diff options
Diffstat (limited to 'drivers/pci')
62 files changed, 6277 insertions, 3880 deletions
diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig index 2a4501dd251..fdc864f9cf2 100644 --- a/drivers/pci/Kconfig +++ b/drivers/pci/Kconfig @@ -59,3 +59,13 @@ config HT_IRQ This allows native hypertransport devices to use interrupts. If unsure say Y. + +config PCI_IOV + bool "PCI IOV support" + depends on PCI + help + I/O Virtualization is a PCI feature supported by some devices + which allows them to create virtual devices which share their + physical resources. + + If unsure, say N. diff --git a/drivers/pci/Makefile b/drivers/pci/Makefile index 3d07ce24f6a..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-$(CONFIG_PCIEPORTBUS) += pcie/ @@ -29,6 +30,8 @@ obj-$(CONFIG_DMAR) += dmar.o iova.o intel-iommu.o obj-$(CONFIG_INTR_REMAP) += dmar.o intr_remapping.o +obj-$(CONFIG_PCI_IOV) += iov.o + # # Some architectures use the generic PCI setup functions # @@ -37,7 +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 38144479477..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 @@ -87,8 +106,8 @@ EXPORT_SYMBOL(pci_read_vpd); * pci_write_vpd - Write entry to Vital Product Data * @dev: pci device struct * @pos: offset in vpd space - * @count: number of bytes to read - * @val: value to write + * @count: number of bytes to write + * @buf: buffer containing write data * */ ssize_t pci_write_vpd(struct pci_dev *dev, loff_t pos, size_t count, const void *buf) @@ -356,7 +375,8 @@ int pci_vpd_truncate(struct pci_dev *dev, size_t size) return -EINVAL; dev->vpd->len = size; - dev->vpd->attr->size = size; + if (dev->vpd->attr) + dev->vpd->attr->size = size; return 0; } diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c index 52b54f053be..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; @@ -133,7 +138,7 @@ int pci_bus_add_child(struct pci_bus *bus) * * Call hotplug for each new devices. */ -void pci_bus_add_devices(struct pci_bus *bus) +void pci_bus_add_devices(const struct pci_bus *bus) { struct pci_dev *dev; struct pci_bus *child; @@ -184,8 +189,10 @@ void pci_enable_bridges(struct pci_bus *bus) list_for_each_entry(dev, &bus->devices, bus_list) { if (dev->subordinate) { - retval = pci_enable_device(dev); - pci_set_master(dev); + if (!pci_is_enabled(dev)) { + retval = pci_enable_device(dev); + pci_set_master(dev); + } pci_enable_bridges(dev->subordinate); } } @@ -199,13 +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); @@ -229,8 +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 26c536b51c5..fa3a11365ec 100644 --- a/drivers/pci/dmar.c +++ b/drivers/pci/dmar.c @@ -31,6 +31,8 @@ #include <linux/iova.h> #include <linux/intel-iommu.h> #include <linux/timer.h> +#include <linux/irq.h> +#include <linux/interrupt.h> #undef PREFIX #define PREFIX "DMAR:" @@ -42,6 +44,7 @@ LIST_HEAD(dmar_drhd_units); static struct acpi_table_header * __initdata dmar_tbl; +static acpi_size dmar_tbl_size; static void __init dmar_register_drhd_unit(struct dmar_drhd_unit *drhd) { @@ -170,13 +173,23 @@ dmar_parse_one_drhd(struct acpi_dmar_header *header) struct dmar_drhd_unit *dmaru; int ret = 0; + drhd = (struct acpi_dmar_hardware_unit *)header; + if (!drhd->address) { + /* Promote an attitude of violence to a BIOS engineer today */ + WARN(1, "Your BIOS is broken; DMAR reported at address zero!\n" + "BIOS vendor: %s; Ver: %s; Product Version: %s\n", + dmi_get_system_info(DMI_BIOS_VENDOR), + dmi_get_system_info(DMI_BIOS_VERSION), + dmi_get_system_info(DMI_PRODUCT_VERSION)); + return -ENODEV; + } dmaru = kzalloc(sizeof(*dmaru), GFP_KERNEL); if (!dmaru) return -ENOMEM; dmaru->hdr = header; - drhd = (struct acpi_dmar_hardware_unit *)header; dmaru->reg_base_addr = drhd->address; + dmaru->segment = drhd->segment; dmaru->include_all = drhd->flags & 0x1; /* BIT0: INCLUDE_ALL */ ret = alloc_iommu(dmaru); @@ -288,8 +301,9 @@ static int __init dmar_table_detect(void) acpi_status status = AE_OK; /* if we could find DMAR table, then there are DMAR devices */ - status = acpi_get_table(ACPI_SIG_DMAR, 0, - (struct acpi_table_header **)&dmar_tbl); + status = acpi_get_table_with_size(ACPI_SIG_DMAR, 0, + (struct acpi_table_header **)&dmar_tbl, + &dmar_tbl_size); if (ACPI_SUCCESS(status) && !dmar_tbl) { printk (KERN_WARNING PREFIX "Unable to map DMAR\n"); @@ -489,6 +503,7 @@ void __init detect_intel_iommu(void) iommu_detected = 1; #endif } + early_acpi_os_unmap_memory(dmar_tbl, dmar_tbl_size); dmar_tbl = NULL; } @@ -506,6 +521,7 @@ int alloc_iommu(struct dmar_drhd_unit *drhd) return -ENOMEM; iommu->seq_id = iommu_allocated++; + sprintf (iommu->name, "dmar%d", iommu->seq_id); iommu->reg = ioremap(drhd->reg_base_addr, VTD_PAGE_SIZE); if (!iommu->reg) { @@ -748,14 +764,77 @@ int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr, } /* + * Disable Queued Invalidation interface. + */ +void dmar_disable_qi(struct intel_iommu *iommu) +{ + unsigned long flags; + u32 sts; + cycles_t start_time = get_cycles(); + + if (!ecap_qis(iommu->ecap)) + return; + + spin_lock_irqsave(&iommu->register_lock, flags); + + sts = dmar_readq(iommu->reg + DMAR_GSTS_REG); + if (!(sts & DMA_GSTS_QIES)) + goto end; + + /* + * Give a chance to HW to complete the pending invalidation requests. + */ + while ((readl(iommu->reg + DMAR_IQT_REG) != + readl(iommu->reg + DMAR_IQH_REG)) && + (DMAR_OPERATION_TIMEOUT > (get_cycles() - start_time))) + cpu_relax(); + + iommu->gcmd &= ~DMA_GCMD_QIE; + + writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG); + + IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, readl, + !(sts & DMA_GSTS_QIES), sts); +end: + spin_unlock_irqrestore(&iommu->register_lock, flags); +} + +/* + * Enable queued invalidation. + */ +static void __dmar_enable_qi(struct intel_iommu *iommu) +{ + u32 cmd, sts; + unsigned long flags; + struct q_inval *qi = iommu->qi; + + qi->free_head = qi->free_tail = 0; + qi->free_cnt = QI_LENGTH; + + spin_lock_irqsave(&iommu->register_lock, flags); + + /* write zero to the tail reg */ + writel(0, iommu->reg + DMAR_IQT_REG); + + 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); + + /* Make sure hardware complete it */ + IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, readl, (sts & DMA_GSTS_QIES), sts); + + spin_unlock_irqrestore(&iommu->register_lock, flags); +} + +/* * Enable Queued Invalidation interface. This is a must to support * interrupt-remapping. Also used by DMA-remapping, which replaces * register based IOTLB invalidation. */ int dmar_enable_qi(struct intel_iommu *iommu) { - u32 cmd, sts; - unsigned long flags; struct q_inval *qi; if (!ecap_qis(iommu->ecap)) @@ -767,20 +846,20 @@ int dmar_enable_qi(struct intel_iommu *iommu) if (iommu->qi) return 0; - iommu->qi = kmalloc(sizeof(*qi), GFP_KERNEL); + iommu->qi = kmalloc(sizeof(*qi), GFP_ATOMIC); if (!iommu->qi) return -ENOMEM; qi = iommu->qi; - qi->desc = (void *)(get_zeroed_page(GFP_KERNEL)); + qi->desc = (void *)(get_zeroed_page(GFP_ATOMIC)); if (!qi->desc) { kfree(qi); iommu->qi = 0; return -ENOMEM; } - qi->desc_status = kmalloc(QI_LENGTH * sizeof(int), GFP_KERNEL); + qi->desc_status = kmalloc(QI_LENGTH * sizeof(int), GFP_ATOMIC); if (!qi->desc_status) { free_page((unsigned long) qi->desc); kfree(qi); @@ -793,19 +872,283 @@ int dmar_enable_qi(struct intel_iommu *iommu) spin_lock_init(&qi->q_lock); - spin_lock_irqsave(&iommu->register_lock, flags); - /* write zero to the tail reg */ - writel(0, iommu->reg + DMAR_IQT_REG); + __dmar_enable_qi(iommu); - dmar_writeq(iommu->reg + DMAR_IQA_REG, virt_to_phys(qi->desc)); + return 0; +} - cmd = iommu->gcmd | DMA_GCMD_QIE; - iommu->gcmd |= DMA_GCMD_QIE; - writel(cmd, iommu->reg + DMAR_GCMD_REG); +/* iommu interrupt handling. Most stuff are MSI-like. */ - /* Make sure hardware complete it */ - IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, readl, (sts & DMA_GSTS_QIES), sts); - spin_unlock_irqrestore(&iommu->register_lock, flags); +enum faulttype { + DMA_REMAP, + INTR_REMAP, + UNKNOWN, +}; + +static const char *dma_remap_fault_reasons[] = +{ + "Software", + "Present bit in root entry is clear", + "Present bit in context entry is clear", + "Invalid context entry", + "Access beyond MGAW", + "PTE Write access is not set", + "PTE Read access is not set", + "Next page table ptr is invalid", + "Root table address invalid", + "Context table ptr is invalid", + "non-zero reserved fields in RTP", + "non-zero reserved fields in CTP", + "non-zero reserved fields in PTE", +}; + +static const char *intr_remap_fault_reasons[] = +{ + "Detected reserved fields in the decoded interrupt-remapped request", + "Interrupt index exceeded the interrupt-remapping table size", + "Present field in the IRTE entry is clear", + "Error accessing interrupt-remapping table pointed by IRTA_REG", + "Detected reserved fields in the IRTE entry", + "Blocked a compatibility format interrupt request", + "Blocked an interrupt request due to source-id verification failure", +}; + +#define MAX_FAULT_REASON_IDX (ARRAY_SIZE(fault_reason_strings) - 1) + +const char *dmar_get_fault_reason(u8 fault_reason, int *fault_type) +{ + if (fault_reason >= 0x20 && (fault_reason <= 0x20 + + ARRAY_SIZE(intr_remap_fault_reasons))) { + *fault_type = INTR_REMAP; + return intr_remap_fault_reasons[fault_reason - 0x20]; + } else if (fault_reason < ARRAY_SIZE(dma_remap_fault_reasons)) { + *fault_type = DMA_REMAP; + return dma_remap_fault_reasons[fault_reason]; + } else { + *fault_type = UNKNOWN; + return "Unknown"; + } +} + +void dmar_msi_unmask(unsigned int irq) +{ + struct intel_iommu *iommu = get_irq_data(irq); + unsigned long flag; + + /* unmask it */ + spin_lock_irqsave(&iommu->register_lock, flag); + writel(0, iommu->reg + DMAR_FECTL_REG); + /* Read a reg to force flush the post write */ + readl(iommu->reg + DMAR_FECTL_REG); + spin_unlock_irqrestore(&iommu->register_lock, flag); +} + +void dmar_msi_mask(unsigned int irq) +{ + unsigned long flag; + struct intel_iommu *iommu = get_irq_data(irq); + + /* mask it */ + spin_lock_irqsave(&iommu->register_lock, flag); + writel(DMA_FECTL_IM, iommu->reg + DMAR_FECTL_REG); + /* Read a reg to force flush the post write */ + readl(iommu->reg + DMAR_FECTL_REG); + spin_unlock_irqrestore(&iommu->register_lock, flag); +} + +void dmar_msi_write(int irq, struct msi_msg *msg) +{ + struct intel_iommu *iommu = get_irq_data(irq); + unsigned long flag; + + spin_lock_irqsave(&iommu->register_lock, flag); + writel(msg->data, iommu->reg + DMAR_FEDATA_REG); + writel(msg->address_lo, iommu->reg + DMAR_FEADDR_REG); + writel(msg->address_hi, iommu->reg + DMAR_FEUADDR_REG); + spin_unlock_irqrestore(&iommu->register_lock, flag); +} + +void dmar_msi_read(int irq, struct msi_msg *msg) +{ + struct intel_iommu *iommu = get_irq_data(irq); + unsigned long flag; + + spin_lock_irqsave(&iommu->register_lock, flag); + msg->data = readl(iommu->reg + DMAR_FEDATA_REG); + msg->address_lo = readl(iommu->reg + DMAR_FEADDR_REG); + msg->address_hi = readl(iommu->reg + DMAR_FEUADDR_REG); + spin_unlock_irqrestore(&iommu->register_lock, flag); +} + +static int dmar_fault_do_one(struct intel_iommu *iommu, int type, + u8 fault_reason, u16 source_id, unsigned long long addr) +{ + const char *reason; + int fault_type; + + reason = dmar_get_fault_reason(fault_reason, &fault_type); + + if (fault_type == INTR_REMAP) + printk(KERN_ERR "INTR-REMAP: Request device [[%02x:%02x.%d] " + "fault index %llx\n" + "INTR-REMAP:[fault reason %02d] %s\n", + (source_id >> 8), PCI_SLOT(source_id & 0xFF), + PCI_FUNC(source_id & 0xFF), addr >> 48, + fault_reason, reason); + else + printk(KERN_ERR + "DMAR:[%s] Request device [%02x:%02x.%d] " + "fault addr %llx \n" + "DMAR:[fault reason %02d] %s\n", + (type ? "DMA Read" : "DMA Write"), + (source_id >> 8), PCI_SLOT(source_id & 0xFF), + PCI_FUNC(source_id & 0xFF), addr, fault_reason, reason); + return 0; +} + +#define PRIMARY_FAULT_REG_LEN (16) +irqreturn_t dmar_fault(int irq, void *dev_id) +{ + struct intel_iommu *iommu = dev_id; + int reg, fault_index; + u32 fault_status; + unsigned long flag; + + spin_lock_irqsave(&iommu->register_lock, flag); + fault_status = readl(iommu->reg + DMAR_FSTS_REG); + if (fault_status) + printk(KERN_ERR "DRHD: handling fault status reg %x\n", + fault_status); + + /* TBD: ignore advanced fault log currently */ + if (!(fault_status & DMA_FSTS_PPF)) + goto clear_rest; + + fault_index = dma_fsts_fault_record_index(fault_status); + reg = cap_fault_reg_offset(iommu->cap); + while (1) { + u8 fault_reason; + u16 source_id; + u64 guest_addr; + int type; + u32 data; + + /* highest 32 bits */ + data = readl(iommu->reg + reg + + fault_index * PRIMARY_FAULT_REG_LEN + 12); + if (!(data & DMA_FRCD_F)) + break; + + fault_reason = dma_frcd_fault_reason(data); + type = dma_frcd_type(data); + + data = readl(iommu->reg + reg + + fault_index * PRIMARY_FAULT_REG_LEN + 8); + source_id = dma_frcd_source_id(data); + + guest_addr = dmar_readq(iommu->reg + reg + + fault_index * PRIMARY_FAULT_REG_LEN); + guest_addr = dma_frcd_page_addr(guest_addr); + /* clear the fault */ + writel(DMA_FRCD_F, iommu->reg + reg + + fault_index * PRIMARY_FAULT_REG_LEN + 12); + + spin_unlock_irqrestore(&iommu->register_lock, flag); + + dmar_fault_do_one(iommu, type, fault_reason, + source_id, guest_addr); + + fault_index++; + if (fault_index > cap_num_fault_regs(iommu->cap)) + fault_index = 0; + spin_lock_irqsave(&iommu->register_lock, flag); + } +clear_rest: + /* clear all the other faults */ + fault_status = readl(iommu->reg + DMAR_FSTS_REG); + writel(fault_status, iommu->reg + DMAR_FSTS_REG); + + spin_unlock_irqrestore(&iommu->register_lock, flag); + return IRQ_HANDLED; +} + +int dmar_set_interrupt(struct intel_iommu *iommu) +{ + int irq, ret; + + /* + * Check if the fault interrupt is already initialized. + */ + if (iommu->irq) + return 0; + + irq = create_irq(); + if (!irq) { + printk(KERN_ERR "IOMMU: no free vectors\n"); + return -EINVAL; + } + + set_irq_data(irq, iommu); + iommu->irq = irq; + + ret = arch_setup_dmar_msi(irq); + if (ret) { + set_irq_data(irq, NULL); + iommu->irq = 0; + destroy_irq(irq); + return 0; + } + + ret = request_irq(irq, dmar_fault, 0, iommu->name, iommu); + if (ret) + printk(KERN_ERR "IOMMU: can't request irq\n"); + return ret; +} + +int __init enable_drhd_fault_handling(void) +{ + struct dmar_drhd_unit *drhd; + + /* + * Enable fault control interrupt. + */ + for_each_drhd_unit(drhd) { + int ret; + struct intel_iommu *iommu = drhd->iommu; + ret = dmar_set_interrupt(iommu); + + if (ret) { + printk(KERN_ERR "DRHD %Lx: failed to enable fault, " + " interrupt, ret %d\n", + (unsigned long long)drhd->reg_base_addr, ret); + return -1; + } + } + + return 0; +} + +/* + * Re-enable Queued Invalidation interface. + */ +int dmar_reenable_qi(struct intel_iommu *iommu) +{ + if (!ecap_qis(iommu->ecap)) + return -ENOENT; + + if (!iommu->qi) + return -ENOENT; + + /* + * First disable queued invalidation. + */ + dmar_disable_qi(iommu); + /* + * Then enable queued invalidation again. Since there is no pending + * invalidation requests now, it's safe to re-enable queued + * invalidation. + */ + __dmar_enable_qi(iommu); return 0; } 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 1c114180106..fbc63d5e459 100644 --- a/drivers/pci/hotplug/acpi_pcihp.c +++ b/drivers/pci/hotplug/acpi_pcihp.c @@ -30,9 +30,8 @@ #include <linux/types.h> #include <linux/pci.h> #include <linux/pci_hotplug.h> +#include <linux/acpi.h> #include <linux/pci-acpi.h> -#include <acpi/acpi.h> -#include <acpi/acpi_bus.h> #define MY_NAME "acpi_pcihp" @@ -333,19 +332,14 @@ acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus, { acpi_status status = AE_NOT_FOUND; acpi_handle handle, phandle; - struct pci_bus *pbus = bus; - struct pci_dev *pdev; - - do { - pdev = pbus->self; - if (!pdev) { - handle = acpi_get_pci_rootbridge_handle( - pci_domain_nr(pbus), pbus->number); + struct pci_bus *pbus; + + handle = NULL; + for (pbus = bus; pbus; pbus = pbus->parent) { + handle = acpi_pci_get_bridge_handle(pbus); + if (handle) break; - } - handle = DEVICE_ACPI_HANDLE(&(pdev->dev)); - pbus = pbus->parent; - } while (!handle); + } /* * _HPP settings apply to all child buses, until another _HPP is @@ -378,12 +372,10 @@ EXPORT_SYMBOL_GPL(acpi_get_hp_params_from_firmware); * * Attempt to take hotplug control from firmware. */ -int acpi_get_hp_hw_control_from_firmware(struct pci_dev *dev, u32 flags) +int acpi_get_hp_hw_control_from_firmware(struct pci_dev *pdev, u32 flags) { acpi_status status; acpi_handle chandle, handle; - struct pci_dev *pdev = dev; - struct pci_bus *parent; struct acpi_buffer string = { ACPI_ALLOCATE_BUFFER, NULL }; flags &= (OSC_PCI_EXPRESS_NATIVE_HP_CONTROL | @@ -408,33 +400,25 @@ int acpi_get_hp_hw_control_from_firmware(struct pci_dev *dev, u32 flags) acpi_get_name(handle, ACPI_FULL_PATHNAME, &string); dbg("Trying to get hotplug control for %s\n", (char *)string.pointer); - status = pci_osc_control_set(handle, flags); + status = acpi_pci_osc_control_set(handle, flags); if (ACPI_SUCCESS(status)) goto got_one; kfree(string.pointer); string = (struct acpi_buffer){ ACPI_ALLOCATE_BUFFER, NULL }; } - pdev = dev; - handle = DEVICE_ACPI_HANDLE(&dev->dev); - while (!handle) { + handle = DEVICE_ACPI_HANDLE(&pdev->dev); + if (!handle) { /* * This hotplug controller was not listed in the ACPI name * space at all. Try to get acpi handle of parent pci bus. */ - if (!pdev || !pdev->bus->parent) - break; - parent = pdev->bus->parent; - dbg("Could not find %s in acpi namespace, trying parent\n", - pci_name(pdev)); - if (!parent->self) - /* Parent must be a host bridge */ - handle = acpi_get_pci_rootbridge_handle( - pci_domain_nr(parent), - parent->number); - else - handle = DEVICE_ACPI_HANDLE(&(parent->self->dev)); - pdev = parent->self; + struct pci_bus *pbus; + for (pbus = pdev->bus; pbus; pbus = pbus->parent) { + handle = acpi_pci_get_bridge_handle(pbus); + if (handle) + break; + } } while (handle) { @@ -453,13 +437,13 @@ int acpi_get_hp_hw_control_from_firmware(struct pci_dev *dev, u32 flags) } dbg("Cannot get control of hotplug hardware for pci %s\n", - pci_name(dev)); + pci_name(pdev)); kfree(string.pointer); return -ENODEV; got_one: - dbg("Gained control for hotplug HW for pci %s (%s)\n", pci_name(dev), - (char *)string.pointer); + dbg("Gained control for hotplug HW for pci %s (%s)\n", + pci_name(pdev), (char *)string.pointer); kfree(string.pointer); return 0; } diff --git a/drivers/pci/hotplug/acpiphp.h b/drivers/pci/hotplug/acpiphp.h index 4fc168b7009..e68d5f20ffb 100644 --- a/drivers/pci/hotplug/acpiphp.h +++ b/drivers/pci/hotplug/acpiphp.h @@ -129,7 +129,6 @@ struct acpiphp_func { struct acpiphp_bridge *bridge; /* Ejectable PCI-to-PCI bridge */ struct list_head sibling; - struct pci_dev *pci_dev; struct notifier_block nb; acpi_handle handle; 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 803d9ddd6e7..3a6064bce56 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -32,12 +32,11 @@ /* * Lifetime rules for pci_dev: - * - The one in acpiphp_func has its refcount elevated by pci_get_slot() - * when the driver is loaded or when an insertion event occurs. It loses - * a refcount when its ejected or the driver unloads. * - The one in acpiphp_bridge has its refcount elevated by pci_get_slot() * when the bridge is scanned and it loses a refcount when the bridge * is removed. + * - When a P2P bridge is present, we elevate the refcount on the subordinate + * bus. It loses the refcount when the the driver unloads. */ #include <linux/init.h> @@ -128,6 +127,7 @@ register_slot(acpi_handle handle, u32 lvl, void *context, void **rv) unsigned long long adr, sun; int device, function, retval; struct pci_bus *pbus = bridge->pci_bus; + struct pci_dev *pdev; if (!acpi_pci_check_ejectable(pbus, handle) && !is_dock_device(handle)) return AE_OK; @@ -211,10 +211,10 @@ register_slot(acpi_handle handle, u32 lvl, void *context, void **rv) newfunc->slot = slot; list_add_tail(&newfunc->sibling, &slot->funcs); - /* associate corresponding pci_dev */ - newfunc->pci_dev = pci_get_slot(pbus, PCI_DEVFN(device, function)); - if (newfunc->pci_dev) { + pdev = pci_get_slot(pbus, PCI_DEVFN(device, function)); + if (pdev) { slot->flags |= (SLOT_ENABLED | SLOT_POWEREDON); + pci_dev_put(pdev); } if (is_dock_device(handle)) { @@ -440,6 +440,12 @@ static void add_p2p_bridge(acpi_handle *handle, struct pci_dev *pci_dev) goto err; } + /* + * Grab a ref to the subordinate PCI bus in case the bus is + * removed via PCI core logical hotplug. The ref pins the bus + * (which we access during module unload). + */ + get_device(&bridge->pci_bus->dev); spin_lock_init(&bridge->res_lock); init_bridge_misc(bridge); @@ -609,7 +615,6 @@ static void cleanup_bridge(struct acpiphp_bridge *bridge) if (ACPI_FAILURE(status)) err("failed to remove notify handler\n"); } - pci_dev_put(func->pci_dev); list_del(list); kfree(func); } @@ -619,6 +624,12 @@ static void cleanup_bridge(struct acpiphp_bridge *bridge) slot = next; } + /* + * Only P2P bridges have a pci_dev + */ + if (bridge->pci_dev) + put_device(&bridge->pci_bus->dev); + pci_dev_put(bridge->pci_dev); list_del(&bridge->list); kfree(bridge); @@ -1087,22 +1098,24 @@ static int __ref enable_device(struct acpiphp_slot *slot) pci_enable_bridges(bus); pci_bus_add_devices(bus); - /* associate pci_dev to our representation */ list_for_each (l, &slot->funcs) { func = list_entry(l, struct acpiphp_func, sibling); - func->pci_dev = pci_get_slot(bus, PCI_DEVFN(slot->device, - func->function)); - if (!func->pci_dev) + dev = pci_get_slot(bus, PCI_DEVFN(slot->device, + func->function)); + if (!dev) continue; - if (func->pci_dev->hdr_type != PCI_HEADER_TYPE_BRIDGE && - func->pci_dev->hdr_type != PCI_HEADER_TYPE_CARDBUS) + if (dev->hdr_type != PCI_HEADER_TYPE_BRIDGE && + dev->hdr_type != PCI_HEADER_TYPE_CARDBUS) { + pci_dev_put(dev); continue; + } status = find_p2p_bridge(func->handle, (u32)1, bus, NULL); if (ACPI_FAILURE(status)) warn("find_p2p_bridge failed (error code = 0x%x)\n", status); + pci_dev_put(dev); } slot->flags |= SLOT_ENABLED; @@ -1128,17 +1141,14 @@ static void disable_bridges(struct pci_bus *bus) */ static int disable_device(struct acpiphp_slot *slot) { - int retval = 0; struct acpiphp_func *func; - struct list_head *l; + struct pci_dev *pdev; /* is this slot already disabled? */ if (!(slot->flags & SLOT_ENABLED)) goto err_exit; - list_for_each (l, &slot->funcs) { - func = list_entry(l, struct acpiphp_func, sibling); - + list_for_each_entry(func, &slot->funcs, sibling) { if (func->bridge) { /* cleanup p2p bridges under this P2P bridge */ cleanup_p2p_bridge(func->bridge->handle, @@ -1146,35 +1156,28 @@ static int disable_device(struct acpiphp_slot *slot) func->bridge = NULL; } - if (func->pci_dev) { - pci_stop_bus_device(func->pci_dev); - if (func->pci_dev->subordinate) { - disable_bridges(func->pci_dev->subordinate); - pci_disable_device(func->pci_dev); + pdev = pci_get_slot(slot->bridge->pci_bus, + PCI_DEVFN(slot->device, func->function)); + if (pdev) { + pci_stop_bus_device(pdev); + if (pdev->subordinate) { + disable_bridges(pdev->subordinate); + pci_disable_device(pdev); } + pci_remove_bus_device(pdev); + pci_dev_put(pdev); } } - list_for_each (l, &slot->funcs) { - func = list_entry(l, struct acpiphp_func, sibling); - + list_for_each_entry(func, &slot->funcs, sibling) { acpiphp_unconfigure_ioapics(func->handle); acpiphp_bus_trim(func->handle); - /* try to remove anyway. - * acpiphp_bus_add might have been failed */ - - if (!func->pci_dev) - continue; - - pci_remove_bus_device(func->pci_dev); - pci_dev_put(func->pci_dev); - func->pci_dev = NULL; } slot->flags &= (~SLOT_ENABLED); - err_exit: - return retval; +err_exit: + return 0; } diff --git a/drivers/pci/hotplug/cpci_hotplug_core.c b/drivers/pci/hotplug/cpci_hotplug_core.c index de94f4feef8..a5b9f6ae507 100644 --- a/drivers/pci/hotplug/cpci_hotplug_core.c +++ b/drivers/pci/hotplug/cpci_hotplug_core.c @@ -72,7 +72,6 @@ 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..2fa47af992a 100644 --- a/drivers/pci/hotplug/cpqphp_ctrl.c +++ b/drivers/pci/hotplug/cpqphp_ctrl.c @@ -81,14 +81,15 @@ 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 +101,17 @@ 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 +132,8 @@ 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 +152,17 @@ 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 +177,23 @@ 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 +211,8 @@ 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 +243,17 @@ 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 +264,16 @@ 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 +434,15 @@ 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 +548,10 @@ 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 +560,8 @@ 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 +586,8 @@ 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 +607,8 @@ 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 +649,16 @@ 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 +681,8 @@ 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 +754,8 @@ 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 +780,8 @@ 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 +898,19 @@ 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 +973,8 @@ 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 +1005,8 @@ 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 +1015,8 @@ 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 +1046,8 @@ 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 +1135,43 @@ 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 +1180,22 @@ 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 +1212,48 @@ 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 +1277,17 @@ 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 +1308,7 @@ 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 +1361,8 @@ 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 +1387,8 @@ 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 +1445,8 @@ 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 +1455,12 @@ 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 +1473,7 @@ 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 +1533,7 @@ 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 +1582,8 @@ 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 +1870,12 @@ 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 +1969,7 @@ 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 +2090,7 @@ 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 +2124,8 @@ 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 +2193,67 @@ 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 +2322,9 @@ 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 +2349,8 @@ 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 +2403,7 @@ 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 +2494,8 @@ 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 +2567,8 @@ 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 +2929,26 @@ 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 a13abf55d78..8450f4a6568 100644 --- a/drivers/pci/hotplug/cpqphp_sysfs.c +++ b/drivers/pci/hotplug/cpqphp_sysfs.c @@ -225,7 +225,8 @@ void cpqhp_shutdown_debugfs(void) void cpqhp_create_debugfs_files(struct controller *ctrl) { - ctrl->dentry = debugfs_create_file(ctrl->pci_dev->dev.bus_id, S_IRUGO, root, ctrl, &debug_ops); + ctrl->dentry = debugfs_create_file(dev_name(&ctrl->pci_dev->dev), + S_IRUGO, root, ctrl, &debug_ops); } void cpqhp_remove_debugfs_files(struct controller *ctrl) diff --git a/drivers/pci/hotplug/fakephp.c b/drivers/pci/hotplug/fakephp.c index d8649e12729..6151389fd90 100644 --- a/drivers/pci/hotplug/fakephp.c +++ b/drivers/pci/hotplug/fakephp.c @@ -1,395 +1,163 @@ -/* - * Fake PCI Hot Plug Controller Driver +/* Works like the fakephp driver used to, except a little better. * - * Copyright (C) 2003 Greg Kroah-Hartman <greg@kroah.com> - * Copyright (C) 2003 IBM Corp. - * Copyright (C) 2003 Rolf Eike Beer <eike-kernel@sf-tec.de> + * - It's possible to remove devices with subordinate busses. + * - New PCI devices that appear via any method, not just a fakephp triggered + * rescan, will be noticed. + * - Devices that are removed via any method, not just a fakephp triggered + * removal, will also be noticed. * - * Based on ideas and code from: - * Vladimir Kondratiev <vladimir.kondratiev@intel.com> - * Rolf Eike Beer <eike-kernel@sf-tec.de> + * Uses nothing from the pci-hotplug subsystem. * - * All rights reserved. - * - * 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. - * - * Send feedback to <greg@kroah.com> */ -/* - * - * This driver will "emulate" removing PCI devices from the system. If - * the "power" file is written to with "0" then the specified PCI device - * will be completely removed from the kernel. - * - * WARNING, this does NOT turn off the power to the PCI device. This is - * a "logical" removal, not a physical or electrical removal. - * - * Use this module at your own risk, you have been warned! - * - * Enabling PCI devices is left as an exercise for the reader... - * - */ -#include <linux/kernel.h> #include <linux/module.h> -#include <linux/pci.h> -#include <linux/pci_hotplug.h> +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/list.h> +#include <linux/kobject.h> +#include <linux/sysfs.h> #include <linux/init.h> -#include <linux/string.h> -#include <linux/slab.h> -#include <linux/workqueue.h> +#include <linux/pci.h> +#include <linux/device.h> #include "../pci.h" -#if !defined(MODULE) - #define MY_NAME "fakephp" -#else - #define MY_NAME THIS_MODULE->name -#endif - -#define dbg(format, arg...) \ - do { \ - if (debug) \ - printk(KERN_DEBUG "%s: " format, \ - MY_NAME , ## arg); \ - } while (0) -#define err(format, arg...) printk(KERN_ERR "%s: " format, MY_NAME , ## arg) -#define info(format, arg...) printk(KERN_INFO "%s: " format, MY_NAME , ## arg) - -#define DRIVER_AUTHOR "Greg Kroah-Hartman <greg@kroah.com>" -#define DRIVER_DESC "Fake PCI Hot Plug Controller Driver" - -struct dummy_slot { - struct list_head node; - struct hotplug_slot *slot; - struct pci_dev *dev; - struct work_struct remove_work; - unsigned long removed; +struct legacy_slot { + struct kobject kobj; + struct pci_dev *dev; + struct list_head list; }; -static int debug; -static int dup_slots; -static LIST_HEAD(slot_list); -static struct workqueue_struct *dummyphp_wq; - -static void pci_rescan_worker(struct work_struct *work); -static DECLARE_WORK(pci_rescan_work, pci_rescan_worker); - -static int enable_slot (struct hotplug_slot *slot); -static int disable_slot (struct hotplug_slot *slot); +static LIST_HEAD(legacy_list); -static struct hotplug_slot_ops dummy_hotplug_slot_ops = { - .owner = THIS_MODULE, - .enable_slot = enable_slot, - .disable_slot = disable_slot, -}; - -static void dummy_release(struct hotplug_slot *slot) +static ssize_t legacy_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct dummy_slot *dslot = slot->private; - - list_del(&dslot->node); - kfree(dslot->slot->info); - kfree(dslot->slot); - pci_dev_put(dslot->dev); - kfree(dslot); + struct legacy_slot *slot = container_of(kobj, typeof(*slot), kobj); + strcpy(buf, "1\n"); + return 2; } -#define SLOT_NAME_SIZE 8 - -static int add_slot(struct pci_dev *dev) +static void remove_callback(void *data) { - struct dummy_slot *dslot; - struct hotplug_slot *slot; - char name[SLOT_NAME_SIZE]; - int retval = -ENOMEM; - static int count = 1; - - slot = kzalloc(sizeof(struct hotplug_slot), GFP_KERNEL); - if (!slot) - goto error; - - slot->info = kzalloc(sizeof(struct hotplug_slot_info), GFP_KERNEL); - if (!slot->info) - goto error_slot; - - slot->info->power_status = 1; - slot->info->max_bus_speed = PCI_SPEED_UNKNOWN; - slot->info->cur_bus_speed = PCI_SPEED_UNKNOWN; - - dslot = kzalloc(sizeof(struct dummy_slot), GFP_KERNEL); - if (!dslot) - goto error_info; - - if (dup_slots) - snprintf(name, SLOT_NAME_SIZE, "fake"); - else - snprintf(name, SLOT_NAME_SIZE, "fake%d", count++); - dbg("slot->name = %s\n", name); - slot->ops = &dummy_hotplug_slot_ops; - slot->release = &dummy_release; - slot->private = dslot; - - retval = pci_hp_register(slot, dev->bus, PCI_SLOT(dev->devfn), name); - if (retval) { - err("pci_hp_register failed with error %d\n", retval); - goto error_dslot; - } - - dbg("slot->name = %s\n", hotplug_slot_name(slot)); - dslot->slot = slot; - dslot->dev = pci_dev_get(dev); - list_add (&dslot->node, &slot_list); - return retval; - -error_dslot: - kfree(dslot); -error_info: - kfree(slot->info); -error_slot: - kfree(slot); -error: - return retval; + pci_remove_bus_device((struct pci_dev *)data); } -static int __init pci_scan_buses(void) +static ssize_t legacy_store(struct kobject *kobj, struct attribute *attr, + const char *buf, size_t len) { - struct pci_dev *dev = NULL; - int lastslot = 0; + struct legacy_slot *slot = container_of(kobj, typeof(*slot), kobj); + unsigned long val; - while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) { - if (PCI_FUNC(dev->devfn) > 0 && - lastslot == PCI_SLOT(dev->devfn)) - continue; - lastslot = PCI_SLOT(dev->devfn); - add_slot(dev); - } + if (strict_strtoul(buf, 0, &val) < 0) + return -EINVAL; - return 0; + if (val) + pci_rescan_bus(slot->dev->bus); + else + sysfs_schedule_callback(&slot->dev->dev.kobj, remove_callback, + slot->dev, THIS_MODULE); + return len; } -static void remove_slot(struct dummy_slot *dslot) -{ - int retval; - - dbg("removing slot %s\n", hotplug_slot_name(dslot->slot)); - retval = pci_hp_deregister(dslot->slot); - if (retval) - err("Problem unregistering a slot %s\n", - hotplug_slot_name(dslot->slot)); -} +static struct attribute *legacy_attrs[] = { + &(struct attribute){ .name = "power", .mode = 0644 }, + NULL, +}; -/* called from the single-threaded workqueue handler to remove a slot */ -static void remove_slot_worker(struct work_struct *work) +static void legacy_release(struct kobject *kobj) { - struct dummy_slot *dslot = - container_of(work, struct dummy_slot, remove_work); - remove_slot(dslot); -} + struct legacy_slot *slot = container_of(kobj, typeof(*slot), kobj); -/** - * pci_rescan_slot - Rescan slot - * @temp: Device template. Should be set: bus and devfn. - * - * Tries hard not to re-enable already existing devices; - * also handles scanning of subfunctions. - */ -static int pci_rescan_slot(struct pci_dev *temp) -{ - struct pci_bus *bus = temp->bus; - struct pci_dev *dev; - int func; - u8 hdr_type; - int count = 0; - - if (!pci_read_config_byte(temp, PCI_HEADER_TYPE, &hdr_type)) { - temp->hdr_type = hdr_type & 0x7f; - if ((dev = pci_get_slot(bus, temp->devfn)) != NULL) - pci_dev_put(dev); - else { - dev = pci_scan_single_device(bus, temp->devfn); - if (dev) { - dbg("New device on %s function %x:%x\n", - bus->name, temp->devfn >> 3, - temp->devfn & 7); - count++; - } - } - /* multifunction device? */ - if (!