ®32); - latency = (reg32 & PCI_EXP_DEVCAP_L0S) >> 6; - latency = calc_L0S_latency(latency, 1); - ep_state->l0s_acceptable_latency = latency; - if (link_state->support_state & PCIE_LINK_STATE_L1) { - latency = (reg32 & PCI_EXP_DEVCAP_L1) >> 9; - latency = calc_L1_latency(latency, 1); - ep_state->l1_acceptable_latency = latency; + pos = pci_find_capability(child, PCI_CAP_ID_EXP); + pci_read_config_dword(child, pos + PCI_EXP_DEVCAP, ®32); + encoding = (reg32 & PCI_EXP_DEVCAP_L0S) >> 6; + acceptable->l0s = calc_l0s_acceptable(encoding); + if (link->aspm_support & PCIE_LINK_STATE_L1) { + encoding = (reg32 & PCI_EXP_DEVCAP_L1) >> 9; + acceptable->l1 = calc_l1_acceptable(encoding); } } } -static unsigned int __pcie_aspm_check_state_one(struct pci_dev *pdev, - unsigned int state) -{ - struct pci_dev *parent_dev, *tmp_dev; - unsigned int latency, l1_latency = 0; - struct pcie_link_state *link_state; - struct endpoint_state *ep_state; - - parent_dev = pdev->bus->self; - link_state = parent_dev->link_state; - state &= link_state->support_state; - if (state == 0) - return 0; - ep_state = &link_state->endpoints[PCI_FUNC(pdev->devfn)]; - - /* - * Check latency for endpoint device. - * TBD: The latency from the endpoint to root complex vary per - * switch's upstream link state above the device. Here we just do a - * simple check which assumes all links above the device can be in L1 - * state, that is we just consider the worst case. If switch's upstream - * link can't be put into L0S/L1, then our check is too strictly. - */ - tmp_dev = pdev; - while (state & (PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1)) { - parent_dev = tmp_dev->bus->self; - link_state = parent_dev->link_state; - if (state & PCIE_LINK_STATE_L0S) { - latency = max_t(unsigned int, - link_state->l0s_upper_latency, - link_state->l0s_down_latency); - if (latency > ep_state->l0s_acceptable_latency) - state &= ~PCIE_LINK_STATE_L0S; - } - if (state & PCIE_LINK_STATE_L1) { - latency = max_t(unsigned int, - link_state->l1_upper_latency, - link_state->l1_down_latency); - if (latency + l1_latency > - ep_state->l1_acceptable_latency) - state &= ~PCIE_LINK_STATE_L1; - } - if (!parent_dev->bus->self) /* parent_dev is a root port */ - break; - else { - /* - * parent_dev is the downstream port of a switch, make - * tmp_dev the upstream port of the switch - */ - tmp_dev = parent_dev->bus->self; - /* - * every switch on the path to root complex need 1 more - * microsecond for L1. Spec doesn't mention L0S. - */ - if (state & PCIE_LINK_STATE_L1) - l1_latency += 1000; - } +/** + * __pcie_aspm_check_state_one - check latency for endpoint device. + * @endpoint: pointer to the struct pci_dev of endpoint device + * + * TBD: The latency from the endpoint to root complex vary per switch's + * upstream link state above the device. Here we just do a simple check + * which assumes all links above the device can be in L1 state, that + * is we just consider the worst case. If switch's upstream link can't + * be put into L0S/L1, then our check is too strictly. + */ +static u32 __pcie_aspm_check_state_one(struct pci_dev *endpoint, u32 state) +{ + u32 l1_switch_latency = 0; + struct aspm_latency *acceptable; + struct pcie_link_state *link; + + link = endpoint->bus->self->link_state; + state &= link->aspm_support; + acceptable = &link->acceptable[PCI_FUNC(endpoint->devfn)]; + + while (link && state) { + if ((state & PCIE_LINK_STATE_L0S) && + (link->latency.l0s > acceptable->l0s)) + state &= ~PCIE_LINK_STATE_L0S; + if ((state & PCIE_LINK_STATE_L1) && + (link->latency.l1 + l1_switch_latency > acceptable->l1)) + state &= ~PCIE_LINK_STATE_L1; + link = link->parent; + /* + * Every switch on the path to root complex need 1 + * more microsecond for L1. Spec doesn't mention L0s. + */ + l1_switch_latency += 1000; } return state; } -static unsigned int pcie_aspm_check_state(struct pci_dev *pdev, - unsigned int state) +static u32 pcie_aspm_check_state(struct pcie_link_state *link, u32 state) { - struct pci_dev *child_dev; + pci_power_t power_state; + struct pci_dev *child; + struct pci_bus *linkbus = link->pdev->subordinate; /* If no child, ignore the link */ - if (list_empty(&pdev->subordinate->devices)) + if (list_empty(&linkbus->devices)) return state; - list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) { - if (child_dev->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE) { - /* - * If downstream component of a link is pci bridge, we - * disable ASPM for now for the link - * */ - state = 0; - break; - } - if ((child_dev->pcie_type != PCI_EXP_TYPE_ENDPOINT && - child_dev->pcie_type != PCI_EXP_TYPE_LEG_END)) + + list_for_each_entry(child, &linkbus->devices, bus_list) { + /* + * If