(hdr_type & 0x80)) - return count; - - /* continue scanning for other functions */ - for (func = 1, temp->devfn++; func < 8; func++, temp->devfn++) { - if (pci_read_config_byte(temp, PCI_HEADER_TYPE, &hdr_type)) - continue; - temp->hdr_type = hdr_type & 0x7f; - - if ((dev = pci_get_slot(bus, temp->devfn)) != NULL) - pci_dev_put(dev); - else { - dev = pci_scan_single_device(bus, temp->devfn); - if (dev) { - dbg("New device on %s function %x:%x\n", - bus->name, temp->devfn >> 3, - temp->devfn & 7); - count++; - } - } - } - } - - return count; + pci_dev_put(slot->dev); + kfree(slot); } +static struct kobj_type legacy_ktype = { + .sysfs_ops = &(struct sysfs_ops){ + .store = legacy_store, .show = legacy_show + }, + .release = &legacy_release, + .default_attrs = legacy_attrs, +}; -/** - * pci_rescan_bus - Rescan PCI bus - * @bus: the PCI bus to rescan - * - * Call pci_rescan_slot for each possible function of the bus. - */ -static void pci_rescan_bus(const struct pci_bus *bus) +static int legacy_add_slot(struct pci_dev *pdev) { - unsigned int devfn; - struct pci_dev *dev; - int retval; - int found = 0; - dev = alloc_pci_dev(); - if (!dev) - return; + struct legacy_slot *slot = kzalloc(sizeof(*slot), GFP_KERNEL); - dev->bus = (struct pci_bus*)bus; - dev->sysdata = bus->sysdata; - for (devfn = 0; devfn < 0x100; devfn += 8) { - dev->devfn = devfn; - found += pci_rescan_slot(dev); - } - - if (found) { - pci_bus_assign_resources(bus); - list_for_each_entry(dev, &bus->devices, bus_list) { - /* Skip already-added devices */ - if (dev->is_added) - continue; - retval = pci_bus_add_device(dev); - if (retval) - dev_err(&dev->dev, - "Error adding device, continuing\n"); - else - add_slot(dev); - } - pci_bus_add_devices(bus); - } - kfree(dev); -} + if (!slot) + return -ENOMEM; -/* recursively scan all buses */ -static void pci_rescan_buses(const struct list_head *list) -{ - const struct list_head *l; - list_for_each(l,list) { - const struct pci_bus *b = pci_bus_b(l); - pci_rescan_bus(b); - pci_rescan_buses(&b->children); + if (kobject_init_and_add(&slot->kobj, &legacy_ktype, + &pci_slots_kset->kobj, "%s", + dev_name(&pdev->dev))) { + dev_warn(&pdev->dev, "Failed to created legacy fake slot\n"); + return -EINVAL; } -} + slot->dev = pci_dev_get(pdev); -/* initiate rescan of all pci buses */ -static inline void pci_rescan(void) { - pci_rescan_buses(&pci_root_buses); -} - -/* called from the single-threaded workqueue handler to rescan all pci buses */ -static void pci_rescan_worker(struct work_struct *work) -{ - pci_rescan(); -} + list_add(&slot->list, &legacy_list); -static int enable_slot(struct hotplug_slot *hotplug_slot) -{ - /* mis-use enable_slot for rescanning of the pci bus */ - cancel_work_sync(&pci_rescan_work); - queue_work(dummyphp_wq, &pci_rescan_work); return 0; } -static int disable_slot(struct hotplug_slot *slot) +static int legacy_notify(struct notifier_block *nb, + unsigned long action, void *data) { - struct dummy_slot *dslot; - struct pci_dev *dev; - int func; - - if (!slot) - return -ENODEV; - dslot = slot->private; - - dbg("%s - physical_slot = %s\n", __func__, hotplug_slot_name(slot)); + struct pci_dev *pdev = to_pci_dev(data); - for (func = 7; func >= 0; func--) { - dev = pci_get_slot(dslot->dev->bus, dslot->dev->devfn + func); - if (!dev) - continue; + if (action == BUS_NOTIFY_ADD_DEVICE) { + legacy_add_slot(pdev); + } else if (action == BUS_NOTIFY_DEL_DEVICE) { + struct legacy_slot *slot; - if (test_and_set_bit(0, &dslot->removed)) { - dbg("Slot already scheduled for removal\n"); - pci_dev_put(dev); - return -ENODEV; - } + list_for_each_entry(slot, &legacy_list, list) + if (slot->dev == pdev) + goto found; - /* remove the device from the pci core */ - pci_remove_bus_device(dev); - - /* queue work item to blow away this sysfs entry and other - * parts. - */ - INIT_WORK(&dslot->remove_work, remove_slot_worker); - queue_work(dummyphp_wq, &dslot->remove_work); - - pci_dev_put(dev); + dev_warn(&pdev->dev, "Missing legacy fake slot?"); + return -ENODEV; +found: + kobject_del(&slot->kobj); + list_del(&slot->list); + kobject_put(&slot->kobj); } + return 0; } -static void cleanup_slots (void) -{ - struct list_head *tmp; - struct list_head *next; - struct dummy_slot *dslot; - - destroy_workqueue(dummyphp_wq); - list_for_each_safe (tmp, next, &slot_list) { - dslot = list_entry (tmp, struct dummy_slot, node); - remove_slot(dslot); - } - -} +static struct notifier_block legacy_notifier = { + .notifier_call = legacy_notify +}; -static int __init dummyphp_init(void) +static int __init init_legacy(void) { - info(DRIVER_DESC "\n"); + struct pci_dev *pdev = NULL; - dummyphp_wq = create_singlethread_workqueue(MY_NAME); - if (!dummyphp_wq) - return -ENOMEM; + /* Add existing devices */ + while ((pdev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, pdev))) + legacy_add_slot(pdev); - return pci_scan_buses(); + /* Be alerted of any new ones */ + bus_register_notifier(&pci_bus_type, &legacy_notifier); + return 0; } +module_init(init_legacy); - -static void __exit dummyphp_exit(void) +static void __exit remove_legacy(void) { - cleanup_slots(); + struct legacy_slot *slot, *tmp; + + bus_unregister_notifier(&pci_bus_type, &legacy_notifier); + + list_for_each_entry_safe(slot, tmp, &legacy_list, list) { + list_del(&slot->list); + kobject_del(&slot->kobj); + kobject_put(&slot->kobj); + } } +module_exit(remove_legacy); -module_init(dummyphp_init); -module_exit(dummyphp_exit); -MODULE_AUTHOR(DRIVER_AUTHOR); -MODULE_DESCRIPTION(DRIVER_DESC); +MODULE_AUTHOR("Trent Piepho <xyzzy@speakeasy.org>"); +MODULE_DESCRIPTION("Legacy version of the fakephp interface"); MODULE_LICENSE("GPL"); -module_param(debug, bool, S_IRUGO | S_IWUSR); -MODULE_PARM_DESC(debug, "Debugging mode enabled or not"); -module_param(dup_slots, bool, S_IRUGO | S_IWUSR); -MODULE_PARM_DESC(dup_slots, "Force duplicate slot names for debugging"); 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..844580489d4 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) @@ -540,10 +550,10 @@ 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 * * Registers a hotplug slot with the pci hotplug subsystem, which will allow @@ -551,8 +561,9 @@ static struct hotplug_slot *get_slot_from_name (const char *name) * * 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 +578,16 @@ 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 +697,6 @@ 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 39ae37589fd..e6cf096498b 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -46,10 +46,10 @@ extern int pciehp_force; extern struct workqueue_struct *pciehp_wq; #define dbg(format, arg...) \ - do { \ - if (pciehp_debug) \ - printk("%s: " format, MY_NAME , ## arg); \ - } while (0) +do { \ + if (pciehp_debug) \ + printk(KERN_DEBUG "%s: " format, MY_NAME , ## arg); \ +} while (0) #define err(format, arg...) \ printk(KERN_ERR "%s: " format, MY_NAME , ## arg) #define info(format, arg...) \ @@ -60,7 +60,7 @@ extern struct workqueue_struct *pciehp_wq; #define ctrl_dbg(ctrl, format, arg...) \ do { \ if (pciehp_debug) \ - dev_printk(, &ctrl->pcie->device, \ + dev_printk(KERN_DEBUG, &ctrl->pcie->device, \ format, ## arg); \ } while (0) #define ctrl_err(ctrl, format, arg...) \ @@ -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; }; @@ -108,10 +107,11 @@ struct controller { u32 slot_cap; u8 cap_base; struct timer_list poll_timer; - int cmd_busy; + unsigned int cmd_busy:1; unsigned int no_cmd_complete:1; unsigned int link_active_reporting:1; unsigned int notification_enabled:1; + unsigned int power_fault_detected; }; #define INT_BUTTON_IGNORE 0 @@ -202,8 +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_acpi.c b/drivers/pci/hotplug/pciehp_acpi.c index 438d795f9fe..96048010e7d 100644 --- a/drivers/pci/hotplug/pciehp_acpi.c +++ b/drivers/pci/hotplug/pciehp_acpi.c @@ -67,37 +67,27 @@ static int __init parse_detect_mode(void) return PCIEHP_DETECT_DEFAULT; } -static struct pcie_port_service_id __initdata port_pci_ids[] = { - { - .vendor = PCI_ANY_ID, - .device = PCI_ANY_ID, - .port_type = PCIE_ANY_PORT, - .service_type = PCIE_PORT_SERVICE_HP, - .driver_data = 0, - }, { /* end: all zeroes */ } -}; - static int __initdata dup_slot_id; static int __initdata acpi_slot_detected; static struct list_head __initdata dummy_slots = LIST_HEAD_INIT(dummy_slots); /* Dummy driver for dumplicate name detection */ -static int __init dummy_probe(struct pcie_device *dev, - const struct pcie_port_service_id *id) +static int __init dummy_probe(struct pcie_device *dev) { int pos; u32 slot_cap; struct slot *slot, *tmp; struct pci_dev *pdev = dev->port; struct pci_bus *pbus = pdev->subordinate; - if (!(slot = kzalloc(sizeof(*slot), GFP_KERNEL))) - return -ENOMEM; /* Note: pciehp_detect_mode != PCIEHP_DETECT_ACPI here */ if (pciehp_get_hp_hw_control_from_firmware(pdev)) return -ENODEV; if (!(pos = pci_find_capability(pdev, PCI_CAP_ID_EXP))) return -ENODEV; pci_read_config_dword(pdev, pos + PCI_EXP_SLTCAP, &slot_cap); + slot = kzalloc(sizeof(*slot), GFP_KERNEL); + if (!slot) + return -ENOMEM; slot->number = slot_cap >> 19; list_for_each_entry(tmp, &dummy_slots, slot_list) { if (tmp->number == slot->number) @@ -111,7 +101,8 @@ static int __init dummy_probe(struct pcie_device *dev, static struct pcie_port_service_driver __initdata dummy_driver = { .name = "pciehp_dummy", - .id_table = port_pci_ids, + .port_type = PCIE_ANY_PORT, + .service = PCIE_PORT_SERVICE_HP, .probe = dummy_probe, }; diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c index 681e3912b82..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); - } } /* @@ -401,7 +291,7 @@ static int get_cur_bus_speed(struct hotplug_slot *hotplug_slot, enum pci_bus_spe return 0; } -static int pciehp_probe(struct pcie_device *dev, const struct pcie_port_service_id *id) +static int pciehp_probe(struct pcie_device *dev) { int rc; struct controller *ctrl; @@ -475,7 +365,7 @@ static void pciehp_remove (struct pcie_device *dev) } #ifdef CONFIG_PM -static int pciehp_suspend (struct pcie_device *dev, pm_message_t state) +static int pciehp_suspend (struct pcie_device *dev) { dev_info(&dev->device, "%s ENTRY\n", __func__); return 0; @@ -503,20 +393,12 @@ static int pciehp_resume (struct pcie_device *dev) } return 0; } -#endif - -static struct pcie_port_service_id port_pci_ids[] = { { - .vendor = PCI_ANY_ID, - .device = PCI_ANY_ID, - .port_type = PCIE_ANY_PORT, - .service_type = PCIE_PORT_SERVICE_HP, - .driver_data = 0, - }, { /* end: all zeroes */ } -}; +#endif /* PM */ static struct pcie_port_service_driver hpdriver_portdrv = { .name = PCIE_MODULE_NAME, - .id_table = &port_pci_ids[0], + .port_type = PCIE_ANY_PORT, + .service = PCIE_PORT_SERVICE_HP, .probe = pciehp_probe, .remove = pciehp_remove, diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index 7a16c6897bb..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; @@ -548,23 +519,21 @@ static int hpc_power_on_slot(struct slot * slot) slot_cmd = POWER_ON; cmd_mask = PCI_EXP_SLTCTL_PCC; - /* Enable detection that we turned off at slot power-off time */ if (!pciehp_poll_mode) { - slot_cmd |= (PCI_EXP_SLTCTL_PFDE | PCI_EXP_SLTCTL_MRLSCE | - PCI_EXP_SLTCTL_PDCE); - cmd_mask |= (PCI_EXP_SLTCTL_PFDE | PCI_EXP_SLTCTL_MRLSCE | - PCI_EXP_SLTCTL_PDCE); + /* Enable power fault detection turned off at power off time */ + slot_cmd |= PCI_EXP_SLTCTL_PFDE; + cmd_mask |= PCI_EXP_SLTCTL_PFDE; } retval = pcie_write_cmd(ctrl, slot_cmd, cmd_mask); - if (retval) { ctrl_err(ctrl, "Write %x command failed!\n", slot_cmd); - return -1; + return retval; } ctrl_dbg(ctrl, "%s: SLOTCTRL %x write cmd %x\n", __func__, ctrl->cap_base + PCI_EXP_SLTCTL, slot_cmd); + ctrl->power_fault_detected = 0; return retval; } @@ -621,18 +590,10 @@ static int hpc_power_off_slot(struct slot * slot) slot_cmd = POWER_OFF; cmd_mask = PCI_EXP_SLTCTL_PCC; - /* - * If we get MRL or presence detect interrupts now, the isr - * will notice the sticky power-fault bit too and issue power - * indicator change commands. This will lead to an endless loop - * of command completions, since the power-fault bit remains on - * till the slot is powered on again. - */ if (!pciehp_poll_mode) { - slot_cmd &= ~(PCI_EXP_SLTCTL_PFDE | PCI_EXP_SLTCTL_MRLSCE | - PCI_EXP_SLTCTL_PDCE); - cmd_mask |= (PCI_EXP_SLTCTL_PFDE | PCI_EXP_SLTCTL_MRLSCE | - PCI_EXP_SLTCTL_PDCE); + /* Disable power fault detection */ + slot_cmd &= ~PCI_EXP_SLTCTL_PFDE; + cmd_mask |= PCI_EXP_SLTCTL_PFDE; } retval = pcie_write_cmd(ctrl, slot_cmd, cmd_mask); @@ -672,10 +633,11 @@ static irqreturn_t pcie_isr(int irq, void *dev_id) detected &= (PCI_EXP_SLTSTA_ABP | PCI_EXP_SLTSTA_PFD | PCI_EXP_SLTSTA_MRLSC | PCI_EXP_SLTSTA_PDC | PCI_EXP_SLTSTA_CC); + detected &= ~intr_loc; intr_loc |= detected; if (!intr_loc) return IRQ_NONE; - if (detected && pciehp_writew(ctrl, PCI_EXP_SLTSTA, detected)) { + if (detected && pciehp_writew(ctrl, PCI_EXP_SLTSTA, intr_loc)) { ctrl_err(ctrl, "%s: Cannot write to SLOTSTATUS\n", __func__); return IRQ_NONE; @@ -709,9 +671,10 @@ static irqreturn_t pcie_isr(int irq, void *dev_id) pciehp_handle_presence_change(p_slot); /* Check Power Fault Detected */ - if (intr_loc & PCI_EXP_SLTSTA_PFD) + if ((intr_loc & PCI_EXP_SLTSTA_PFD) && !ctrl->power_fault_detected) { + ctrl->power_fault_detected = 1; pciehp_handle_power_fault(p_slot); - + } return IRQ_HANDLED; } @@ -882,8 +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..a4494d78e7c 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, @@ -679,7 +678,7 @@ 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 +715,7 @@ 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.h b/drivers/pci/hotplug/shpchp.h index 6aba0b6cf2e..974e924ca96 100644 --- a/drivers/pci/hotplug/shpchp.h +++ b/drivers/pci/hotplug/shpchp.h @@ -48,10 +48,10 @@ extern int shpchp_debug; extern struct workqueue_struct *shpchp_wq; #define dbg(format, arg...) \ - do { \ - if (shpchp_debug) \ - printk("%s: " format, MY_NAME , ## arg); \ - } while (0) +do { \ + if (shpchp_debug) \ + printk(KERN_DEBUG "%s: " format, MY_NAME , ## arg); \ +} while (0) #define err(format, arg...) \ printk(KERN_ERR "%s: " format, MY_NAME , ## arg) #define info(format, arg...) \ @@ -62,7 +62,7 @@ extern struct workqueue_struct *shpchp_wq; #define ctrl_dbg(ctrl, format, arg...) \ do { \ if (shpchp_debug) \ - dev_printk(, &ctrl->pci_dev->dev, \ + dev_printk(KERN_DEBUG, &ctrl->pci_dev->dev, \ format, ## arg); \ } while (0) #define ctrl_err(ctrl, format, arg...) \ 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/hotplug/shpchp_pci.c b/drivers/pci/hotplug/shpchp_pci.c index 138f161becc..aa315e52529 100644 --- a/drivers/pci/hotplug/shpchp_pci.c +++ b/drivers/pci/hotplug/shpchp_pci.c @@ -137,7 +137,7 @@ int __ref shpchp_configure_device(struct slot *p_slot) busnr)) break; } - if (busnr >= end) { + if (busnr > end) { ctrl_err(ctrl, "No free bus for hot-added bridge\n"); pci_dev_put(dev); diff --git a/drivers/pci/htirq.c b/drivers/pci/htirq.c index bf7d6ce9bbb..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); @@ -158,6 +160,7 @@ int ht_create_irq(struct pci_dev *dev, int idx) /** * ht_destroy_irq - destroy an irq created with ht_create_irq + * @irq: irq to be destroyed * * This reverses ht_create_irq removing the specified irq from * existence. The irq should be free before this happens. diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c index f3f686581a9..cd389162735 100644 --- a/drivers/pci/intel-iommu.c +++ b/drivers/pci/intel-iommu.c @@ -36,6 +36,7 @@ #include <linux/iova.h> #include <linux/iommu.h> #include <linux/intel-iommu.h> +#include <linux/sysdev.h> #include <asm/cacheflush.h> #include <asm/iommu.h> #include "pci.h" @@ -55,8 +56,12 @@ #define DOMAIN_MAX_ADDR(gaw) ((((u64)1) << gaw) - 1) #define IOVA_PFN(addr) ((addr) >> PAGE_SHIFT) -#define DMA_32BIT_PFN IOVA_PFN(DMA_32BIT_MASK) -#define DMA_64BIT_PFN IOVA_PFN(DMA_64BIT_MASK) +#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 /* global iommu list, set NULL for ignored DMAR units */ static struct intel_iommu **g_iommus; @@ -164,7 +169,8 @@ static inline void context_clear_entry(struct context_entry *context) * 1: writable * 2-6: reserved * 7: super page - * 8-11: available + * 8-10: available + * 11: snoop behavior * 12-63: Host physcial address */ struct dma_pte { @@ -186,6 +192,11 @@ static inline void dma_set_pte_writable(struct dma_pte *pte) pte->val |= DMA_PTE_WRITE; } +static inline void dma_set_pte_snp(struct dma_pte *pte) +{ + pte->val |= DMA_PTE_SNP; +} + static inline void dma_set_pte_prot(struct dma_pte *pte, unsigned long prot) { pte->val = (pte->val & ~3) | (prot & 3); @@ -231,6 +242,7 @@ struct dmar_domain { int flags; /* flags to find out type of domain */ int iommu_coherency;/* indicate coherency of iommu access */ + int iommu_snooping; /* indicate snooping control feature*/ int iommu_count; /* reference count of iommu */ spinlock_t iommu_lock; /* protect iommu set in domain */ u64 max_addr; /* maximum mapped address */ @@ -240,7 +252,8 @@ struct dmar_domain { struct device_domain_info { struct list_head link; /* link to domain siblings */ struct list_head global; /* link to global list */ - u8 bus; /* PCI bus numer */ + int segment; /* PCI domain */ + u8 bus; /* PCI bus number */ u8 devfn; /* PCI devfn number */ struct pci_dev *dev; /* it's NULL for PCIE-to-PCI bridge */ struct dmar_domain *domain; /* pointer to domain */ @@ -421,7 +434,6 @@ static struct intel_iommu *domain_get_iommu(struct dmar_domain *domain) return g_iommus[iommu_id]; } -/* "Coherency" capability may be different across iommus */ static void domain_update_iommu_coherency(struct dmar_domain *domain) { int i; @@ -438,7 +450,30 @@ static void domain_update_iommu_coherency(struct dmar_domain *domain) } } -static struct intel_iommu *device_to_iommu(u8 bus, u8 devfn) +static void domain_update_iommu_snooping(struct dmar_domain *domain) +{ + int i; + + domain->iommu_snooping = 1; + + i = find_first_bit(&domain->iommu_bmp, g_num_of_iommus); + for (; i < g_num_of_iommus; ) { + if (!ecap_sc_support(g_iommus[i]->ecap)) { + domain->iommu_snooping = 0; + break; + } + i = find_next_bit(&domain->iommu_bmp, g_num_of_iommus, i+1); + } +} + +/* Some capabilities may be different across iommus */ +static void domain_update_iommu_cap(struct dmar_domain *domain) +{ + domain_update_iommu_coherency(domain); + domain_update_iommu_snooping(domain); +} + +static struct intel_iommu *device_to_iommu(int segment, u8 bus, u8 devfn) { struct dmar_drhd_unit *drhd = NULL; int i; @@ -446,12 +481,20 @@ static struct intel_iommu *device_to_iommu(u8 bus, u8 devfn) for_each_drhd_unit(drhd) { if (drhd->ignored) continue; + if (segment != drhd->segment) + continue; - for (i = 0; i < drhd->devices_cnt; i++) + for (i = 0; i < drhd->devices_cnt; i++) { if (drhd->devices[i] && drhd->devices[i]->bus->number == bus && drhd->devices[i]->devfn == devfn) return drhd->iommu; + if (drhd->devices[i] && + drhd->devices[i]->subordinate && + drhd->devices[i]->subordinate->number <= bus && + drhd->devices[i]->subordinate->subordinate >= bus) + return drhd->iommu; + } if (drhd->include_all) return drhd->iommu; @@ -689,15 +732,17 @@ static void dma_pte_clear_one(struct dmar_domain *domain, u64 addr) static void dma_pte_clear_range(struct dmar_domain *domain, u64 start, u64 end) { int addr_width = agaw_to_width(domain->agaw); + int npages; start &= (((u64)1) << addr_width) - 1; end &= (((u64)1) << addr_width) - 1; /* in case it's partial page */ - start = PAGE_ALIGN(start); - end &= PAGE_MASK; + start &= PAGE_MASK; + end = PAGE_ALIGN(end); + npages = (end - start) / VTD_PAGE_SIZE; /* we don't need lock here, nobody else touches the iova range */ - while (start < end) { + while (npages--) { dma_pte_clear_one(domain, start); start += VTD_PAGE_SIZE; } @@ -1004,194 +1049,6 @@ static int iommu_disable_translation(struct intel_iommu *iommu) return 0; } -/* iommu interrupt handling. Most stuff are MSI-like. */ - -static const char *fault_reason_strings[] = -{ - "Software", - "Present bit in root entry is clear", - "Present bit in context entry is clear", - "Invalid context entry", - "Access beyond MGAW", - "PTE Write access is not set", - "PTE Read access is not set", - "Next page table ptr is invalid", - "Root table address invalid", - "Context table ptr is invalid", - "non-zero reserved fields in RTP", - "non-zero reserved fields in CTP", - "non-zero reserved fields in PTE", -}; -#define MAX_FAULT_REASON_IDX (ARRAY_SIZE(fault_reason_strings) - 1) - -const char *dmar_get_fault_reason(u8 fault_reason) -{ - if (fault_reason > MAX_FAULT_REASON_IDX) - return "Unknown"; - else - return fault_reason_strings[fault_reason]; -} - -void dmar_msi_unmask(unsigned int irq) -{ - struct intel_iommu *iommu = get_irq_data(irq); - unsigned long flag; - - /* unmask it */ - spin_lock_irqsave(&iommu->register_lock, flag); - writel(0, iommu->reg + DMAR_FECTL_REG); - /* Read a reg to force flush the post write */ - readl(iommu->reg + DMAR_FECTL_REG); - spin_unlock_irqrestore(&iommu->register_lock, flag); -} - -void dmar_msi_mask(unsigned int irq) -{ - unsigned long flag; - struct intel_iommu *iommu = get_irq_data(irq); - - /* mask it */ - spin_lock_irqsave(&iommu->register_lock, flag); - writel(DMA_FECTL_IM, iommu->reg + DMAR_FECTL_REG); - /* Read a reg to force flush the post write */ - readl(iommu->reg + DMAR_FECTL_REG); - spin_unlock_irqrestore(&iommu->register_lock, flag); -} - -void dmar_msi_write(int irq, struct msi_msg *msg) -{ - struct intel_iommu *iommu = get_irq_data(irq); - unsigned long flag; - - spin_lock_irqsave(&iommu->register_lock, flag); - writel(msg->data, iommu->reg + DMAR_FEDATA_REG); - writel(msg->address_lo, iommu->reg + DMAR_FEADDR_REG); - writel(msg->address_hi, iommu->reg + DMAR_FEUADDR_REG); - spin_unlock_irqrestore(&iommu->register_lock, flag); -} - -void dmar_msi_read(int irq, struct msi_msg *msg) -{ - struct intel_iommu *iommu = get_irq_data(irq); - unsigned long flag; - - spin_lock_irqsave(&iommu->register_lock, flag); - msg->data = readl(iommu->reg + DMAR_FEDATA_REG); - msg->address_lo = readl(iommu->reg + DMAR_FEADDR_REG); - msg->address_hi = readl(iommu->reg + DMAR_FEUADDR_REG); - spin_unlock_irqrestore(&iommu->register_lock, flag); -} - -static int iommu_page_fault_do_one(struct intel_iommu *iommu, int type, - u8 fault_reason, u16 source_id, unsigned long long addr) -{ - const char *reason; - - reason = dmar_get_fault_reason(fault_reason); - - printk(KERN_ERR - "DMAR:[%s] Request device [%02x:%02x.%d] " - "fault addr %llx \n" - "DMAR:[fault reason %02d] %s\n", - (type ? "DMA Read" : "DMA Write"), - (source_id >> 8), PCI_SLOT(source_id & 0xFF), - PCI_FUNC(source_id & 0xFF), addr, fault_reason, reason); - return 0; -} - -#define PRIMARY_FAULT_REG_LEN (16) -static irqreturn_t iommu_page_fault(int irq, void *dev_id) -{ - struct intel_iommu *iommu = dev_id; - int reg, fault_index; - u32 fault_status; - unsigned long flag; - - spin_lock_irqsave(&iommu->register_lock, flag); - fault_status = readl(iommu->reg + DMAR_FSTS_REG); - - /* TBD: ignore advanced fault log currently */ - if (!(fault_status & DMA_FSTS_PPF)) - goto clear_overflow; - - fault_index = dma_fsts_fault_record_index(fault_status); - reg = cap_fault_reg_offset(iommu->cap); - while (1) { - u8 fault_reason; - u16 source_id; - u64 guest_addr; - int type; - u32 data; - - /* highest 32 bits */ - data = readl(iommu->reg + reg + - fault_index * PRIMARY_FAULT_REG_LEN + 12); - if (!(data & DMA_FRCD_F)) - break; - - fault_reason = dma_frcd_fault_reason(data); - type = dma_frcd_type(data); - - data = readl(iommu->reg + reg + - fault_index * PRIMARY_FAULT_REG_LEN + 8); - source_id = dma_frcd_source_id(data); - - guest_addr = dmar_readq(iommu->reg + reg + - fault_index * PRIMARY_FAULT_REG_LEN); - guest_addr = dma_frcd_page_addr(guest_addr); - /* clear the fault */ - writel(DMA_FRCD_F, iommu->reg + reg + - fault_index * PRIMARY_FAULT_REG_LEN + 12); - - spin_unlock_irqrestore(&iommu->register_lock, flag); - - iommu_page_fault_do_one(iommu, type, fault_reason, - source_id, guest_addr); - - fault_index++; - if (fault_index > cap_num_fault_regs(iommu->cap)) - fault_index = 0; - spin_lock_irqsave(&iommu->register_lock, flag); - } -clear_overflow: - /* clear primary fault overflow */ - fault_status = readl(iommu->reg + DMAR_FSTS_REG); - if (fault_status & DMA_FSTS_PFO) - writel(DMA_FSTS_PFO, iommu->reg + DMAR_FSTS_REG); - - spin_unlock_irqrestore(&iommu->register_lock, flag); - return IRQ_HANDLED; -} - -int dmar_set_interrupt(struct intel_iommu *iommu) -{ - int irq, ret; - - irq = create_irq(); - if (!irq) { - printk(KERN_ERR "IOMMU: no free vectors\n"); - return -EINVAL; - } - - set_irq_data(irq, iommu); - iommu->irq = irq; - - ret = arch_setup_dmar_msi(irq); - if (ret) { - set_irq_data(irq, NULL); - iommu->irq = 0; - destroy_irq(irq); - return 0; - } - - /* Force fault register is cleared */ - iommu_page_fault(irq, iommu); - - ret = request_irq(irq, iommu_page_fault, 0, iommu->name, iommu); - if (ret) - printk(KERN_ERR "IOMMU: can't request irq\n"); - return ret; -} static int iommu_init_domains(struct intel_iommu *iommu) { @@ -1363,7 +1220,7 @@ static void dmar_init_reserved_ranges(void) if (!r->flags || !(r->flags & IORESOURCE_MEM)) continue; addr = r->start; - addr &= PAGE_MASK; + addr &= PHYSICAL_PAGE_MASK; size = r->end - addr; size = PAGE_ALIGN(size); iova = reserve_iova(&reserved_iova_list, IOVA_PFN(addr), @@ -1429,6 +1286,11 @@ static int domain_init(struct dmar_domain *domain, int guest_width) else domain->iommu_coherency = 0; + if (ecap_sc_support(iommu->ecap)) + domain->iommu_snooping = 1; + else + domain->iommu_snooping = 0; + domain->iommu_count = 1; /* always allocate the top pgd */ @@ -1464,7 +1326,7 @@ static void domain_exit(struct dmar_domain *domain) } static int domain_context_mapping_one(struct dmar_domain *domain, - u8 bus, u8 devfn) + int segment, u8 bus, u8 devfn) { struct context_entry *context; unsigned long flags; @@ -1479,7 +1341,7 @@ static int domain_context_mapping_one(struct dmar_domain *domain, bus, PCI_SLOT(devfn), PCI_FUNC(devfn)); BUG_ON(!domain->pgd); - iommu = device_to_iommu(bus, devfn); + iommu = device_to_iommu(segment, bus, devfn); if (!iommu) return -ENODEV; @@ -1557,7 +1419,7 @@ static int domain_context_mapping_one(struct dmar_domain *domain, spin_lock_irqsave(&domain->iommu_lock, flags); if (!test_and_set_bit(iommu->seq_id, &domain->iommu_bmp)) { domain->iommu_count++; - domain_update_iommu_coherency(domain); + domain_update_iommu_cap(domain); } spin_unlock_irqrestore(&domain->iommu_lock, flags); return 0; @@ -1569,8 +1431,8 @@ domain_context_mapping(struct dmar_domain *domain, struct pci_dev *pdev) int ret; struct pci_dev *tmp, *parent; - ret = domain_context_mapping_one(domain, pdev->bus->number, - pdev->devfn); + ret = domain_context_mapping_one(domain, pci_domain_nr(pdev->bus), + pdev->bus->number, pdev->devfn); if (ret) return ret; @@ -1581,18 +1443,23 @@ domain_context_mapping(struct dmar_domain *domain, struct pci_dev *pdev) /* Secondary interface's bus number and devfn 0 */ parent = pdev->bus->self; while (parent != tmp) { - ret = domain_context_mapping_one(domain, parent->bus->number, - parent->devfn); + ret = domain_context_mapping_one(domain, + pci_domain_nr(parent->bus), + parent->bus->number, + parent->devfn); if (ret) return ret; parent = parent->bus->self; } if (tmp->is_pcie) /* this is a PCIE-to-PCI bridge */ return domain_context_mapping_one(domain, - tmp->subordinate->number, 0); + pci_domain_nr(tmp->subordinate), + tmp->subordinate->number, 0); else /* this is a legacy PCI bridge */ return domain_context_mapping_one(domain, - tmp->bus->number, tmp->devfn); + pci_domain_nr(tmp->bus), + tmp->bus->number, + tmp->devfn); } static int domain_context_mapped(struct pci_dev *pdev) @@ -1601,12 +1468,12 @@ static int domain_context_mapped(struct pci_dev *pdev) struct pci_dev *tmp, *parent; struct intel_iommu *iommu; - iommu = device_to_iommu(pdev->bus->number, pdev->devfn); + iommu = device_to_iommu(pci_domain_nr(pdev->bus), pdev->bus->number, + pdev->devfn); if (!iommu) return -ENODEV; - ret = device_context_mapped(iommu, - pdev->bus->number, pdev->devfn); + ret = device_context_mapped(iommu, pdev->bus->number, pdev->devfn); if (!ret) return ret; /* dependent device mapping */ @@ -1617,17 +1484,17 @@ static int domain_context_mapped(struct pci_dev *pdev) parent = pdev->bus->self; while (parent != tmp) { ret = device_context_mapped(iommu, parent->bus->number, - parent->devfn); + parent->devfn); if (!ret) return ret; parent = parent->bus->self; } if (tmp->is_pcie) - return device_context_mapped(iommu, - tmp->subordinate->number, 0); + return device_context_mapped(iommu, tmp->subordinate->number, + 0); else - return device_context_mapped(iommu, - tmp->bus->number, tmp->devfn); + return device_context_mapped(iommu, tmp->bus->number, + tmp->devfn); } static int @@ -1657,6 +1524,8 @@ domain_page_mapping(struct dmar_domain *domain, dma_addr_t iova, 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++; @@ -1692,7 +1561,7 @@ static void domain_remove_dev_info(struct dmar_domain *domain) info->dev->dev.archdata.iommu = NULL; spin_unlock_irqrestore(&device_domain_lock, flags); - iommu = device_to_iommu(info->bus, info->devfn); + iommu = device_to_iommu(info->segment, info->bus, info->devfn); iommu_detach_dev(iommu, info->bus, info->devfn); free_devinfo_mem(info); @@ -1727,11 +1596,14 @@ static struct dmar_domain *get_domain_for_dev(struct pci_dev *pdev, int gaw) struct pci_dev *dev_tmp; unsigned long flags; int bus = 0, devfn = 0; + int segment; domain = find_domain(pdev); if (domain) return domain; + segment = pci_domain_nr(pdev->bus); + dev_tmp = pci_find_upstream_pcie_bridge(pdev); if (dev_tmp) { if (dev_tmp->is_pcie) { @@ -1743,7 +1615,8 @@ static struct dmar_domain *get_domain_for_dev(struct pci_dev *pdev, int gaw) } spin_lock_irqsave(&device_domain_lock, flags); list_for_each_entry(info, &device_domain_list, global) { - if (info->bus == bus && info->devfn == devfn) { + if (info->segment == segment && + info->bus == bus && info->devfn == devfn) { found = info->domain; break; } @@ -1781,6 +1654,7 @@ static struct dmar_domain *get_domain_for_dev(struct pci_dev *pdev, int gaw) domain_exit(domain); goto error; } + info->segment = segment; info->bus = bus; info->devfn = devfn; info->dev = NULL; @@ -1792,7 +1666,8 @@ static struct dmar_domain *get_domain_for_dev(struct pci_dev *pdev, int gaw) found = NULL; spin_lock_irqsave(&device_domain_lock, flags); list_for_each_entry(tmp, &device_domain_list, global) { - if (tmp->bus == bus && tmp->devfn == devfn) { + if (tmp->segment == segment && + tmp->bus == bus && tmp->devfn == devfn) { found = tmp->domain; break; } @@ -1812,6 +1687,7 @@ found_domain: info = alloc_devinfo_mem(); if (!info) goto error; + info->segment = segment; info->bus = pdev->bus->number; info->devfn = pdev->devfn; info->dev = pdev; @@ -1970,7 +1846,7 @@ static inline void iommu_prepare_isa(void) ret = iommu_prepare_identity_map(pdev, 0, 16*1024*1024); if (ret) - printk("IOMMU: Failed to create 0-64M identity map, " + printk(KERN_ERR "IOMMU: Failed to create 0-64M identity map, " "floppy might not work\n"); } @@ -1987,7 +1863,7 @@ static int __init init_dmars(void) struct dmar_rmrr_unit *rmrr; struct pci_dev *pdev; struct intel_iommu *iommu; - int i, ret, unit = 0; + int i, ret; /* * for each drhd @@ -2043,11 +1919,40 @@ static int __init init_dmars(void) } } + /* + * Start from the sane iommu hardware state. + */ for_each_drhd_unit(drhd) { if (drhd->ignored) continue; iommu = drhd->iommu; + + /* + * If the queued invalidation is already initialized by us + * (for example, while enabling interrupt-remapping) then + * we got the things already rolling from a sane state. + */ + if (iommu->qi) + continue; + + /* + * Clear any previous faults. + */ + dmar_fault(-1, iommu); + /* + * Disable queued invalidation if supported and already enabled + * before OS handover. + */ + dmar_disable_qi(iommu); + } + + for_each_drhd_unit(drhd) { + if (drhd->ignored) + continue; + + iommu = drhd->iommu; + if (dmar_enable_qi(iommu)) { /* * Queued Invalidate not enabled, use Register Based @@ -2109,7 +2014,6 @@ static int __init init_dmars(void) if (drhd->ignored) continue; iommu = drhd->iommu; - sprintf (iommu->name, "dmar%d", unit++); iommu_flush_write_buffer(iommu); @@ -2171,15 +2075,15 @@ __intel_alloc_iova(struct device *dev, struct dmar_domain *domain, struct pci_dev *pdev = to_pci_dev(dev); struct iova *iova = NULL; - if (dma_mask <= DMA_32BIT_MASK || dmar_forcedac) + if (dma_mask <= DMA_BIT_MASK(32) || dmar_forcedac) iova = iommu_alloc_iova(domain, size, dma_mask); else { /* * First try to allocate an io virtual address in - * DMA_32BIT_MASK and if that fails then try allocating + * DMA_BIT_MASK(32) and if that fails then try allocating * from higher range */ - iova = iommu_alloc_iova(domain, size, DMA_32BIT_MASK); + iova = iommu_alloc_iova(domain, size, DMA_BIT_MASK(32)); if (!iova) iova = iommu_alloc_iova(domain, size, dma_mask); } @@ -2264,7 +2168,8 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr, * is not a big problem */ ret = domain_page_mapping(domain, start_paddr, - ((u64)paddr) & PAGE_MASK, size, prot); + ((u64)paddr) & PHYSICAL_PAGE_MASK, + size, prot); if (ret) goto error; @@ -2279,16 +2184,18 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr, error: if (iova) __free_iova(&domain->iovad, iova); - printk(KERN_ERR"Device %s request: %lx@%llx dir %d --- failed\n", + printk(KERN_ERR"Device %s request: %zx@%llx dir %d --- failed\n", pci_name(pdev), size, (unsigned long long)paddr, dir); return 0; } -dma_addr_t intel_map_single(struct device *hwdev, phys_addr_t paddr, - size_t size, int