downstream component of a link is pci bridge, we + * disable ASPM for now for the link + */ + if (child->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE) + return 0; + + if ((child->pcie_type != PCI_EXP_TYPE_ENDPOINT && + child->pcie_type != PCI_EXP_TYPE_LEG_END)) continue; /* Device not in D0 doesn't need check latency */ - if (child_dev->current_state == PCI_D1 || - child_dev->current_state == PCI_D2 || - child_dev->current_state == PCI_D3hot || - child_dev->current_state == PCI_D3cold) + power_state = child->current_state; + if (power_state == PCI_D1 || power_state == PCI_D2 || + power_state == PCI_D3hot || power_state == PCI_D3cold) continue; - state = __pcie_aspm_check_state_one(child_dev, state); + state = __pcie_aspm_check_state_one(child, state); } return state; } @@ -482,90 +453,71 @@ static void __pcie_aspm_config_one_dev(struct pci_dev *pdev, unsigned int state) pci_write_config_word(pdev, pos + PCI_EXP_LNKCTL, reg16); } -static void __pcie_aspm_config_link(struct pci_dev *pdev, unsigned int state) +static void __pcie_aspm_config_link(struct pcie_link_state *link, u32 state) { - struct pci_dev *child_dev; - int valid = 1; - struct pcie_link_state *link_state = pdev->link_state; + struct pci_dev *child, *parent = link->pdev; + struct pci_bus *linkbus = parent->subordinate; /* If no child, disable the link */ - if (list_empty(&pdev->subordinate->devices)) + if (list_empty(&linkbus->devices)) state = 0; /* - * if the downstream component has pci bridge function, don't do ASPM - * now + * If the downstream component has pci bridge function, don't + * do ASPM now. */ - list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) { - if (child_dev->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE) { - valid = 0; - break; - } + list_for_each_entry(child, &linkbus->devices, bus_list) { + if (child->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE) + return; } - if (!valid) - return; - /* - * spec 2.0 suggests all functions should be configured the same - * setting for ASPM. Enabling ASPM L1 should be done in upstream - * component first and then downstream, and vice versa for disabling - * ASPM L1. Spec doesn't mention L0S. + * Spec 2.0 suggests all functions should be configured the + * same setting for ASPM. Enabling ASPM L1 should be done in + * upstream component first and then downstream, and vice + * versa for disabling ASPM L1. Spec doesn't mention L0S. */ if (state & PCIE_LINK_STATE_L1) - __pcie_aspm_config_one_dev(pdev, state); + __pcie_aspm_config_one_dev(parent, state); - list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) - __pcie_aspm_config_one_dev(child_dev, state); + list_for_each_entry(child, &linkbus->devices, bus_list) + __pcie_aspm_config_one_dev(child, state); if (!(state & PCIE_LINK_STATE_L1)) - __pcie_aspm_config_one_dev(pdev, state); + __pcie_aspm_config_one_dev(parent, state); - link_state->enabled_state = state; + link->aspm_enabled = state; } -static struct pcie_link_state *get_root_port_link(struct pcie_link_state *link) +/* Check the whole hierarchy, and configure each link in the hierarchy */ +static void __pcie_aspm_configure_link_state(struct pcie_link_state *link, + u32 state) { - struct pcie_link_state *root_port_link = link; - while (root_port_link->parent) - root_port_link = root_port_link->parent; - return root_port_link; -} + struct pcie_link_state *leaf, *root = link->root; -/* check the whole hierarchy, and configure each link in the hierarchy */ -static void __pcie_aspm_configure_link_state(struct pci_dev *pdev, - unsigned int state) -{ - struct pcie_link_state *link_state = pdev->link_state; - struct pcie_link_state *root_port_link = get_root_port_link(link_state); - struct pcie_link_state *leaf; + state &= (PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1); - state &= PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1; - - /* check all links who have specific root port link */ - list_for_each_entry(leaf, &link_list, sibiling) { - if (!list_empty(&leaf->children) || - get_root_port_link(leaf) != root_port_link) + /* Check all links who have specific root port link */ + list_for_each_entry(leaf, &link_list, sibling) { + if (!list_empty(&leaf->children) || (leaf->root != root)) continue; - state = pcie_aspm_check_state(leaf->pdev, state); + state = pcie_aspm_check_state(leaf, state); } - /* check root port link too in case it hasn't children */ - state = pcie_aspm_check_state(root_port_link->pdev, state); - - if (link_state->enabled_state == state) + /* Check root port link too in case it hasn't children */ + state = pcie_aspm_check_state(root, state); + if (link->aspm_enabled == state) return; - /* - * we