dir) +static dma_addr_t intel_map_page(struct device *dev, struct page *page, + unsigned long offset, size_t size, + enum dma_data_direction dir, + struct dma_attrs *attrs) { - return __intel_map_single(hwdev, paddr, size, dir, - to_pci_dev(hwdev)->dma_mask); + return __intel_map_single(dev, page_to_phys(page) + offset, size, + dir, to_pci_dev(dev)->dma_mask); } static void flush_unmaps(void) @@ -2352,8 +2259,9 @@ static void add_unmap(struct dmar_domain *dom, struct iova *iova) spin_unlock_irqrestore(&async_umap_flush_lock, flags); } -void intel_unmap_single(struct device *dev, dma_addr_t dev_addr, size_t size, - int dir) +static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr, + size_t size, enum dma_data_direction dir, + struct dma_attrs *attrs) { struct pci_dev *pdev = to_pci_dev(dev); struct dmar_domain *domain; @@ -2375,7 +2283,7 @@ void intel_unmap_single(struct device *dev, dma_addr_t dev_addr, size_t size, start_addr = iova->pfn_lo << PAGE_SHIFT; size = aligned_size((u64)dev_addr, size); - pr_debug("Device %s unmapping: %lx@%llx\n", + pr_debug("Device %s unmapping: %zx@%llx\n", pci_name(pdev), size, (unsigned long long)start_addr); /* clear the whole page */ @@ -2397,8 +2305,14 @@ void intel_unmap_single(struct device *dev, dma_addr_t dev_addr, size_t size, } } -void *intel_alloc_coherent(struct device *hwdev, size_t size, - dma_addr_t *dma_handle, gfp_t flags) +static void intel_unmap_single(struct device *dev, dma_addr_t dev_addr, size_t size, + int dir) +{ + intel_unmap_page(dev, dev_addr, size, dir, NULL); +} + +static void *intel_alloc_coherent(struct device *hwdev, size_t size, + dma_addr_t *dma_handle, gfp_t flags) { void *vaddr; int order; @@ -2421,8 +2335,8 @@ void *intel_alloc_coherent(struct device *hwdev, size_t size, return NULL; } -void intel_free_coherent(struct device *hwdev, size_t size, void *vaddr, - dma_addr_t dma_handle) +static void intel_free_coherent(struct device *hwdev, size_t size, void *vaddr, + dma_addr_t dma_handle) { int order; @@ -2433,10 +2347,9 @@ void intel_free_coherent(struct device *hwdev, size_t size, void *vaddr, free_pages((unsigned long)vaddr, order); } -#define SG_ENT_VIRT_ADDRESS(sg) (sg_virt((sg))) - -void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist, - int nelems, int dir) +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); @@ -2444,7 +2357,7 @@ void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist, unsigned long start_addr; struct iova *iova; size_t size = 0; - void *addr; + phys_addr_t addr; struct scatterlist *sg; struct intel_iommu *iommu; @@ -2460,7 +2373,7 @@ void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist, if (!iova) return; for_each_sg(sglist, sg, nelems, i) { - addr = SG_ENT_VIRT_ADDRESS(sg); + addr = page_to_phys(sg_page(sg)) + sg->offset; size += aligned_size((u64)addr, sg->length); } @@ -2487,16 +2400,16 @@ static int intel_nontranslate_map_sg(struct device *hddev, for_each_sg(sglist, sg, nelems, i) { BUG_ON(!sg_page(sg)); - sg->dma_address = virt_to_bus(SG_ENT_VIRT_ADDRESS(sg)); + sg->dma_address = page_to_phys(sg_page(sg)) + sg->offset; sg->dma_length = sg->length; } return nelems; } -int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, - int dir) +static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, + enum dma_data_direction dir, struct dma_attrs *attrs) { - void *addr; + phys_addr_t addr; int i; struct pci_dev *pdev = to_pci_dev(hwdev); struct dmar_domain *domain; @@ -2520,8 +2433,7 @@ int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, iommu = domain_get_iommu(domain); for_each_sg(sglist, sg, nelems, i) { - addr = SG_ENT_VIRT_ADDRESS(sg); - addr = (void *)virt_to_phys(addr); + addr = page_to_phys(sg_page(sg)) + sg->offset; size += aligned_size((u64)addr, sg->length); } @@ -2544,12 +2456,11 @@ int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, start_addr = iova->pfn_lo << PAGE_SHIFT; offset = 0; for_each_sg(sglist, sg, nelems, i) { - addr = SG_ENT_VIRT_ADDRESS(sg); - addr = (void *)virt_to_phys(addr); + 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) & PAGE_MASK, - size, prot); + ((u64)addr) & PHYSICAL_PAGE_MASK, + size, prot); if (ret) { /* clear the page */ dma_pte_clear_range(domain, start_addr, @@ -2574,13 +2485,19 @@ int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, return nelems; } -static struct dma_mapping_ops intel_dma_ops = { +static int intel_mapping_error(struct device *dev, dma_addr_t dma_addr) +{ + return !dma_addr; +} + +struct dma_map_ops intel_dma_ops = { .alloc_coherent = intel_alloc_coherent, .free_coherent = intel_free_coherent, - .map_single = intel_map_single, - .unmap_single = intel_unmap_single, .map_sg = intel_map_sg, .unmap_sg = intel_unmap_sg, + .map_page = intel_map_page, + .unmap_page = intel_unmap_page, + .mapping_error = intel_mapping_error, }; static inline int iommu_domain_cache_init(void) @@ -2707,6 +2624,150 @@ static void __init init_no_remapping_devices(void) } } +#ifdef CONFIG_SUSPEND +static int init_iommu_hw(void) +{ + struct dmar_drhd_unit *drhd; + struct intel_iommu *iommu = NULL; + + for_each_active_iommu(iommu, drhd) + if (iommu->qi) + dmar_reenable_qi(iommu); + + for_each_active_iommu(iommu, drhd) { + iommu_flush_write_buffer(iommu); + + 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_disable_protect_mem_regions(iommu); + iommu_enable_translation(iommu); + } + + return 0; +} + +static void iommu_flush_all(void) +{ + struct dmar_drhd_unit *drhd; + struct intel_iommu *iommu; + + for_each_active_iommu(iommu, drhd) { + 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); + } +} + +static int iommu_suspend(struct sys_device *dev, pm_message_t state) +{ + struct dmar_drhd_unit *drhd; + struct intel_iommu *iommu = NULL; + unsigned long flag; + + for_each_active_iommu(iommu, drhd) { + iommu->iommu_state = kzalloc(sizeof(u32) * MAX_SR_DMAR_REGS, + GFP_ATOMIC); + if (!iommu->iommu_state) + goto nomem; + } + + iommu_flush_all(); + + for_each_active_iommu(iommu, drhd) { + iommu_disable_translation(iommu); + + spin_lock_irqsave(&iommu->register_lock, flag); + + iommu->iommu_state[SR_DMAR_FECTL_REG] = + readl(iommu->reg + DMAR_FECTL_REG); + iommu->iommu_state[SR_DMAR_FEDATA_REG] = + readl(iommu->reg + DMAR_FEDATA_REG); + iommu->iommu_state[SR_DMAR_FEADDR_REG] = + readl(iommu->reg + DMAR_FEADDR_REG); + iommu->iommu_state[SR_DMAR_FEUADDR_REG] = + readl(iommu->reg + DMAR_FEUADDR_REG); + + spin_unlock_irqrestore(&iommu->register_lock, flag); + } + return 0; + +nomem: + for_each_active_iommu(iommu, drhd) + kfree(iommu->iommu_state); + + return -ENOMEM; +} + +static int iommu_resume(struct sys_device *dev) +{ + struct dmar_drhd_unit *drhd; + struct intel_iommu *iommu = NULL; + unsigned long flag; + + if (init_iommu_hw()) { + WARN(1, "IOMMU setup failed, DMAR can not resume!\n"); + return -EIO; + } + + for_each_active_iommu(iommu, drhd) { + + spin_lock_irqsave(&iommu->register_lock, flag); + + writel(iommu->iommu_state[SR_DMAR_FECTL_REG], + iommu->reg + DMAR_FECTL_REG); + writel(iommu->iommu_state[SR_DMAR_FEDATA_REG], + iommu->reg + DMAR_FEDATA_REG); + writel(iommu->iommu_state[SR_DMAR_FEADDR_REG], + iommu->reg + DMAR_FEADDR_REG); + writel(iommu->iommu_state[SR_DMAR_FEUADDR_REG], + iommu->reg + DMAR_FEUADDR_REG); + + spin_unlock_irqrestore(&iommu->register_lock, flag); + } + + for_each_active_iommu(iommu, drhd) + kfree(iommu->iommu_state); + + return 0; +} + +static struct sysdev_class iommu_sysclass = { + .name = "iommu", + .resume = iommu_resume, + .suspend = iommu_suspend, +}; + +static struct sys_device device_iommu = { + .cls = &iommu_sysclass, +}; + +static int __init init_iommu_sysfs(void) +{ + int error; + + error = sysdev_class_register(&iommu_sysclass); + if (error) + return error; + + error = sysdev_register(&device_iommu); + if (error) + sysdev_class_unregister(&iommu_sysclass); + + return error; +} + +#else +static int __init init_iommu_sysfs(void) +{ + return 0; +} +#endif /* CONFIG_PM */ + int __init intel_iommu_init(void) { int ret = 0; @@ -2742,6 +2803,7 @@ 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); @@ -2758,6 +2820,7 @@ static int vm_domain_add_dev_info(struct dmar_domain *domain, if (!info) return -ENOMEM; + info->segment = pci_domain_nr(pdev->bus); info->bus = pdev->bus->number; info->devfn = pdev->devfn; info->dev = pdev; @@ -2772,6 +2835,33 @@ static int vm_domain_add_dev_info(struct dmar_domain *domain, return 0; } +static void iommu_detach_dependent_devices(struct intel_iommu *iommu, + struct pci_dev *pdev) +{ + struct pci_dev *tmp, *parent; + + if (!iommu || !pdev) + return; + + /* dependent device detach */ + tmp = pci_find_upstream_pcie_bridge(pdev); + /* Secondary interface's bus number and devfn 0 */ + if (tmp) { + parent = pdev->bus->self; + while (parent != tmp) { + iommu_detach_dev(iommu, parent->bus->number, + parent->devfn); + parent = parent->bus->self; + } + if (tmp->is_pcie) /* this is a PCIE-to-PCI bridge */ + iommu_detach_dev(iommu, + tmp->subordinate->number, 0); + else /* this is a legacy PCI bridge */ + iommu_detach_dev(iommu, tmp->bus->number, + tmp->devfn); + } +} + static void vm_domain_remove_one_dev_info(struct dmar_domain *domain, struct pci_dev *pdev) { @@ -2781,13 +2871,15 @@ static void vm_domain_remove_one_dev_info(struct dmar_domain *domain, int found = 0; struct list_head *entry, *tmp; - iommu = device_to_iommu(pdev->bus->number, pdev->devfn); + iommu = device_to_iommu(pci_domain_nr(pdev->bus), pdev->bus->number, + pdev->devfn); if (!iommu) return; spin_lock_irqsave(&device_domain_lock, flags); list_for_each_safe(entry, tmp, &domain->devices) { info = list_entry(entry, struct device_domain_info, link); + /* No need to compare PCI domain; it has to be the same */ if (info->bus == pdev->bus->number && info->devfn == pdev->devfn) { list_del(&info->link); @@ -2797,6 +2889,7 @@ static void vm_domain_remove_one_dev_info(struct dmar_domain *domain, spin_unlock_irqrestore(&device_domain_lock, flags); iommu_detach_dev(iommu, info->bus, info->devfn); + iommu_detach_dependent_devices(iommu, pdev); free_devinfo_mem(info); spin_lock_irqsave(&device_domain_lock, flags); @@ -2811,7 +2904,8 @@ static void vm_domain_remove_one_dev_info(struct dmar_domain *domain, * owned by this domain, clear this iommu in iommu_bmp * update iommu count and coherency */ - if (device_to_iommu(info->bus, info->devfn) == iommu) + if (iommu == device_to_iommu(info->segment, info->bus, + info->devfn)) found = 1; } @@ -2820,7 +2914,7 @@ static void vm_domain_remove_one_dev_info(struct dmar_domain *domain, spin_lock_irqsave(&domain->iommu_lock, tmp_flags); clear_bit(iommu->seq_id, &domain->iommu_bmp); domain->iommu_count--; - domain_update_iommu_coherency(domain); + domain_update_iommu_cap(domain); spin_unlock_irqrestore(&domain->iommu_lock, tmp_flags); } @@ -2844,17 +2938,18 @@ static void vm_domain_remove_all_dev_info(struct dmar_domain *domain) spin_unlock_irqrestore(&device_domain_lock, flags1); - iommu = device_to_iommu(info->bus, info->devfn); + 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); /* clear this iommu in iommu_bmp, update iommu count - * and coherency + * and capabilities */ spin_lock_irqsave(&domain->iommu_lock, flags2); if (test_and_clear_bit(iommu->seq_id, &domain->iommu_bmp)) { domain->iommu_count--; - domain_update_iommu_coherency(domain); + domain_update_iommu_cap(domain); } spin_unlock_irqrestore(&domain->iommu_lock, flags2); @@ -3031,7 +3126,8 @@ static int intel_iommu_attach_device(struct iommu_domain *domain, } } - iommu = device_to_iommu(pdev->bus->number, pdev->devfn); + iommu = device_to_iommu(pci_domain_nr(pdev->bus), pdev->bus->number, + pdev->devfn); if (!iommu) return -ENODEV; @@ -3077,6 +3173,8 @@ static int intel_iommu_map_range(struct iommu_domain *domain, prot |= DMA_PTE_READ; if (iommu_prot & IOMMU_WRITE) prot |= DMA_PTE_WRITE; + if ((iommu_prot & IOMMU_CACHE) && dmar_domain->iommu_snooping) + prot |= DMA_PTE_SNP; max_addr = (iova & VTD_PAGE_MASK) + VTD_PAGE_ALIGN(size); if (dmar_domain->max_addr < max_addr) { @@ -3130,6 +3228,17 @@ static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain, return phys; } +static int intel_iommu_domain_has_cap(struct iommu_domain *domain, + unsigned long cap) +{ + struct dmar_domain *dmar_domain = domain->priv; + + if (cap == IOMMU_CAP_CACHE_COHERENCY) + return dmar_domain->iommu_snooping; + + return 0; +} + static struct iommu_ops intel_iommu_ops = { .domain_init = intel_iommu_domain_init, .domain_destroy = intel_iommu_domain_destroy, @@ -3138,6 +3247,7 @@ static struct iommu_ops intel_iommu_ops = { .map = intel_iommu_map_range, .unmap = intel_iommu_unmap_range, .iova_to_phys = intel_iommu_iova_to_phys, + .domain_has_cap = intel_iommu_domain_has_cap, }; static void __devinit quirk_iommu_rwbf(struct pci_dev *dev) diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c index 45effc5726c..3a0cb0bb059 100644 --- a/drivers/pci/intr_remapping.c +++ b/drivers/pci/intr_remapping.c @@ -6,13 +6,23 @@ #include <linux/irq.h> #include <asm/io_apic.h> #include <asm/smp.h> +#include <asm/cpu.h> #include <linux/intel-iommu.h> #include "intr_remapping.h" +#include <acpi/acpi.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; @@ -20,16 +30,13 @@ struct irq_2_iommu { u8 irte_mask; }; -#ifdef CONFIG_SPARSE_IRQ -static struct irq_2_iommu *get_one_free_irq_2_iommu(int cpu) +#ifdef CONFIG_GENERIC_HARDIRQS +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; } @@ -46,7 +53,7 @@ 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; @@ -54,7 +61,7 @@ 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; @@ -63,14 +70,14 @@ 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 */ @@ -116,21 +123,22 @@ int get_irte(int irq, struct irte *entry) { int index; struct irq_2_iommu *irq_iommu; + unsigned long flags; if (!entry) return -1; - spin_lock(&irq_2_ir_lock); + spin_lock_irqsave(&irq_2_ir_lock, flags); irq_iommu = valid_irq_2_iommu(irq); if (!irq_iommu) { - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return -1; } index = irq_iommu->irte_index + irq_iommu->sub_handle; *entry = *(irq_iommu->iommu->ir_table->base + index); - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return 0; } @@ -140,6 +148,7 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count) struct irq_2_iommu *irq_iommu; u16 index, start_index; unsigned int mask = 0; + unsigned long flags; int i; if (!count) @@ -169,7 +178,7 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count) return -1; } - spin_lock(&irq_2_ir_lock); + spin_lock_irqsave(&irq_2_ir_lock, flags); do { for (i = index; i < index + count; i++) if (table->base[i].present) @@ -181,7 +190,7 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count) index = (index + count) % INTR_REMAP_TABLE_ENTRIES; if (index == start_index) { - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); printk(KERN_ERR "can't allocate an IRTE\n"); return -1; } @@ -192,7 +201,7 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count) irq_iommu = irq_2_iommu_alloc(irq); if (!irq_iommu) { - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); printk(KERN_ERR "can't allocate irq_2_iommu\n"); return -1; } @@ -202,7 +211,7 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count) irq_iommu->sub_handle = 0; irq_iommu->irte_mask = mask; - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return index; } @@ -222,30 +231,32 @@ int map_irq_to_irte_handle(int irq, u16 *sub_handle) { int index; struct irq_2_iommu *irq_iommu; + unsigned long flags; - spin_lock(&irq_2_ir_lock); + spin_lock_irqsave(&irq_2_ir_lock, flags); irq_iommu = valid_irq_2_iommu(irq); if (!irq_iommu) { - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return -1; } *sub_handle = irq_iommu->sub_handle; index = irq_iommu->irte_index; - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return index; } int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle) { struct irq_2_iommu *irq_iommu; + unsigned long flags; - spin_lock(&irq_2_ir_lock); + spin_lock_irqsave(&irq_2_ir_lock, flags); irq_iommu = irq_2_iommu_alloc(irq); if (!irq_iommu) { - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); printk(KERN_ERR "can't allocate irq_2_iommu\n"); return -1; } @@ -255,7 +266,7 @@ int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle) irq_iommu->sub_handle = subhandle; irq_iommu->irte_mask = 0; - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return 0; } @@ -263,11 +274,12 @@ int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle) int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index) { struct irq_2_iommu *irq_iommu; + unsigned long flags; - spin_lock(&irq_2_ir_lock); + spin_lock_irqsave(&irq_2_ir_lock, flags); irq_iommu = valid_irq_2_iommu(irq); if (!irq_iommu) { - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return -1; } @@ -276,7 +288,7 @@ int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index) irq_iommu->sub_handle = 0; irq_2_iommu(irq)->irte_mask = 0; - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return 0; } @@ -288,11 +300,12 @@ int modify_irte(int irq, struct irte *irte_modified) struct irte *irte; struct intel_iommu *iommu; struct irq_2_iommu *irq_iommu; + unsigned