must change the hierarchy. See comments in + * We must change the hierarchy. See comments in * __pcie_aspm_config_link for the order **/ if (state & PCIE_LINK_STATE_L1) { - list_for_each_entry(leaf, &link_list, sibiling) { - if (get_root_port_link(leaf) == root_port_link) - __pcie_aspm_config_link(leaf->pdev, state); + list_for_each_entry(leaf, &link_list, sibling) { + if (leaf->root == root) + __pcie_aspm_config_link(leaf, state); } } else { - list_for_each_entry_reverse(leaf, &link_list, sibiling) { - if (get_root_port_link(leaf) == root_port_link) - __pcie_aspm_config_link(leaf->pdev, state); + list_for_each_entry_reverse(leaf, &link_list, sibling) { + if (leaf->root == root) + __pcie_aspm_config_link(leaf, state); } } } @@ -574,45 +526,42 @@ static void __pcie_aspm_configure_link_state(struct pci_dev *pdev, * pcie_aspm_configure_link_state: enable/disable PCI express link state * @pdev: the root port or switch downstream port */ -static void pcie_aspm_configure_link_state(struct pci_dev *pdev, - unsigned int state) +static void pcie_aspm_configure_link_state(struct pcie_link_state *link, + u32 state) { down_read(&pci_bus_sem); mutex_lock(&aspm_lock); - __pcie_aspm_configure_link_state(pdev, state); + __pcie_aspm_configure_link_state(link, state); mutex_unlock(&aspm_lock); up_read(&pci_bus_sem); } -static void free_link_state(struct pci_dev *pdev) +static void free_link_state(struct pcie_link_state *link) { - kfree(pdev->link_state); - pdev->link_state = NULL; + link->pdev->link_state = NULL; + kfree(link); } static int pcie_aspm_sanity_check(struct pci_dev *pdev) { - struct pci_dev *child_dev; - int child_pos; + struct pci_dev *child; + int pos; u32 reg32; - /* - * Some functions in a slot might not all be PCIE functions, very - * strange. Disable ASPM for the whole slot + * Some functions in a slot might not all be PCIE functions, + * very strange. Disable ASPM for the whole slot */ - list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) { - child_pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP); - if (!child_pos) + list_for_each_entry(child, &pdev->subordinate->devices, bus_list) { + pos = pci_find_capability(child, PCI_CAP_ID_EXP); + if (!pos) return -EINVAL; - /* * Disable ASPM for pre-1.1 PCIe device, we follow MS to use * RBER bit to determine if a function is 1.1 version device */ - pci_read_config_dword(child_dev, child_pos + PCI_EXP_DEVCAP, - ®32); + pci_read_config_dword(child, pos + PCI_EXP_DEVCAP, ®32); if (!(reg32 & PCI_EXP_DEVCAP_RBER) && !aspm_force) { - dev_printk(KERN_INFO, &child_dev->dev, "disabling ASPM" + dev_printk(KERN_INFO, &child->dev, "disabling ASPM" " on pre-1.1 PCIe device. You can enable it" " with 'pcie_aspm=force'\n"); return -EINVAL; @@ -621,6 +570,47 @@ static int pcie_aspm_sanity_check(struct pci_dev *pdev) return 0; } +static struct pcie_link_state *pcie_aspm_setup_link_state(struct pci_dev *pdev) +{ + struct pcie_link_state *link; + int blacklist = !!pcie_aspm_sanity_check(pdev); + + link = kzalloc(sizeof(*link), GFP_KERNEL); + if (!link) + return NULL; + INIT_LIST_HEAD(&link->sibling); + INIT_LIST_HEAD(&link->children); + INIT_LIST_HEAD(&link->link); + link->pdev = pdev; + if (pdev->pcie_type == PCI_EXP_TYPE_DOWNSTREAM) { + struct pcie_link_state *parent; + parent = pdev->bus->parent->self->link_state; + if (!parent) { + kfree(link); + return NULL; + } + link->parent = parent; + list_add(&link->link, &parent->children); + } + /* Setup a pointer to the root port link */ + if (!link->parent) + link->root = link; + else + link->root = link->parent->root; + + list_add(&link->sibling, &link_list); + + pdev->link_state = link; + + /* Check ASPM capability */ + pcie_aspm_cap_init(link, blacklist); + + /* Check Clock PM capability */ + pcie_clkpm_cap_init(link, blacklist); + + return link; +} + /* * pcie_aspm_init_link_state: Initiate PCI express link state. * It is called after the pcie and its children devices are scaned. @@ -628,75 +618,47 @@ static int pcie_aspm_sanity_check(struct pci_dev *pdev) */ void pcie_aspm_init_link_state(struct pci_dev *pdev) { - unsigned int state; - struct pcie_link_state *link_state; - int error = 0; - int blacklist; + u32 state; + struct pcie_link_state *link; if (aspm_disabled || !pdev->is_pcie || pdev->link_state) return; if (pdev->pcie_type != PCI_EXP_TYPE_ROOT_PORT && - pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM) + pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM) + return; + + /* VIA has a strange chipset, root port is under a bridge */ + if (pdev->pcie_type == PCI_EXP_TYPE_ROOT_PORT && + pdev->bus->self) return; + down_read(&pci_bus_sem); if (list_empty(&pdev->subordinate->devices)) goto out; - blacklist = !!pcie_aspm_sanity_check(pdev); - mutex_lock(&aspm_lock); - - link_state = kzalloc(sizeof(*link_state), GFP_KERNEL); - if (!link_state) - goto unlock_out; - - link_state->downstream_has_switch = pcie_aspm_downstream_has_switch(pdev); - INIT_LIST_HEAD(&link_state->children); - INIT_LIST_HEAD(&link_state->link); - if (pdev->bus->self) {/* this is a switch */ - struct pcie_link_state *parent_link_state; - - parent_link_state = pdev->bus->parent->self->link_state; - if (!parent_link_state) { - kfree(link_state); - goto unlock_out; - } - list_add(&link_state->link, &parent_link_state->children); - link_state->parent = parent_link_state; - } - - pdev->link_state = link_state; - - if (!blacklist) { - pcie_aspm_configure_common_clock(pdev); - pcie_aspm_cap_init(pdev); + link = pcie_aspm_setup_link_state(pdev); + if (!link) + goto unlock; + /* + * Setup initial ASPM state + * + * If link has switch, delay the link config. The leaf link + * initialization will config the whole hierarchy. But we must + * make sure BIOS doesn't set unsupported link state. + */ + if (pcie_aspm_downstream_has_switch(link)) { + state = pcie_aspm_check_state(link, link->aspm_default); + __pcie_aspm_config_link(link, state); } else { - link_state->enabled_state = PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1; - link_state->bios_aspm_state = 0; - /* Set support state to 0, so we will disable ASPM later */ - link_state->support_state = 0; + state = policy_to_aspm_state(link); + __pcie_aspm_configure_link_state(link, state); } - link_state->pdev = pdev; - list_add(&link_state->sibiling, &link_list); - - if (link_state->downstream_has_switch) { - /* - * If link has switch, delay the link config. The leaf link - * initialization will config the whole hierarchy. but we must - * make sure BIOS doesn't set unsupported link state - **/ - state = pcie_aspm_check_state(pdev, link_state->bios_aspm_state); - __pcie_aspm_config_link(pdev, state); - } else - __pcie_aspm_configure_link_state(pdev, - policy_to_aspm_state(pdev)); - - pcie_check_clock_pm(pdev, blacklist); - -unlock_out: - if (error) - free_link_state(pdev); + /* Setup initial Clock PM state */ + state = (link->clkpm_capable) ? policy_to_clkpm_state(link) : 0; + pcie_set_clkpm(link, state); +unlock: mutex_unlock(&aspm_lock); out: up_read(&pci_bus_sem); @@ -725,11 +687,11 @@ void pcie_aspm_exit_link_state(struct pci_dev *pdev) /* All functions are removed, so just disable ASPM for the link */ __pcie_aspm_config_one_dev(parent, 0); - list_del(&link_state->sibiling); + list_del(&link_state->sibling); list_del(&link_state->link); /* Clock PM is for endpoint device */ - free_link_state(parent); + free_link_state(link_state); out: mutex_unlock(&aspm_lock); up_read(&pci_bus_sem); @@ -749,7 +711,7 @@ void pcie_aspm_pm_state_change(struct pci_dev *pdev) * devices changed PM state, we should recheck if latency meets all * functions' requirement */ - pcie_aspm_configure_link_state(pdev, link_state->enabled_state); + pcie_aspm_configure_link_state(link_state, link_state->aspm_enabled); } /* @@ -772,14 +734,12 @@ void pci_disable_link_state(struct pci_dev *pdev, int state) down_read(&pci_bus_sem); mutex_lock(&aspm_lock); link_state = parent->link_state; - link_state->support_state &= - ~(state & (PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1)); - if (state & PCIE_LINK_STATE_CLKPM) - link_state->clk_pm_capable = 0; - - __pcie_aspm_configure_link_state(parent, link_state->enabled_state); - if (!link_state->clk_pm_capable && link_state->clk_pm_enabled) - pcie_set_clock_pm(parent, 0); + link_state->aspm_support &= ~state; + __pcie_aspm_configure_link_state(link_state, link_state->aspm_enabled); + if (state & PCIE_LINK_STATE_CLKPM) { + link_state->clkpm_capable = 0; + pcie_set_clkpm(link_state, 0); + } mutex_unlock(&aspm_lock); up_read(&pci_bus_sem); } @@ -788,7 +748,6 @@ EXPORT_SYMBOL(pci_disable_link_state); static int pcie_aspm_set_policy(const char *val, struct kernel_param *kp) { int i; - struct pci_dev *pdev; struct pcie_link_state *link_state; for (i = 0; i < ARRAY_SIZE(policy_str); i++) @@ -802,14 +761,10 @@ static int pcie_aspm_set_policy(const char *val, struct kernel_param *kp) down_read(&pci_bus_sem); mutex_lock(&aspm_lock); aspm_policy = i; - list_for_each_entry(link_state, &link_list, sibiling) { - pdev = link_state->pdev; - __pcie_aspm_configure_link_state(pdev, - policy_to_aspm_state(pdev)); - if (link_state->clk_pm_capable && - link_state->clk_pm_enabled != policy_to_clkpm_state(pdev)) - pcie_set_clock_pm(pdev, policy_to_clkpm_state(pdev)); - + list_for_each_entry(link_state, &link_list, sibling) { + __pcie_aspm_configure_link_state(link_state, + policy_to_aspm_state(link_state)); + pcie_set_clkpm(link_state, policy_to_clkpm_state(link_state)); } mutex_unlock(&aspm_lock); up_read(&pci_bus_sem); @@ -838,7 +793,7 @@ static ssize_t link_state_show(struct device *dev, struct pci_dev *pci_device = to_pci_dev(dev); struct pcie_link_state *link_state = pci_device->link_state; - return sprintf(buf, "%d\n", link_state->enabled_state); + return sprintf(buf, "%d\n", link_state->aspm_enabled); } static ssize_t link_state_store(struct device *dev, @@ -846,7 +801,7 @@ static ssize_t link_state_store(struct device *dev, const char *buf, size_t n) { - struct pci_dev *pci_device = to_pci_dev(dev); + struct pci_dev *pdev = to_pci_dev(dev); int state; if (n < 1) @@ -854,7 +809,7 @@ static ssize_t link_state_store(struct device *dev, state = buf[0]-'0'; if (state >= 0 && state <= 3) { /* setup link aspm state */ - pcie_aspm_configure_link_state(pci_device, state); + pcie_aspm_configure_link_state(pdev->link_state, state); return n; } @@ -868,7 +823,7 @@ static ssize_t clk_ctl_show(struct device *dev, struct pci_dev *pci_device = to_pci_dev(dev); struct pcie_link_state *link_state = pci_device->link_state; - return sprintf(buf, "%d\n", link_state->clk_pm_enabled); + return sprintf(buf, "%d\n", link_state->clkpm_enabled); } static ssize_t clk_ctl_store(struct device *dev, @@ -876,7 +831,7 @@ static ssize_t clk_ctl_store(struct device *dev, const char *buf, size_t n) { - struct pci_dev *pci_device = to_pci_dev(dev); + struct pci_dev *pdev = to_pci_dev(dev); int state; if (n < 1) @@ -885,7 +840,7 @@ static ssize_t clk_ctl_store(struct device *dev, down_read(&pci_bus_sem); mutex_lock(&aspm_lock); - pcie_set_clock_pm(pci_device, !!state); + pcie_set_clkpm_nocheck(pdev->link_state, !!state); mutex_unlock(&aspm_lock); up_read(&pci_bus_sem); @@ -904,10 +859,10 @@ void pcie_aspm_create_sysfs_dev_files(struct pci_dev *pdev) pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM) || !link_state) return; - if (link_state->support_state) + if (link_state->aspm_support) sysfs_add_file_to_group(&pdev->dev.kobj, &dev_attr_link_state.attr, power_group); - if (link_state->clk_pm_capable) + if (link_state->clkpm_capable) sysfs_add_file_to_group(&pdev->dev.kobj, &dev_attr_clk_ctl.attr, power_group); } @@ -920,10 +875,10 @@ void pcie_aspm_remove_sysfs_dev_files(struct pci_dev *pdev) pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM) || !link_state) return; - if (link_state->support_state) + if (link_state->aspm_support) sysfs_remove_file_from_group(&pdev->dev.kobj, &dev_attr_link_state.attr, power_group); - if (link_state->clk_pm_capable) + if (link_state->clkpm_capable) sysfs_remove_file_from_group(&pdev->dev.kobj, &dev_attr_clk_ctl.attr, power_group); } diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c index e3998250386..13ffdc35ea0 100644 --- a/drivers/pci/pcie/portdrv_core.c +++ b/drivers/pci/pcie/portdrv_core.c @@ -275,7 +275,7 @@ static void pcie_device_init(struct pci_dev *parent, struct pcie_device *dev, memset(device, 0, sizeof(struct device)); device->bus = &pcie_port_bus_type; device->driver = NULL; - device->driver_data = NULL; + dev_set_drvdata(device, NULL); device->release = release_pcie_device; /* callback to free pcie dev */ dev_set_name(device, "%s:pcie%02x", pci_name(parent), get_descriptor_id(port_type, service_type)); diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index f1ae2475fff..40e75f6a505 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -193,7 +193,7 @@ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type, res->flags |= pci_calc_resource_flags(l) | IORESOURCE_SIZEALIGN; if (type == pci_bar_io) { l &= PCI_BASE_ADDRESS_IO_MASK; - mask = PCI_BASE_ADDRESS_IO_MASK & 0xffff; + mask = PCI_BASE_ADDRESS_IO_MASK & IO_SPACE_LIMIT; } else { l &= PCI_BASE_ADDRESS_MEM_MASK; mask = (u32)PCI_BASE_ADDRESS_MEM_MASK; @@ -237,6 +237,8 @@ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type, dev_printk(KERN_DEBUG, &dev->dev, "reg %x 64bit mmio: %pR\n", pos, res); } + + res->flags |= IORESOURCE_MEM_64; } else { sz = pci_size(l, sz, mask); @@ -287,7 +289,7 @@ void __devinit pci_read_bridge_bases(struct pci_bus *child) struct resource *res; int i; - if (!child->parent) /* It's a host bus, nothing to read */ + if (pci_is_root_bus(child)) /* It's a host bus, nothing to read */ return; if (dev->transparent) { @@ -362,7 +364,10 @@ void __devinit pci_read_bridge_bases(struct pci_bus *child) } } if (base <= limit) { - res->flags = (mem_base_lo & PCI_MEMORY_RANGE_TYPE_MASK) | IORESOURCE_MEM | IORESOURCE_PREFETCH; + res->flags = (mem_base_lo & PCI_PREF_RANGE_TYPE_MASK) | + IORESOURCE_MEM | IORESOURCE_PREFETCH; + if (res->flags & PCI_PREF_RANGE_TYPE_64) + res->flags |= IORESOURCE_MEM_64; res->start = base; res->end = limit + 0xfffff; dev_printk(KERN_DEBUG, &dev->dev, "bridge %sbit mmio pref: %pR\n", diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 3067673d54f..06b96562396 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1058,6 +1058,11 @@ static void __devinit quirk_no_ata_d3(struct pci_dev *pdev) } DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_SERVERWORKS, PCI_ANY_ID, quirk_no_ata_d3); DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_ATI, PCI_ANY_ID, quirk_no_ata_d3); +/* ALi loses some register settings that we cannot then restore */ +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_AL, PCI_ANY_ID, quirk_no_ata_d3); +/* VIA comes back fine but we need to keep it alive or ACPI GTM failures + occur when mode detecting */ +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_VIA, PCI_ANY_ID, quirk_no_ata_d3); /* This was originally an Alpha specific thing, but it really fits here. * The i82375 PCI/EISA bridge appears as non-classified. Fix that. @@ -1133,6 +1138,7 @@ static void __init asus_hides_smbus_hostbridge(struct pci_dev *dev) switch (dev->subsystem_device) { case 0x1751: /* M2N notebook */ case 0x1821: /* M5N notebook */ + case 0x1897: /* A6L notebook */ asus_hides_smbus = 1; } else if (dev->device == PCI_DEVICE_ID_INTEL_82855PM_HB) @@ -1163,6 +1169,7 @@ static void __init asus_hides_smbus_hostbridge(struct pci_dev *dev) switch (dev->subsystem_device) { case 0x12bc: /* HP D330L */ case 0x12bd: /* HP D530 */ + case 0x006a: /* HP Compaq nx9500 */ asus_hides_smbus = 1; } else if (dev->device == PCI_DEVICE_ID_INTEL_82875_HB) @@ -2016,6 +2023,28 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5709S, quirk_brcm_570x_limit_vpd); +/* Originally in EDAC sources for i82875P: + * Intel tells BIOS developers to hide device 6 which + * configures the overflow device access containing + * the DRBs - this is where we expose device 6. + * http://www.x86-secret.com/articles/tweak/pat/patsecrets-2.htm + */ +static void __devinit quirk_unhide_mch_dev6(struct pci_dev *dev) +{ + u8 reg; + + if (pci_read_config_byte(dev, 0xF4, ®) == 0 && !(reg & 0x02)) { + dev_info(&dev->dev, "Enabling MCH 'Overflow' Device\n"); + pci_write_config_byte(dev, 0xF4, reg | 0x02); + } +} + +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82865_HB, + quirk_unhide_mch_dev6); +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82875_HB, + quirk_unhide_mch_dev6); + + #ifdef CONFIG_PCI_MSI /* Some chipsets do not support MSI. We cannot easily rely on setting * PCI_BUS_FLAGS_NO_MSI in its bus flags because there are actually @@ -2461,6 +2490,8 @@ static void __devinit quirk_i82576_sriov(struct pci_dev *dev) DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x10c9, quirk_i82576_sriov); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x10e6, quirk_i82576_sriov); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x10e7, quirk_i82576_sriov); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x10e8, quirk_i82576_sriov); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x150a, quirk_i82576_sriov); #endif /* CONFIG_PCI_IOV */ diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c index 86503c14ce7..176615e7231 100644 --- a/drivers/pci/remove.c +++ b/drivers/pci/remove.c @@ -32,8 +32,6 @@ static void pci_stop_dev(struct pci_dev *dev) static void pci_destroy_dev(struct pci_dev *dev) { - pci_stop_dev(dev); - /* Remove the device from the device lists, and prevent any further * list accesses from this device */ down_write(&pci_bus_sem); diff --git a/drivers/pci/search.c b/drivers/pci/search.c index 710d4ea6956..e8cb5051c31 100644 --- a/drivers/pci/search.c +++ b/drivers/pci/search.c @@ -29,7 +29,7 @@ pci_find_upstream_pcie_bridge(struct pci_dev *pdev) if (pdev->is_pcie) return NULL; while (1) { - if (!pdev->bus->parent) + if (pci_is_root_bus(pdev->bus)) break; pdev = pdev->bus->self; /* a p2p bridge */ @@ -115,36 +115,6 @@ pci_find_next_bus(const struct pci_bus *from) #ifdef CONFIG_PCI_LEGACY /** - * pci_find_slot - locate PCI device from a given PCI slot - * @bus: number of PCI bus on which desired PCI device resides - * @devfn: encodes number of PCI slot in which the desired PCI - * device resides and the logical device number within that slot - * in case of multi-function devices. - * - * Given a PCI bus and slot/function number, the desired PCI device - * is located in system global list of PCI devices. If the device - * is found, a pointer to its data structure is returned. If no - * device is found, %NULL is returned. - * - * NOTE: Do not use this function any more; use pci_get_slot() instead, as - * the PCI device returned by this function can disappear