long flags; - spin_lock(&irq_2_ir_lock); + spin_lock_irqsave(&irq_2_ir_lock, flags); irq_iommu = valid_irq_2_iommu(irq); if (!irq_iommu) { - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return -1; } @@ -301,11 +314,11 @@ 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 | (1 << 1)); + set_64bit((unsigned long *)irte, irte_modified->low); __iommu_flush_cache(iommu, irte, sizeof(*irte)); rc = qi_flush_iec(iommu, index, 0); - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return rc; } @@ -316,11 +329,12 @@ int flush_irte(int irq) int index; struct intel_iommu *iommu; struct irq_2_iommu *irq_iommu; + unsigned long flags; - spin_lock(&irq_2_ir_lock); + spin_lock_irqsave(&irq_2_ir_lock, flags); irq_iommu = valid_irq_2_iommu(irq); if (!irq_iommu) { - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return -1; } @@ -329,7 +343,7 @@ int flush_irte(int irq) index = irq_iommu->irte_index + irq_iommu->sub_handle; rc = qi_flush_iec(iommu, index, irq_iommu->irte_mask); - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return rc; } @@ -362,11 +376,12 @@ int free_irte(int irq) struct irte *irte; struct intel_iommu *iommu; struct irq_2_iommu *irq_iommu; + unsigned long flags; - spin_lock(&irq_2_ir_lock); + spin_lock_irqsave(&irq_2_ir_lock, flags); irq_iommu = valid_irq_2_iommu(irq); if (!irq_iommu) { - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return -1; } @@ -377,7 +392,7 @@ int free_irte(int irq) if (!irq_iommu->sub_handle) { for (i = 0; i < (1 << irq_iommu->irte_mask); i++) - set_64bit((unsigned long *)irte, 0); + set_64bit((unsigned long *)(irte + i), 0); rc = qi_flush_iec(iommu, index, irq_iommu->irte_mask); } @@ -386,7 +401,7 @@ int free_irte(int irq) irq_iommu->sub_handle = 0; irq_iommu->irte_mask = 0; - spin_unlock(&irq_2_ir_lock); + spin_unlock_irqrestore(&irq_2_ir_lock, flags); return rc; } @@ -406,6 +421,7 @@ static void iommu_set_intr_remapping(struct intel_iommu *iommu, int mode) /* Set interrupt-remapping table pointer */ cmd = iommu->gcmd | DMA_GCMD_SIRTP; + iommu->gcmd |= DMA_GCMD_SIRTP; writel(cmd, iommu->reg + DMAR_GCMD_REG); IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, @@ -438,12 +454,12 @@ static int setup_intr_remapping(struct intel_iommu *iommu, int mode) struct page *pages; ir_table = iommu->ir_table = kzalloc(sizeof(struct ir_table), - GFP_KERNEL); + GFP_ATOMIC); if (!iommu->ir_table) return -ENOMEM; - pages = alloc_pages(GFP_KERNEL | __GFP_ZERO, INTR_REMAP_PAGE_ORDER); + pages = alloc_pages(GFP_ATOMIC | __GFP_ZERO, INTR_REMAP_PAGE_ORDER); if (!pages) { printk(KERN_ERR "failed to allocate pages of order %d\n", @@ -458,11 +474,85 @@ static int setup_intr_remapping(struct intel_iommu *iommu, int mode) return 0; } +/* + * Disable Interrupt Remapping. + */ +static void iommu_disable_intr_remapping(struct intel_iommu *iommu) +{ + unsigned long flags; + u32 sts; + + if (!ecap_ir_support(iommu->ecap)) + return; + + /* + * global invalidation of interrupt entry cache before disabling + * interrupt-remapping. + */ + qi_global_iec(iommu); + + spin_lock_irqsave(&iommu->register_lock, flags); + + sts = dmar_readq(iommu->reg + DMAR_GSTS_REG); + if (!(sts & DMA_GSTS_IRES)) + goto end; + + iommu->gcmd &= ~DMA_GCMD_IRE; + writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG); + + IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, + readl, !(sts & DMA_GSTS_IRES), sts); + +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; int setup = 0; + for_each_drhd_unit(drhd) { + struct intel_iommu *iommu = drhd->iommu; + + /* + * If the queued invalidation is already initialized, + * shouldn't disable it. + */ + if (iommu->qi) + continue; + + /* + * Clear previous faults. + */ + dmar_fault(-1, iommu); + + /* + * Disable intr remapping and queued invalidation, if already + * enabled prior to OS handover. + */ + iommu_disable_intr_remapping(iommu); + + dmar_disable_qi(iommu); + } + /* * check for the Interrupt-remapping support */ @@ -586,3 +676,54 @@ int __init parse_ioapics_under_ir(void) return ir_supported; } + +void disable_intr_remapping(void) +{ + struct dmar_drhd_unit *drhd; + struct intel_iommu *iommu = NULL; + + /* + * Disable Interrupt-remapping for all the DRHD's now. + */ + for_each_iommu(iommu, drhd) { + if (!ecap_ir_support(iommu->ecap)) + continue; + + iommu_disable_intr_remapping(iommu); + } +} + +int reenable_intr_remapping(int eim) +{ + struct dmar_drhd_unit *drhd; + int setup = 0; + struct intel_iommu *iommu = NULL; + + for_each_iommu(iommu, drhd) + if (iommu->qi) + dmar_reenable_qi(iommu); + + /* + * Setup Interrupt-remapping for all the DRHD's now. + */ + for_each_iommu(iommu, drhd) { + if (!ecap_ir_support(iommu->ecap)) + continue; + + /* Set up interrupt remapping for iommu.*/ + iommu_set_intr_remapping(iommu, eim); + setup = 1; + } + + if (!setup) + goto error; + + return 0; + +error: + /* + * handle error condition gracefully here! + */ + return -1; +} + diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c new file mode 100644 index 00000000000..03c7706c0a0 --- /dev/null +++ b/drivers/pci/iov.c @@ -0,0 +1,683 @@ +/* + * drivers/pci/iov.c + * + * Copyright (C) 2009 Intel Corporation, Yu Zhao <yu.zhao@intel.com> + * + * PCI Express I/O Virtualization (IOV) support. + * Single Root IOV 1.0 + */ + +#include <linux/pci.h> +#include <linux/mutex.h> +#include <linux/string.h> +#include <linux/delay.h> +#include "pci.h" + +#define VIRTFN_ID_LEN 16 + +static inline u8 virtfn_bus(struct pci_dev *dev, int id) +{ + return dev->bus->number + ((dev->devfn + dev->sriov->offset + + dev->sriov->stride * id) >> 8); +} + +static inline u8 virtfn_devfn(struct pci_dev *dev, int id) +{ + return (dev->devfn + dev->sriov->offset + + dev->sriov->stride * id) & 0xff; +} + +static struct pci_bus *virtfn_add_bus(struct pci_bus *bus, int busnr) +{ + int rc; + struct pci_bus *child; + + if (bus->number == busnr) + return bus; + + child = pci_find_bus(pci_domain_nr(bus), busnr); + if (child) + return child; + + child = pci_add_new_bus(bus, NULL, busnr); + if (!child) + return NULL; + + child->subordinate = busnr; + child->dev.parent = bus->bridge; + rc = pci_bus_add_child(child); + if (rc) { + pci_remove_bus(child); + return NULL; + } + + return child; +} + +static void virtfn_remove_bus(struct pci_bus *bus, int busnr) +{ + struct pci_bus *child; + + if (bus->number == busnr) + return; + + child = pci_find_bus(pci_domain_nr(bus), busnr); + BUG_ON(!child); + + if (list_empty(&child->devices)) + pci_remove_bus(child); +} + +static int virtfn_add(struct pci_dev *dev, int id, int reset) +{ + int i; + int rc; + u64 size; + char buf[VIRTFN_ID_LEN]; + struct pci_dev *virtfn; + struct resource *res; + struct pci_sriov *iov = dev->sriov; + + virtfn = alloc_pci_dev(); + if (!virtfn) + return -ENOMEM; + + mutex_lock(&iov->dev->sriov->lock); + virtfn->bus = virtfn_add_bus(dev->bus, virtfn_bus(dev, id)); + if (!virtfn->bus) { + kfree(virtfn); + mutex_unlock(&iov->dev->sriov->lock); + return -ENOMEM; + } + virtfn->devfn = virtfn_devfn(dev, id); + virtfn->vendor = dev->vendor; + pci_read_config_word(dev, iov->pos + PCI_SRIOV_VF_DID, &virtfn->device); + pci_setup_device(virtfn); + virtfn->dev.parent = dev->dev.parent; + + for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) { + res = dev->resource + PCI_IOV_RESOURCES + i; + if (!res->parent) + continue; + virtfn->resource[i].name = pci_name(virtfn); + virtfn->resource[i].flags = res->flags; + size = resource_size(res); + do_div(size, iov->total); + virtfn->resource[i].start = res->start + size * id; + virtfn->resource[i].end = virtfn->resource[i].start + size - 1; + rc = request_resource(res, &virtfn->resource[i]); + BUG_ON(rc); + } + + if (reset) + __pci_reset_function(virtfn); + + pci_device_add(virtfn, virtfn->bus); + mutex_unlock(&iov->dev->sriov->lock); + + virtfn->physfn = pci_dev_get(dev); + virtfn->is_virtfn = 1; + + rc = pci_bus_add_device(virtfn); + if (rc) + goto failed1; + sprintf(buf, "virtfn%u", id); + rc = sysfs_create_link(&dev->dev.kobj, &virtfn->dev.kobj, buf); + if (rc) + goto failed1; + rc = sysfs_create_link(&virtfn->dev.kobj, &dev->dev.kobj, "physfn"); + if (rc) + goto failed2; + + kobject_uevent(&virtfn->dev.kobj, KOBJ_CHANGE); + + return 0; + +failed2: + sysfs_remove_link(&dev->dev.kobj, buf); +failed1: + pci_dev_put(dev); + mutex_lock(&iov->dev->sriov->lock); + pci_remove_bus_device(virtfn); + virtfn_remove_bus(dev->bus, virtfn_bus(dev, id)); + mutex_unlock(&iov->dev->sriov->lock); + + return rc; +} + +static void virtfn_remove(struct pci_dev *dev, int id, int reset) +{ + char buf[VIRTFN_ID_LEN]; + struct pci_bus *bus; + struct pci_dev *virtfn; + struct pci_sriov *iov = dev->sriov; + + bus = pci_find_bus(pci_domain_nr(dev->bus), virtfn_bus(dev, id)); + if (!bus) + return; + + virtfn = pci_get_slot(bus, virtfn_devfn(dev, id)); + if (!virtfn) + return; + + pci_dev_put(virtfn); + + if (reset) { + device_release_driver(&virtfn->dev); + __pci_reset_function(virtfn); + } + + sprintf(buf, "virtfn%u", id); + sysfs_remove_link(&dev->dev.kobj, buf); + sysfs_remove_link(&virtfn->dev.kobj, "physfn"); + + mutex_lock(&iov->dev->sriov->lock); + pci_remove_bus_device(virtfn); + virtfn_remove_bus(dev->bus, virtfn_bus(dev, id)); + mutex_unlock(&iov->dev->sriov->lock); + + pci_dev_put(dev); +} + +static int sriov_migration(struct pci_dev *dev) +{ + u16 status; + struct pci_sriov *iov = dev->sriov; + + if (!iov->nr_virtfn) + return 0; + + if (!(iov->cap & PCI_SRIOV_CAP_VFM)) + return 0; + + pci_read_config_word(dev, iov->pos + PCI_SRIOV_STATUS, &status); + if (!(status & PCI_SRIOV_STATUS_VFM)) + return 0; + + schedule_work(&iov->mtask); + + return 1; +} + +static void sriov_migration_task(struct work_struct *work) +{ + int i; + u8 state; + u16 status; + struct pci_sriov *iov = container_of(work, struct pci_sriov, mtask); + + for (i = iov->initial; i < iov->nr_virtfn; i++) { + state = readb(iov->mstate + i); + if (state == PCI_SRIOV_VFM_MI) { + writeb(PCI_SRIOV_VFM_AV, iov->mstate + i); + state = readb(iov->mstate + i); + if (state == PCI_SRIOV_VFM_AV) + virtfn_add(iov->self, i, 1); + } else if (state == PCI_SRIOV_VFM_MO) { + virtfn_remove(iov->self, i, 1); + writeb(PCI_SRIOV_VFM_UA, iov->mstate + i); + state = readb(iov->mstate + i); + if (state == PCI_SRIOV_VFM_AV) + virtfn_add(iov->self, i, 0); + } + } + + pci_read_config_word(iov->self, iov->pos + PCI_SRIOV_STATUS, &status); + status &= ~PCI_SRIOV_STATUS_VFM; + pci_write_config_word(iov->self, iov->pos + PCI_SRIOV_STATUS, status); +} + +static int sriov_enable_migration(struct pci_dev *dev, int nr_virtfn) +{ + int bir; + u32 table; + resource_size_t pa; + struct pci_sriov *iov = dev->sriov; + + if (nr_virtfn <= iov->initial) + return 0; + + pci_read_config_dword(dev, iov->pos + PCI_SRIOV_VFM, &table); + bir = PCI_SRIOV_VFM_BIR(table); + if (bir > PCI_STD_RESOURCE_END) + return -EIO; + + table = PCI_SRIOV_VFM_OFFSET(table); + if (table + nr_virtfn > pci_resource_len(dev, bir)) + return -EIO; + + pa = pci_resource_start(dev, bir) + table; + iov->mstate = ioremap(pa, nr_virtfn); + if (!iov->mstate) + return -ENOMEM; + + INIT_WORK(&iov->mtask, sriov_migration_task); + + iov->ctrl |= PCI_SRIOV_CTRL_VFM | PCI_SRIOV_CTRL_INTR; + pci_write_config_word(dev, iov->pos + PCI_SRIOV_CTRL, iov->ctrl); + + return 0; +} + +static void sriov_disable_migration(struct pci_dev *dev) +{ + struct pci_sriov *iov = dev->sriov; + + iov->ctrl &= ~(PCI_SRIOV_CTRL_VFM | PCI_SRIOV_CTRL_INTR); + pci_write_config_word(dev, iov->pos + PCI_SRIOV_CTRL, iov->ctrl); + + cancel_work_sync(&iov->mtask); + iounmap(iov->mstate); +} + +static int sriov_enable(struct pci_dev *dev, int nr_virtfn) +{ + int rc; + int i, j; + int nres; + u16 offset, stride, initial; + struct resource *res; + struct pci_dev *pdev; + struct pci_sriov *iov = dev->sriov; + + if (!nr_virtfn) + return 0; + + if (iov->nr_virtfn) + return -EINVAL; + + pci_read_config_word(dev, iov->pos + PCI_SRIOV_INITIAL_VF, &initial); + if (initial > iov->total || + (!(iov->cap & PCI_SRIOV_CAP_VFM) && (initial != iov->total))) + return -EIO; + + if (nr_virtfn < 0 || nr_virtfn > iov->total || + (!(iov->cap & PCI_SRIOV_CAP_VFM) && (nr_virtfn > initial))) + return -EINVAL; + + pci_write_config_word(dev, iov->pos + PCI_SRIOV_NUM_VF, nr_virtfn); + pci_read_config_word(dev, iov->pos + PCI_SRIOV_VF_OFFSET, &offset); + pci_read_config_word(dev, iov->pos + PCI_SRIOV_VF_STRIDE, &stride); + if (!offset || (nr_virtfn > 1 && !stride)) + return -EIO; + + nres = 0; + for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) { + res = dev->resource + PCI_IOV_RESOURCES + i; + if (res->parent) + nres++; + } + if (nres != iov->nres) { + dev_err(&dev->dev, "not enough MMIO resources for SR-IOV\n"); + return -ENOMEM; + } + + iov->offset = offset; + iov->stride = stride; + + if (virtfn_bus(dev, nr_virtfn - 1) > dev->bus->subordinate) { + dev_err(&dev->dev, "SR-IOV: bus number out of range\n"); + return -ENOMEM; + } + + if (iov->link != dev->devfn) { + pdev = pci_get_slot(dev->bus, iov->link); + if (!pdev) + return -ENODEV; + + pci_dev_put(pdev); + + if (!pdev->is_physfn) + return -ENODEV; + + rc = sysfs_create_link(&dev->dev.kobj, + &pdev->dev.kobj, "dep_link"); + if (rc) + return rc; + } + + iov->ctrl |= PCI_SRIOV_CTRL_VFE | PCI_SRIOV_CTRL_MSE; + pci_block_user_cfg_access(dev); + pci_write_config_word(dev, iov->pos + PCI_SRIOV_CTRL, iov->ctrl); + msleep(100); + pci_unblock_user_cfg_access(dev); + + iov->initial = initial; + if (nr_virtfn < initial) + initial = nr_virtfn; + + for (i = 0; i < initial; i++) { + rc = virtfn_add(dev, i, 0); + if (rc) + goto failed; + } + + if (iov->cap & PCI_SRIOV_CAP_VFM) { + rc = sriov_enable_migration(dev, nr_virtfn); + if (rc) + goto failed; + } + + kobject_uevent(&dev->dev.kobj, KOBJ_CHANGE); + iov->nr_virtfn = nr_virtfn; + + return 0; + +failed: + for (j = 0; j < i; j++) + virtfn_remove(dev, j, 0); + + iov->ctrl &= ~(PCI_SRIOV_CTRL_VFE | PCI_SRIOV_CTRL_MSE); + pci_block_user_cfg_access(dev); + pci_write_config_word(dev, iov->pos + PCI_SRIOV_CTRL, iov->ctrl); + ssleep(1); + pci_unblock_user_cfg_access(dev); + + if (iov->link != dev->devfn) + sysfs_remove_link(&dev->dev.kobj, "dep_link"); + + return rc; +} + +static void sriov_disable(struct pci_dev *dev) +{ + int i; + struct pci_sriov *iov = dev->sriov; + + if (!iov->nr_virtfn) + return; + + if (iov->cap & PCI_SRIOV_CAP_VFM) + sriov_disable_migration(dev); + + for (i = 0; i < iov->nr_virtfn; i++) + virtfn_remove(dev, i, 0); + + iov->ctrl &= ~(PCI_SRIOV_CTRL_VFE | PCI_SRIOV_CTRL_MSE); + pci_block_user_cfg_access(dev); + pci_write_config_word(dev, iov->pos + PCI_SRIOV_CTRL, iov->ctrl); + ssleep(1); + pci_unblock_user_cfg_access(dev); + + if (iov->link != dev->devfn) + sysfs_remove_link(&dev->dev.kobj, "dep_link"); + + iov->nr_virtfn = 0; +} + +static int sriov_init(struct pci_dev *dev, int pos) +{ + int i; + int rc; + int nres; + u32 pgsz; + u16 ctrl, total, offset, stride; + struct pci_sriov *iov; + struct resource *res; + struct pci_dev *pdev; + + if (dev->pcie_type != PCI_EXP_TYPE_RC_END && + dev->pcie_type != PCI_EXP_TYPE_ENDPOINT) + return -ENODEV; + + pci_read_config_word(dev, pos + PCI_SRIOV_CTRL, &ctrl); + if (ctrl & PCI_SRIOV_CTRL_VFE) { + pci_write_config_word(dev, pos + PCI_SRIOV_CTRL, 0); + ssleep(1); + } + + pci_read_config_word(dev, pos + PCI_SRIOV_TOTAL_VF, &total); + if (!total) + return 0; + + ctrl = 0; + list_for_each_entry(pdev, &dev->bus->devices, bus_list) + if (pdev->is_physfn) + goto found; + + pdev = NULL; + if (pci_ari_enabled(dev->bus)) + ctrl |= PCI_SRIOV_CTRL_ARI; + +found: + pci_write_config_word(dev, pos + PCI_SRIOV_CTRL, ctrl); + pci_write_config_word(dev, pos + PCI_SRIOV_NUM_VF, total); + pci_read_config_word(dev, pos + PCI_SRIOV_VF_OFFSET, &offset); + pci_read_config_word(dev, pos + PCI_SRIOV_VF_STRIDE, &stride); + if (!offset || (total > 1 && !stride)) + return -EIO; + + pci_read_config_dword(dev, pos + PCI_SRIOV_SUP_PGSIZE, &pgsz); + i = PAGE_SHIFT > 12 ? PAGE_SHIFT - 12 : 0; + pgsz &= ~((1 << i) - 1); + if (!pgsz) + return -EIO; + + pgsz &= ~(pgsz - 1); + pci_write_config_dword(dev, pos + PCI_SRIOV_SYS_PGSIZE, pgsz); + + nres = 0; + for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) { + res = dev->resource + PCI_IOV_RESOURCES + i; + i += __pci_read_base(dev, pci_bar_unknown, res, + pos + PCI_SRIOV_BAR + i * 4); + if (!res->flags) + continue; + if (resource_size(res) & (PAGE_SIZE - 1)) { + rc = -EIO; + goto failed; + } + res->end = res->start + resource_size(res) * total - 1; + nres++; + } + + iov = kzalloc(sizeof(*iov), GFP_KERNEL); + if (!iov) { + rc = -ENOMEM; + goto failed; + } + + iov->pos = pos; + iov->nres = nres; + iov->ctrl = ctrl; + iov->total = total; + iov->offset = offset; + iov->stride = stride; + iov->pgsz = pgsz; + 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 { + iov->dev = dev; + mutex_init(&iov->lock); + } + + dev->sriov = iov; + dev->is_physfn = 1; + + return 0; + +failed: + for (i = 0; i < PCI_SRIOV_NUM_BARS; i++) { + res = dev->resource + PCI_IOV_RESOURCES + i; + res->flags = 0; + } + + return rc; +} + +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 + pci_dev_put(dev->sriov->dev); + + kfree(dev->sriov); + dev->sriov = NULL; +} + +static void sriov_restore_state(struct pci_dev *dev) +{ + int i; + u16 ctrl; + struct pci_sriov *iov = dev->sriov; + + pci_read_config_word(dev, iov->pos + PCI_SRIOV_CTRL, &ctrl); + if (ctrl & PCI_SRIOV_CTRL_VFE) + return; + + for (i = PCI_IOV_RESOURCES; i <= PCI_IOV_RESOURCE_END; i++) + pci_update_resource(dev, i); + + pci_write_config_dword(dev, iov->pos + PCI_SRIOV_SYS_PGSIZE, iov->pgsz); + pci_write_config_word(dev, iov->pos + PCI_SRIOV_NUM_VF, iov->nr_virtfn); + pci_write_config_word(dev, iov->pos + PCI_SRIOV_CTRL, iov->ctrl); + if (iov->ctrl & PCI_SRIOV_CTRL_VFE) + msleep(100); +} + +/** + * pci_iov_init - initialize the IOV capability + * @dev: the PCI device + * + * Returns 0 on success, or negative on failure. + */ +int pci_iov_init(struct pci_dev *dev) +{ + int pos; + + if (!dev->is_pcie) + return -ENODEV; + + pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_SRIOV); + if (pos) + return sriov_init(dev, pos); + + return -ENODEV; +} + +/** + * pci_iov_release - release resources used by the IOV capability + * @dev: the PCI device + */ +void pci_iov_release(struct pci_dev *dev) +{ + if (dev->is_physfn) + sriov_release(dev); +} + +/** + * pci_iov_resource_bar - get position of the SR-IOV BAR + * @dev: the PCI device + * @resno: the resource number + * @type: the BAR type to be filled in + * + * Returns position of the BAR encapsulated in the SR-IOV capability. + */ +int pci_iov_resource_bar(struct pci_dev *dev, int resno, + enum pci_bar_type *type) +{ + if (resno < PCI_IOV_RESOURCES || resno > PCI_IOV_RESOURCE_END) + return 0; + + BUG_ON(!dev->is_physfn); + + *type = pci_bar_unknown; + + return dev->sriov->pos + PCI_SRIOV_BAR + + 4 * (resno - PCI_IOV_RESOURCES); +} + +/** + * pci_restore_iov_state - restore the state of the IOV capability + * @dev: the PCI device + */ +void pci_restore_iov_state(struct pci_dev *dev) +{ + if (dev->is_physfn) + sriov_restore_state(dev); +} + +/** + * pci_iov_bus_range - find bus range used by Virtual Function + * @bus: the PCI bus + * + * Returns max number of buses (exclude current one) used by Virtual + * Functions. + */ +int pci_iov_bus_range(struct pci_bus *bus) +{ + int max = 0; + u8 busnr; + struct pci_dev *dev; + + list_for_each_entry(dev, &bus->devices, bus_list) { + if (!dev->is_physfn) + continue; + busnr = virtfn_bus(dev, dev->sriov->total - 1); + if (busnr > max) + max = busnr; + } + + return max ? max - bus->number : 0; +} + +/** + * pci_enable_sriov - enable the SR-IOV capability + * @dev: the PCI device + * @nr_virtfn: number of virtual functions to enable + * + * Returns 0 on success, or negative on failure. + */ +int pci_enable_sriov(struct pci_dev *dev, int nr_virtfn) +{ + might_sleep(); + + if (!dev->is_physfn) + return -ENODEV; + + return sriov_enable(dev, nr_virtfn); +} +EXPORT_SYMBOL_GPL(pci_enable_sriov); + +/** + * pci_disable_sriov - disable the SR-IOV capability + * @dev: the PCI device + */ +void pci_disable_sriov(struct pci_dev *dev) +{ + might_sleep(); + + if (!dev->is_physfn) + return; + + sriov_disable(dev); +} +EXPORT_SYMBOL_GPL(pci_disable_sriov); + +/** + * pci_sriov_migration - notify SR-IOV core of Virtual Function Migration + * @dev: the PCI device + * + * Returns IRQ_HANDLED if the IRQ is handled, or IRQ_NONE if not. + * + * Physical Function driver is responsible to register IRQ handler using + * VF Migration Interrupt Message Number, and call this function when the + * interrupt is generated by the hardware. + */ +irqreturn_t pci_sriov_migration(struct pci_dev *dev) +{ + if (!dev->is_physfn) + return IRQ_NONE; + + return sriov_migration(dev) ? IRQ_HANDLED : IRQ_NONE; +} +EXPORT_SYMBOL_GPL(pci_sriov_migration); diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c index baba2eb5367..d9f06fbfa0b 100644 --- a/drivers/pci/msi.c +++ b/drivers/pci/msi.c @@ -27,65 +27,65 @@ static int pci_msi_enable = 1; /* Arch hooks */ -int __attribute__ ((weak)) -arch_msi_check_device(struct pci_dev *dev, int nvec, int type) +#ifndef arch_msi_check_device +int arch_msi_check_device(struct pci_dev *dev, int nvec, int type) { return 0; } +#endif -int __attribute__ ((weak)) -arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *entry) -{ - return 0; -} - -int __attribute__ ((weak)) -arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) +#ifndef arch_setup_msi_irqs +int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) { struct msi_desc *entry; int ret; + /* + * If an architecture wants to support multiple MSI, it needs to + * override arch_setup_msi_irqs() + */ + if (type == PCI_CAP_ID_MSI && nvec > 1) + return 1; + list_for_each_entry(entry, &dev->msi_list, list) { ret = arch_setup_msi_irq(dev, entry); - if (ret) + if (ret < 0) return ret; + if (ret > 0) + return -ENOSPC; } return 0; } +#endif -void __attribute__ ((weak)) arch_teardown_msi_irq(unsigned int irq) -{ - return; -} - -void __attribute__ ((weak)) -arch_teardown_msi_irqs(struct pci_dev *dev) +#ifndef arch_teardown_msi_irqs +void arch_teardown_msi_irqs(struct pci_dev *dev) { struct msi_desc *entry; list_for_each_entry(entry, &dev->msi_list, list) { - if (entry->irq != 0) - arch_teardown_msi_irq(entry->irq); + int i, nvec; + if (entry->irq == 0) + continue; + nvec = 1 << entry->msi_attrib.multiple; + for (i = 0; i < nvec; i++) + arch_teardown_msi_irq(entry->irq + i); } } +#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) @@ -111,27 +111,14 @@ static inline __attribute_const__ u32 msi_mask(unsigned x) return (1 << (1 << x)) - 1; } -static void msix_flush_writes(struct irq_desc *desc) +static inline __attribute_const__ u32 msi_capable_mask(u16 control) { - struct msi_desc *entry; + return msi_mask((control >> 1) & 7); +} - entry = get_irq_desc_msi(desc); - BUG_ON(!entry || !entry->dev); - switch (entry->msi_attrib.type) { - case PCI_CAP_ID_MSI: - /* nothing to do */ - break; - case PCI_CAP_ID_MSIX: - { - int offset = entry->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE + - PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET; - readl(entry->mask_base + offset); - break; - } - default: - BUG(); - break; - } +static inline __attribute_const__ u32 msi_enabled_mask(u16 control) +{ + return msi_mask((control >> 4) & 7); } /* @@ -139,53 +126,72 @@ static void msix_flush_writes(struct irq_desc *desc) * 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 int msi_set_mask_bits(struct irq_desc *desc, u32 mask, u32 flag) +static void msi_mask_irq(struct msi_desc *desc, u32 mask, u32 flag) { - struct msi_desc *entry; + u32 mask_bits = desc->masked; - entry = get_irq_desc_msi(desc); - BUG_ON(!entry || !entry->dev); - switch (entry->msi_attrib.type) { - case PCI_CAP_ID_MSI: - if (entry->msi_attrib.maskbit) { - int pos; - u32 mask_bits; - - pos = (long)entry->mask_base; - pci_read_config_dword(entry->dev, pos, &mask_bits); - mask_bits &= ~(mask); - mask_bits |= flag & mask; - pci_write_config_dword(entry->dev, pos, mask_bits); - } else { - return 0; - } - break; - case PCI_CAP_ID_MSIX: - { - int offset = entry->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE + - PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET; - writel(flag, entry->mask_base + offset); - readl(entry->mask_base + offset); - break; - } - default: - BUG(); - break; + if (!desc->msi_attrib.maskbit) + return; + + mask_bits &= ~mask; + mask_bits |= flag; + pci_write_config_dword(desc->dev, desc->mask_pos, mask_bits); + desc->masked = mask_bits; +} + +/* + * This internal function does not flush PCI writes to the device. + * All users must ensure that they read from the device before either + * assuming that the device state is up to date, or returning out of this + * 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) +{ + u32 mask_bits = desc->masked; + unsigned offset = desc->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE + + PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET; + mask_bits &= ~1; + mask_bits |= flag; + writel(mask_bits, desc->mask_base + offset); + desc->masked = mask_bits; +} + +static void msi_set_mask_bit(unsigned irq, u32 flag) +{ + struct msi_desc *desc = get_irq_msi(irq); + + if (desc->msi_attrib.is_msix) { + msix_mask_irq(desc, flag); + readl(desc->mask_base); /* Flush write to device */ + } else { + unsigned offset = irq - desc->dev->irq; + msi_mask_irq(desc, 1 << offset, flag << offset); } - entry->msi_attrib.masked = !!flag; - return 1; +} + +void mask_msi_irq(unsigned int irq) +{ + msi_set_mask_bit(irq, 1); +} + +void unmask_msi_irq(unsigned int irq) +{ + msi_set_mask_bit(irq, 0); } void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) { struct msi_desc *entry = get_irq_desc_msi(desc); - switch(entry->msi_attrib.type) { - case PCI_CAP_ID_MSI: - { + if (entry->msi_attrib.is_msix) { + 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); + } else { struct pci_dev *dev = entry->dev; int pos = entry->msi_attrib.pos; u16 data; @@ -201,21 +207,6 @@ void read_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) pci_read_config_word(dev, msi_data_reg(pos, 0), &data); } msg->data = data; - break; - } - case PCI_CAP_ID_MSIX: - { - void __iomem *base; - 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); - break; - } - default: - BUG(); } } @@ -229,11 +220,25 @@ void read_msi_msg(unsigned int irq, struct msi_msg *msg) void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) { struct msi_desc *entry = get_irq_desc_msi(desc); - switch (entry->msi_attrib.type) { - case PCI_CAP_ID_MSI: - { + if (entry->msi_attrib.is_msix) { + void __iomem *base; + 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); + } else { struct pci_dev *dev = entry->dev; int pos = entry->msi_attrib.pos; + u16 msgctl; + + pci_read_config_word(dev, msi_control_reg(pos), &msgctl); + msgctl &= ~PCI_MSI_FLAGS_QSIZE; + msgctl |= entry->msi_attrib.multiple << 4; + pci_write_config_word(dev, msi_control_reg(pos), msgctl); pci_write_config_dword(dev, msi_lower_address_reg(pos), msg->address_lo); @@ -246,23 +251,6 @@ void write_msi_msg_desc(struct irq_desc *desc, struct msi_msg *msg) pci_write_config_word(dev, msi_data_reg(pos, 0), msg->data); } - break; - } - case PCI_CAP_ID_MSIX: - { - void __iomem *base; - 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); - break; - } - default: - BUG(); } entry->msg = *msg; } @@ -274,37 +262,18 @@ void write_msi_msg(unsigned int irq, struct msi_msg *msg) write_msi_msg_desc(desc, msg); } -void mask_msi_irq(unsigned int irq) -{ - struct irq_desc *desc = irq_to_desc(irq); - - msi_set_mask_bits(desc, 1, 1); - msix_flush_writes(desc); -} - -void unmask_msi_irq(unsigned int irq) -{ - struct irq_desc *desc = irq_to_desc(irq); - - msi_set_mask_bits(desc, 1, 0); - msix_flush_writes(desc); -} - static int msi_free_irqs(struct pci_dev* dev); -static struct msi_desc* alloc_msi_entry(void) +static struct msi_desc *alloc_msi_entry(struct pci_dev *dev) { - struct msi_desc *entry; - - entry = kzalloc(sizeof(struct msi_desc), GFP_KERNEL); - if (!entry) + struct msi_desc *desc = kzalloc(sizeof(*desc), GFP_KERNEL); + if (!desc) return NULL; - INIT_LIST_HEAD(&entry->list); - entry->irq = 0; - entry->dev = NULL; + INIT_LIST_HEAD(&desc->list); + desc->dev = dev; - return entry; + return desc; } static void pci_intx_for_msi(struct pci_dev *dev, int enable) @@ -326,17 +295,13 @@ 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); - if (entry->msi_attrib.maskbit) { - struct irq_desc *desc = irq_to_desc(dev->irq); - msi_set_mask_bits(desc, entry->msi_attrib.maskbits_mask, - entry->msi_attrib.masked); - } pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &control); + msi_mask_irq(entry, msi_capable_mask(control), entry->masked); control &= ~PCI_MSI_FLAGS_QSIZE; - control |= PCI_MSI_FLAGS_ENABLE; + control |= (entry->msi_attrib.multiple << 4) | PCI_MSI_FLAGS_ENABLE; pci_write_config_word(dev, pos + PCI_MSI_FLAGS, control); } @@ -348,23 +313,22 @@ 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) { - struct irq_desc *desc = irq_to_desc(entry->irq); write_msi_msg(entry->irq, &entry->msg); - msi_set_mask_bits(desc, 1, entry->msi_attrib.masked); + 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); } @@ -378,52 +342,48 @@ EXPORT_SYMBOL_GPL(pci_restore_msi_state); /** * msi_capability_init - configure device's MSI capability structure * @dev: pointer to the pci_dev data structure of MSI device function + * @nvec: number of interrupts to allocate * - * Setup the MSI capability structure of device function with a single - * MSI irq, regardless of device function is capable of handling - * multiple messages. A return of zero indicates the successful setup - * of an entry zero with the new MSI irq or non-zero for otherwise. - **/ -static int msi_capability_init(struct pci_dev *dev) + * Setup the MSI capability structure of the device with the requested + * number of interrupts. A return value of zero indicates the successful + * setup of an entry with the new MSI irq. A negative return value indicates + * an error, and a positive return value indicates the number of interrupts + * which could have been allocated. + */ +static int msi_capability_init(struct pci_dev *dev, int nvec) { struct msi_desc *entry; int pos, ret; u16 control; - - msi_set_enable(dev, 0); /* Ensure msi is disabled as I set it up */ + unsigned mask; 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(); + entry = alloc_msi_entry(dev); if (!entry) return -ENOMEM; - entry->msi_attrib.type = PCI_CAP_ID_MSI; + entry->msi_attrib.is_msix = 0; entry->msi_attrib.is_64 = is_64bit_address(control); entry->msi_attrib.entry_nr = 0; entry->msi_attrib.maskbit = is_mask_bit_support(control); - entry->msi_attrib.masked = 1; entry->msi_attrib.default_irq = dev->irq; /* Save IOAPIC IRQ */ entry->msi_attrib.pos = pos; - entry->dev = dev; - if (entry->msi_attrib.maskbit) { - unsigned int base, maskbits, temp; - - base = msi_mask_bits_reg(pos, entry->msi_attrib.is_64); - entry->mask_base = (void __iomem *)(long)base; - - /* All MSIs are unmasked by default, Mask them all */ - pci_read_config_dword(dev, base, &maskbits); - temp = msi_mask((control & PCI_MSI_FLAGS_QMASK) >> 1); - maskbits |= temp; - pci_write_config_dword(dev, base, maskbits); - entry->msi_attrib.maskbits_mask = temp; - } + + 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); + mask = msi_capable_mask(control); + msi_mask_irq(entry, mask, mask); + list_add_tail(&entry->list, &dev->msi_list); /* Configure MSI capability structure */ - ret = arch_setup_msi_irqs(dev, 1, PCI_CAP_ID_MSI); + ret = arch_setup_msi_irqs(dev, nvec, PCI_CAP_ID_MSI); if (ret) { msi_free_irqs(dev); return ret; @@ -431,7 +391,7 @@ static int msi_capability_init(struct pci_dev *dev) /* 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; @@ -459,11 +419,14 @@ 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); @@ -474,28 +437,26 @@ 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(); + entry = alloc_msi_entry(dev); if (!entry) break; j = entries[i].entry; - entry->msi_attrib.type = PCI_CAP_ID_MSIX; + entry->msi_attrib.is_msix = 1; entry->msi_attrib.is_64 = 1; entry->msi_attrib.entry_nr = j; - entry->msi_attrib.maskbit = 1; - entry->msi_attrib.masked = 1; entry->msi_attrib.default_irq = dev->irq; entry->msi_attrib.pos = pos; - entry->dev = dev; entry->mask_base = base; list_add_tail(&entry->list, &dev->msi_list); } ret = arch_setup_msi_irqs(dev, nvec, PCI_CAP_ID_MSIX); - if (ret) { + if (ret < 0) { + /* If we had some success report the number of irqs + * we succeeded in setting up. */ int avail = 0; list_for_each_entry(entry, &dev->msi_list, list) { if (entry->irq != 0) { @@ -503,27 +464,41 @@ static int msix_capability_init(struct pci_dev *dev, } } - msi_free_irqs(dev); + if (avail != 0) + ret = avail; + } - /* If we had some success report the number of irqs - * we succeeded in setting up. - */ - if (avail == 0) - avail = ret; - return avail; + if (ret) { + msi_free_irqs(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_OFFSET); + 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; + control &= ~PCI_MSIX_FLAGS_MASKALL; + pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control); + return 0; } @@ -575,61 +550,75 @@ static int pci_msi_check_device(struct pci_dev* dev, int nvec, int type) } /** - * pci_enable_msi - configure device's MSI capability structure - * @dev: pointer to the pci_dev data structure of MSI device function + * pci_enable_msi_block - configure device's MSI capability structure + * @dev: device to configure + * @nvec: number of interrupts to configure * - * Setup the MSI capability structure of device function with - * a single MSI irq upon its software driver call to request for - * MSI mode enabled on its hardware device function. A return of zero - * indicates the successful setup of an entry zero with the new MSI - * irq or non-zero for otherwise. - **/ -int pci_enable_msi(struct pci_dev* dev) + * Allocate IRQs for a device with the MSI capability. + * This function returns a negative errno if an error occurs. If it + * is unable to allocate the number of interrupts requested, it returns + * the number of interrupts it might be able to