at any moment in - * time. - */ -struct pci_dev *pci_find_slot(unsigned int bus, unsigned int devfn) -{ - struct pci_dev *dev = NULL; - - while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { - if (dev->bus->number == bus && dev->devfn == devfn) { - pci_dev_put(dev); - return dev; - } - } - return NULL; -} -EXPORT_SYMBOL(pci_find_slot); - -/** * pci_find_device - begin or continue searching for a PCI device by vendor/device id * @vendor: PCI vendor id to match, or %PCI_ANY_ID to match all vendor ids * @device: PCI device id to match, or %PCI_ANY_ID to match all device ids diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c index a00f85471b6..b636e245445 100644 --- a/drivers/pci/setup-bus.c +++ b/drivers/pci/setup-bus.c @@ -58,7 +58,6 @@ static void pbus_assign_resources_sorted(const struct pci_bus *bus) res = list->res; idx = res - &list->dev->resource[0]; if (pci_assign_resource(list->dev, idx)) { - /* FIXME: get rid of this */ res->start = 0; res->end = 0; res->flags = 0; @@ -143,6 +142,7 @@ static void pci_setup_bridge(struct pci_bus *bus) struct pci_dev *bridge = bus->self; struct pci_bus_region region; u32 l, bu, lu, io_upper16; + int pref_mem64; if (pci_is_enabled(bridge)) return; @@ -198,16 +198,22 @@ static void pci_setup_bridge(struct pci_bus *bus) pci_write_config_dword(bridge, PCI_PREF_LIMIT_UPPER32, 0); /* Set up PREF base/limit. */ + pref_mem64 = 0; bu = lu = 0; pcibios_resource_to_bus(bridge, ®ion, bus->resource[2]); if (bus->resource[2]->flags & IORESOURCE_PREFETCH) { + int width = 8; l = (region.start >> 16) & 0xfff0; l |= region.end & 0xfff00000; - bu = upper_32_bits(region.start); - lu = upper_32_bits(region.end); - dev_info(&bridge->dev, " PREFETCH window: %#016llx-%#016llx\n", - (unsigned long long)region.start, - (unsigned long long)region.end); + if (bus->resource[2]->flags & IORESOURCE_MEM_64) { + pref_mem64 = 1; + bu = upper_32_bits(region.start); + lu = upper_32_bits(region.end); + width = 16; + } + dev_info(&bridge->dev, " PREFETCH window: %#0*llx-%#0*llx\n", + width, (unsigned long long)region.start, + width, (unsigned long long)region.end); } else { l = 0x0000fff0; @@ -215,9 +221,11 @@ static void pci_setup_bridge(struct pci_bus *bus) } pci_write_config_dword(bridge, PCI_PREF_MEMORY_BASE, l); - /* Set the upper 32 bits of PREF base & limit. */ - pci_write_config_dword(bridge, PCI_PREF_BASE_UPPER32, bu); - pci_write_config_dword(bridge, PCI_PREF_LIMIT_UPPER32, lu); + if (pref_mem64) { + /* Set the upper 32 bits of PREF base & limit. */ + pci_write_config_dword(bridge, PCI_PREF_BASE_UPPER32, bu); + pci_write_config_dword(bridge, PCI_PREF_LIMIT_UPPER32, lu); + } pci_write_config_word(bridge, PCI_BRIDGE_CONTROL, bus->bridge_ctl); } @@ -255,8 +263,25 @@ static void pci_bridge_check_ranges(struct pci_bus *bus) pci_read_config_dword(bridge, PCI_PREF_MEMORY_BASE, &pmem); pci_write_config_dword(bridge, PCI_PREF_MEMORY_BASE, 0x0); } - if (pmem) + if (pmem) { b_res[2].flags |= IORESOURCE_MEM | IORESOURCE_PREFETCH; + if ((pmem & PCI_PREF_RANGE_TYPE_MASK) == PCI_PREF_RANGE_TYPE_64) + b_res[2].flags |= IORESOURCE_MEM_64; + } + + /* double check if bridge does support 64 bit pref */ + if (b_res[2].flags & IORESOURCE_MEM_64) { + u32 mem_base_hi, tmp; + pci_read_config_dword(bridge, PCI_PREF_BASE_UPPER32, + &mem_base_hi); + pci_write_config_dword(bridge, PCI_PREF_BASE_UPPER32, + 0xffffffff); + pci_read_config_dword(bridge, PCI_PREF_BASE_UPPER32, &tmp); + if (!tmp) + b_res[2].flags &= ~IORESOURCE_MEM_64; + pci_write_config_dword(bridge, PCI_PREF_BASE_UPPER32, + mem_base_hi); + } } /* Helper function for sizing routines: find first available @@ -336,6 +361,7 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask, unsigned long resource_size_t aligns[12]; /* Alignments from 1Mb to 2Gb */ int order, max_order; struct resource *b_res = find_free_bus_resource(bus, type); + unsigned int mem64_mask = 0; if (!b_res) return 0; @@ -344,9 +370,12 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask, unsigned long max_order = 0; size = 0; + mem64_mask = b_res->flags & IORESOURCE_MEM_64; + b_res->flags &= ~IORESOURCE_MEM_64; + list_for_each_entry(dev, &bus->devices, bus_list) { int i; - + for (i = 0; i < PCI_NUM_RESOURCES; i++) { struct resource *r = &dev->resource[i]; resource_size_t r_size; @@ -372,6 +401,7 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask, unsigned long aligns[order] += align; if (order > max_order) max_order = order; + mem64_mask &= r->flags & IORESOURCE_MEM_64; } } @@ -396,6 +426,7 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask, unsigned long b_res->start = min_align; b_res->end = size + min_align - 1; b_res->flags |= IORESOURCE_STARTALIGN; + b_res->flags |= mem64_mask; return 1; } diff --git a/drivers/pci/setup-res.c b/drivers/pci/setup-res.c index 3039fcb86af..1898c7b4790 100644 --- a/drivers/pci/setup-res.c +++ b/drivers/pci/setup-res.c @@ -99,17 +99,17 @@ void pci_update_resource(struct pci_dev *dev, int resno) int pci_claim_resource(struct pci_dev *dev, int resource) { struct resource *res = &dev->resource[resource]; - struct resource *root = NULL; - char *dtype = resource < PCI_BRIDGE_RESOURCES ? "device" : "bridge"; + struct resource *root; int err; - root = pcibios_select_root(dev, res); + root = pci_find_parent_resource(dev, res); err = -EINVAL; if (root != NULL) - err = insert_resource(root, res); + err = request_resource(root, res); if (err) { + const char *dtype = resource < PCI_BRIDGE_RESOURCES ? "device" : "bridge"; dev_err(&dev->dev, "BAR %d: %s of %s %pR\n", resource, root ? "address space collision on" : @@ -135,23 +135,16 @@ void pci_disable_bridge_window(struct pci_dev *dev) } #endif /* CONFIG_PCI_QUIRKS */ -int pci_assign_resource(struct pci_dev *dev, int resno) +static int __pci_assign_resource(struct pci_bus *bus, struct pci_dev *dev, + int resno) { - struct pci_bus *bus = dev->bus; struct resource *res = dev->resource + resno; resource_size_t size, min, align; int ret; size = resource_size(res); min = (res->flags & IORESOURCE_IO) ? PCIBIOS_MIN_IO : PCIBIOS_MIN_MEM; - align = resource_alignment(res); - if (!align) { - dev_info(&dev->dev, "BAR %d: can't allocate resource (bogus " - "alignment) %pR flags %#lx\n", - resno, res, res->flags); - return -EINVAL; - } /* First, try exact prefetching match.. */ ret = pci_bus_alloc_resource(bus, res, size, align, min, @@ -169,10 +162,7 @@ int pci_assign_resource(struct pci_dev *dev, int resno) pcibios_align_resource, dev); } - if (ret) { - dev_info(&dev->dev, "BAR %d: can't allocate %s resource %pR\n", - resno, res->flags & IORESOURCE_IO ? "I/O" : "mem", res); - } else { + if (!ret) { res->flags &= ~IORESOURCE_STARTALIGN; if (resno < PCI_BRIDGE_RESOURCES) pci_update_resource(dev, resno); @@ -181,6 +171,39 @@ int pci_assign_resource(struct pci_dev *dev, int resno) return ret; } +int pci_assign_resource(struct pci_dev *dev, int resno) +{ + struct resource *res = dev->resource + resno; + resource_size_t align; + struct pci_bus *bus; + int ret; + + align = resource_alignment(res); + if (!align) { + dev_info(&dev->dev, "BAR %d: can't allocate resource (bogus " + "alignment) %pR flags %#lx\n", + resno, res, res->flags); + return -EINVAL; + } + + bus = dev->bus; + while ((ret = __pci_assign_resource(bus, dev, resno))) { + if (bus->parent && bus->self->transparent) + bus = bus->parent; + else + bus = NULL; + if (bus) + continue; + break; + } + + if (ret) + dev_info(&dev->dev, "BAR %d: can't allocate %s resource %pR\n", + resno, res->flags & IORESOURCE_IO ? "I/O" : "mem", res); + + return ret; +} + #if 0 int pci_assign_resource_fixed(struct pci_dev *dev, int resno) { diff --git a/drivers/pci/slot.c b/drivers/pci/slot.c index fe95ce20bcb..8c02b6c53bd 100644 --- a/drivers/pci/slot.c +++ b/drivers/pci/slot.c @@ -307,6 +307,45 @@ void pci_destroy_slot(struct pci_slot *slot) } EXPORT_SYMBOL_GPL(pci_destroy_slot); +#if defined(CONFIG_HOTPLUG_PCI) || defined(CONFIG_HOTPLUG_PCI_MODULE) +#include <linux/pci_hotplug.h> +/** + * pci_hp_create_link - create symbolic link to the hotplug driver module. + * @pci_slot: struct pci_slot + * + * Helper function for pci_hotplug_core.c to create symbolic link to + * the hotplug driver module. + */ +void pci_hp_create_module_link(struct pci_slot *pci_slot) +{ + struct hotplug_slot *slot = pci_slot->hotplug; + struct kobject *kobj = NULL; + int no_warn; + + if (!slot || !slot->ops) + return; + kobj = kset_find_obj(module_kset, slot->ops->mod_name); + if (!kobj) + return; + no_warn = sysfs_create_link(&pci_slot->kobj, kobj, "module"); + kobject_put(kobj); +} +EXPORT_SYMBOL_GPL(pci_hp_create_module_link); + +/** + * pci_hp_remove_link - remove symbolic link to the hotplug driver module. + * @pci_slot: struct pci_slot + * + * Helper function for pci_hotplug_core.c to remove symbolic link to + * the hotplug driver module. + */ +void pci_hp_remove_module_link(struct pci_slot *pci_slot) +{ + sysfs_remove_link(&pci_slot->kobj, "module"); +} +EXPORT_SYMBOL_GPL(pci_hp_remove_module_link); +#endif + static int pci_slot_init(void) { struct kset *pci_bus_kset; diff --git a/drivers/pci/syscall.c b/drivers/pci/syscall.c index ec22284eed3..e1c1ec54089 100644 --- a/drivers/pci/syscall.c +++ b/drivers/pci/syscall.c @@ -9,7 +9,6 @@ #include <linux/errno.h> #include <linux/pci.h> -#include <linux/smp_lock.h> #include <linux/syscalls.h> #include <asm/uaccess.h> #include "pci.h" |