allocate. If it successfully + * allocates at least the number of interrupts requested, it returns 0 and + * updates the @dev's irq member to the lowest new interrupt number; the + * other interrupt numbers allocated to this device are consecutive. + */ +int pci_enable_msi_block(struct pci_dev *dev, unsigned int nvec) { - int status; + int status, pos, maxvec; + u16 msgctl; + + pos = pci_find_capability(dev, PCI_CAP_ID_MSI); + if (!pos) + return -EINVAL; + pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &msgctl); + maxvec = 1 << ((msgctl & PCI_MSI_FLAGS_QMASK) >> 1); + if (nvec > maxvec) + return maxvec; - status = pci_msi_check_device(dev, 1, PCI_CAP_ID_MSI); + status = pci_msi_check_device(dev, nvec, PCI_CAP_ID_MSI); if (status) return status; WARN_ON(!!dev->msi_enabled); - /* Check whether driver already requested for MSI-X irqs */ + /* Check whether driver already requested MSI-X irqs */ if (dev->msix_enabled) { dev_info(&dev->dev, "can't enable MSI " "(MSI-X already enabled)\n"); return -EINVAL; } - status = msi_capability_init(dev); + + status = msi_capability_init(dev, nvec); return status; } -EXPORT_SYMBOL(pci_enable_msi); +EXPORT_SYMBOL(pci_enable_msi_block); -void pci_msi_shutdown(struct pci_dev* dev) +void pci_msi_shutdown(struct pci_dev *dev) { - struct msi_desc *entry; + 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)); - entry = list_entry(dev->msi_list.next, struct msi_desc, list); - /* Return the the pci reset with msi irqs unmasked */ - if (entry->msi_attrib.maskbit) { - u32 mask = entry->msi_attrib.maskbits_mask; - struct irq_desc *desc = irq_to_desc(dev->irq); - msi_set_mask_bits(desc, mask, ~mask); - } - if (!entry->dev || entry->msi_attrib.type != PCI_CAP_ID_MSI) - return; + pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &ctrl); + mask = msi_capable_mask(ctrl); + msi_mask_irq(desc, mask, ~mask); /* Restore dev->irq to its default pin-assertion irq */ - dev->irq = entry->msi_attrib.default_irq; + dev->irq = desc->msi_attrib.default_irq; } + void pci_disable_msi(struct pci_dev* dev) { struct msi_desc *entry; @@ -640,7 +629,7 @@ void pci_disable_msi(struct pci_dev* dev) pci_msi_shutdown(dev); entry = list_entry(dev->msi_list.next, struct msi_desc, list); - if (!entry->dev || entry->msi_attrib.type != PCI_CAP_ID_MSI) + if (entry->msi_attrib.is_msix) return; msi_free_irqs(dev); @@ -652,18 +641,19 @@ static int msi_free_irqs(struct pci_dev* dev) struct msi_desc *entry, *tmp; list_for_each_entry(entry, &dev->msi_list, list) { - if (entry->irq) - BUG_ON(irq_has_action(entry->irq)); + int i, nvec; + if (!entry->irq) + continue; + nvec = 1 << entry->msi_attrib.multiple; + for (i = 0; i < nvec; i++) + BUG_ON(irq_has_action(entry->irq + i)); } arch_teardown_msi_irqs(dev); list_for_each_entry_safe(entry, tmp, &dev->msi_list, list) { - if (entry->msi_attrib.type == PCI_CAP_ID_MSIX) { - writel(1, entry->mask_base + entry->msi_attrib.entry_nr - * PCI_MSIX_ENTRY_SIZE - + PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET); - + if (entry->msi_attrib.is_msix) { + msix_mask_irq(entry, 1); if (list_is_last(&entry->list, &dev->msi_list)) iounmap(entry->mask_base); } @@ -675,6 +665,23 @@ static int msi_free_irqs(struct pci_dev* dev) } /** + * pci_msix_table_size - return the number of device's MSI-X table entries + * @dev: pointer to the pci_dev data structure of MSI-X device function + */ +int pci_msix_table_size(struct pci_dev *dev) +{ + int pos; + u16 control; + + pos = pci_find_capability(dev, PCI_CAP_ID_MSIX); + if (!pos) + return 0; + + pci_read_config_word(dev, msi_control_reg(pos), &control); + return multi_msix_capable(control); +} + +/** * pci_enable_msix - configure device's MSI-X capability structure * @dev: pointer to the pci_dev data structure of MSI-X device function * @entries: pointer to an array of MSI-X entries @@ -686,14 +693,13 @@ static int msi_free_irqs(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) { - int status, pos, nr_entries; + int status, nr_entries; int i, j; - u16 control; if (!entries) return -EINVAL; @@ -702,11 +708,9 @@ int pci_enable_msix(struct pci_dev* dev, struct msix_entry *entries, int nvec) if (status) return status; - pos = pci_find_capability(dev, PCI_CAP_ID_MSIX); - pci_read_config_word(dev, msi_control_reg(pos), &control); - nr_entries = multi_msix_capable(control); + 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++) { diff --git a/drivers/pci/msi.h b/drivers/pci/msi.h index 3898f523714..a0662842550 100644 --- a/drivers/pci/msi.h +++ b/drivers/pci/msi.h @@ -16,27 +16,15 @@ #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 -#define multi_msi_capable(control) \ - (1 << ((control & PCI_MSI_FLAGS_QMASK) >> 1)) -#define multi_msi_enable(control, num) \ - control |= (((num >> 1) << 4) & PCI_MSI_FLAGS_QSIZE); + (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 msi_enable(control, num) multi_msi_enable(control, num); \ - control |= PCI_MSI_FLAGS_ENABLE #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-acpi.c b/drivers/pci/pci-acpi.c index deea8a187eb..ea15b053745 100644 --- a/drivers/pci/pci-acpi.c +++ b/drivers/pci/pci-acpi.c @@ -18,221 +18,6 @@ #include <linux/pci-acpi.h> #include "pci.h" -struct acpi_osc_data { - acpi_handle handle; - u32 support_set; - u32 control_set; - u32 control_query; - int is_queried; - struct list_head sibiling; -}; -static LIST_HEAD(acpi_osc_data_list); - -struct acpi_osc_args { - u32 capbuf[3]; -}; - -static DEFINE_MUTEX(pci_acpi_lock); - -static struct acpi_osc_data *acpi_get_osc_data(acpi_handle handle) -{ - struct acpi_osc_data *data; - - list_for_each_entry(data, &acpi_osc_data_list, sibiling) { - if (data->handle == handle) - return data; - } - data = kzalloc(sizeof(*data), GFP_KERNEL); - if (!data) - return NULL; - INIT_LIST_HEAD(&data->sibiling); - data->handle = handle; - list_add_tail(&data->sibiling, &acpi_osc_data_list); - return data; -} - -static u8 OSC_UUID[16] = {0x5B, 0x4D, 0xDB, 0x33, 0xF7, 0x1F, 0x1C, 0x40, - 0x96, 0x57, 0x74, 0x41, 0xC0, 0x3D, 0xD7, 0x66}; - -static acpi_status acpi_run_osc(acpi_handle handle, - struct acpi_osc_args *osc_args, u32 *retval) -{ - acpi_status status; - struct acpi_object_list input; - union acpi_object in_params[4]; - struct acpi_buffer output = {ACPI_ALLOCATE_BUFFER, NULL}; - union acpi_object *out_obj; - u32 errors, flags = osc_args->capbuf[OSC_QUERY_TYPE]; - - /* Setting up input parameters */ - input.count = 4; - input.pointer = in_params; - in_params[0].type = ACPI_TYPE_BUFFER; - in_params[0].buffer.length = 16; - in_params[0].buffer.pointer = OSC_UUID; - in_params[1].type = ACPI_TYPE_INTEGER; - in_params[1].integer.value = 1; - in_params[2].type = ACPI_TYPE_INTEGER; - in_params[2].integer.value = 3; - in_params[3].type = ACPI_TYPE_BUFFER; - in_params[3].buffer.length = 12; - in_params[3].buffer.pointer = (u8 *)osc_args->capbuf; - - status = acpi_evaluate_object(handle, "_OSC", &input, &output); - if (ACPI_FAILURE(status)) - return status; - - if (!output.length) - return AE_NULL_OBJECT; - - out_obj = output.pointer; - if (out_obj->type != ACPI_TYPE_BUFFER) { - printk(KERN_DEBUG "Evaluate _OSC returns wrong type\n"); - status = AE_TYPE; - goto out_kfree; - } - /* Need to ignore the bit0 in result code */ - errors = *((u32 *)out_obj->buffer.pointer) & ~(1 << 0); - if (errors) { - if (errors & OSC_REQUEST_ERROR) - printk(KERN_DEBUG "_OSC request fails\n"); - if (errors & OSC_INVALID_UUID_ERROR) - printk(KERN_DEBUG "_OSC invalid UUID\n"); - if (errors & OSC_INVALID_REVISION_ERROR) - printk(KERN_DEBUG "_OSC invalid revision\n"); - if (errors & OSC_CAPABILITIES_MASK_ERROR) { - if (flags & OSC_QUERY_ENABLE) - goto out_success; - printk(KERN_DEBUG "_OSC FW not grant req. control\n"); - status = AE_SUPPORT; - goto out_kfree; - } - status = AE_ERROR; - goto out_kfree; - } -out_success: - *retval = *((u32 *)(out_obj->buffer.pointer + 8)); - status = AE_OK; - -out_kfree: - kfree(output.pointer); - return status; -} - -static acpi_status __acpi_query_osc(u32 flags, struct acpi_osc_data *osc_data) -{ - acpi_status status; - u32 support_set, result; - struct acpi_osc_args osc_args; - - /* do _OSC query for all possible controls */ - support_set = osc_data->support_set | (flags & OSC_SUPPORT_MASKS); - osc_args.capbuf[OSC_QUERY_TYPE] = OSC_QUERY_ENABLE; - osc_args.capbuf[OSC_SUPPORT_TYPE] = support_set; - osc_args.capbuf[OSC_CONTROL_TYPE] = OSC_CONTROL_MASKS; - - status = acpi_run_osc(osc_data->handle, &osc_args, &result); - if (ACPI_SUCCESS(status)) { - osc_data->support_set = support_set; - osc_data->control_query = result; - osc_data->is_queried = 1; - } - - return status; -} - -/* - * pci_acpi_osc_support: Invoke _OSC indicating support for the given feature - * @flags: Bitmask of flags to support - * - * See the ACPI spec for the definition of the flags - */ -int pci_acpi_osc_support(acpi_handle handle, u32 flags) -{ - acpi_status status; - acpi_handle tmp; - struct acpi_osc_data *osc_data; - int rc = 0; - - status = acpi_get_handle(handle, "_OSC", &tmp); - if (ACPI_FAILURE(status)) - return -ENOTTY; - - mutex_lock(&pci_acpi_lock); - osc_data = acpi_get_osc_data(handle); - if (!osc_data) { - printk(KERN_ERR "acpi osc data array is full\n"); - rc = -ENOMEM; - goto out; - } - - __acpi_query_osc(flags, osc_data); -out: - mutex_unlock(&pci_acpi_lock); - return rc; -} - -/** - * pci_osc_control_set - commit requested control to Firmware - * @handle: acpi_handle for the target ACPI object - * @flags: driver's requested control bits - * - * Attempt to take control from Firmware on requested control bits. - **/ -acpi_status pci_osc_control_set(acpi_handle handle, u32 flags) -{ - acpi_status status; - u32 control_req, control_set, result; - acpi_handle tmp; - struct acpi_osc_data *osc_data; - struct acpi_osc_args osc_args; - - status = acpi_get_handle(handle, "_OSC", &tmp); - if (ACPI_FAILURE(status)) - return status; - - mutex_lock(&pci_acpi_lock); - osc_data = acpi_get_osc_data(handle); - if (!osc_data) { - printk(KERN_ERR "acpi osc data array is full\n"); - status = AE_ERROR; - goto out; - } - - control_req = (flags & OSC_CONTROL_MASKS); - if (!control_req) { - status = AE_TYPE; - goto out; - } - - /* No need to evaluate _OSC if the control was already granted. */ - if ((osc_data->control_set & control_req) == control_req) - goto out; - - if (!osc_data->is_queried) { - status = __acpi_query_osc(osc_data->support_set, osc_data); - if (ACPI_FAILURE(status)) - goto out; - } - - if ((osc_data->control_query & control_req) != control_req) { - status = AE_SUPPORT; - goto out; - } - - control_set = osc_data->control_set | control_req; - osc_args.capbuf[OSC_QUERY_TYPE] = 0; - osc_args.capbuf[OSC_SUPPORT_TYPE] = osc_data->support_set; - osc_args.capbuf[OSC_CONTROL_TYPE] = control_set; - status = acpi_run_osc(handle, &osc_args, &result); - if (ACPI_SUCCESS(status)) - osc_data->control_set = result; -out: - mutex_unlock(&pci_acpi_lock); - return status; -} -EXPORT_SYMBOL(pci_osc_control_set); - /* * _SxD returns the D-state with the highest power * (lowest D-state number) supported in the S-state "x". @@ -386,12 +171,12 @@ static int __init acpi_pci_init(void) { int ret; - if (acpi_gbl_FADT.boot_flags & BAF_MSI_NOT_SUPPORTED) { + if (acpi_gbl_FADT.boot_flags & ACPI_FADT_NO_MSI) { printk(KERN_INFO"ACPI FADT declares the system doesn't support MSI, so disable it\n"); pci_no_msi(); } - if (acpi_gbl_FADT.boot_flags & BAF_PCIE_ASPM_CONTROL) { + if (acpi_gbl_FADT.boot_flags & ACPI_FADT_NO_ASPM) { printk(KERN_INFO"ACPI FADT declares the system doesn't support PCIe ASPM, so disable it\n"); pcie_no_aspm(); } diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 93eac142358..d76c4c85367 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -99,6 +99,52 @@ store_new_id(struct device_driver *driver, const char *buf, size_t count) } static DRIVER_ATTR(new_id, S_IWUSR, NULL, store_new_id); +/** + * store_remove_id - remove a PCI device ID from this driver + * @driver: target device driver + * @buf: buffer for scanning device ID data + * @count: input size + * + * Removes a dynamic pci device ID to this driver. + */ +static ssize_t +store_remove_id(struct device_driver *driver, const char *buf, size_t count) +{ + struct pci_dynid *dynid, *n; + struct pci_driver *pdrv = to_pci_driver(driver); + __u32 vendor, device, subvendor = PCI_ANY_ID, + subdevice = PCI_ANY_ID, class = 0, class_mask = 0; + int fields = 0; + int retval = -ENODEV; + + fields = sscanf(buf, "%x %x %x %x %x %x", + &vendor, &device, &subvendor, &subdevice, + &class, &class_mask); + if (fields < 2) + return -EINVAL; + + spin_lock(&pdrv->dynids.lock); + list_for_each_entry_safe(dynid, n, &pdrv->dynids.list, node) { + struct pci_device_id *id = &dynid->id; + if ((id->vendor == vendor) && + (id->device == device) && + (subvendor == PCI_ANY_ID || id->subvendor == subvendor) && + (subdevice == PCI_ANY_ID || id->subdevice == subdevice) && + !((id->class ^ class) & class_mask)) { + list_del(&dynid->node); + kfree(dynid); + retval = 0; + break; + } + } + spin_unlock(&pdrv->dynids.lock); + + if (retval) + return retval; + return count; +} +static DRIVER_ATTR(remove_id, S_IWUSR, NULL, store_remove_id); + static void pci_free_dynids(struct pci_driver *drv) { @@ -125,6 +171,20 @@ static void pci_remove_newid_file(struct pci_driver *drv) { driver_remove_file(&drv->driver, &driver_attr_new_id); } + +static int +pci_create_removeid_file(struct pci_driver *drv) +{ + int error = 0; + if (drv->probe != NULL) + error = driver_create_file(&drv->driver,&driver_attr_remove_id); + return error; +} + +static void pci_remove_removeid_file(struct pci_driver *drv) +{ + driver_remove_file(&drv->driver, &driver_attr_remove_id); +} #else /* !CONFIG_HOTPLUG */ static inline void pci_free_dynids(struct pci_driver *drv) {} static inline int pci_create_newid_file(struct pci_driver *drv) @@ -132,6 +192,11 @@ static inline int pci_create_newid_file(struct pci_driver *drv) return 0; } static inline void pci_remove_newid_file(struct pci_driver *drv) {} +static inline int pci_create_removeid_file(struct pci_driver *drv) +{ + return 0; +} +static inline void pci_remove_removeid_file(struct pci_driver *drv) {} #endif /** @@ -212,10 +277,9 @@ static int pci_call_probe(struct pci_driver *drv, struct pci_dev *dev, node = dev_to_node(&dev->dev); if (node >= 0) { int cpu; - node_to_cpumask_ptr(nodecpumask, node); get_online_cpus(); - cpu = cpumask_any_and(nodecpumask, cpu_online_mask); + cpu = cpumask_any_and(cpumask_of_node(node), cpu_online_mask); if (cpu < nr_cpu_ids) error = work_on_cpu(cpu, local_pci_probe, &ddi); else @@ -352,53 +416,60 @@ static int pci_legacy_suspend(struct device *dev, pm_message_t state) { struct pci_dev * pci_dev = to_pci_dev(dev); struct pci_driver * drv = pci_dev->driver; - int i = 0; + + pci_dev->state_saved = false; if (drv && drv->suspend) { pci_power_t prev = pci_dev->current_state; + int error; - pci_dev->state_saved = false; - - i = drv->suspend(pci_dev, state); - suspend_report_result(drv->suspend, i); - if (i) - return i; - - if (pci_dev->state_saved) - goto Fixup; + error = drv->suspend(pci_dev, state); + suspend_report_result(drv->suspend, error); + if (error) + return error; - if (pci_dev->current_state != PCI_D0 + if (!pci_dev->state_saved && pci_dev->current_state != PCI_D0 && pci_dev->current_state != PCI_UNKNOWN) { WARN_ONCE(pci_dev->current_state != prev, "PCI PM: Device state not saved by %pF\n", drv->suspend); - goto Fixup; } } - pci_save_state(pci_dev); - /* - * This is for compatibility with existing code with legacy PM support. - */ - pci_pm_set_unknown_state(pci_dev); - - Fixup: pci_fixup_device(pci_fixup_suspend, pci_dev); - return i; + return 0; } static int pci_legacy_suspend_late(struct device *dev, pm_message_t state) { struct pci_dev * pci_dev = to_pci_dev(dev); struct pci_driver * drv = pci_dev->driver; - int i = 0; if (drv && drv->suspend_late) { - i = drv->suspend_late(pci_dev, state); - suspend_report_result(drv->suspend_late, i); + pci_power_t prev = pci_dev->current_state; + int error; + + error = drv->suspend_late(pci_dev, state); + suspend_report_result(drv->suspend_late, error); + if (error) + return error; + + if (!pci_dev->state_saved && pci_dev->current_state != PCI_D0 + && pci_dev->current_state != PCI_UNKNOWN) { + WARN_ONCE(pci_dev->current_state != prev, + "PCI PM: Device state not saved by %pF\n", + drv->suspend_late); + return 0; + } } - return i; + + if (!pci_dev->state_saved) + pci_save_state(pci_dev); + + pci_pm_set_unknown_state(pci_dev); + + return 0; } static int pci_legacy_resume_early(struct device *dev) @@ -423,6 +494,23 @@ static int pci_legacy_resume(struct device *dev) /* Auxiliary functions used by the new power management framework */ +/** + * pci_restore_standard_config - restore standard config registers of PCI device + * @pci_dev: PCI device to handle + */ +static int pci_restore_standard_config(struct pci_dev *pci_dev) +{ + pci_update_current_state(pci_dev, PCI_UNKNOWN); + + if (pci_dev->current_state != PCI_D0) { + int error = pci_set_power_state(pci_dev, PCI_D0); + if (error) + return error; + } + + return pci_dev->state_saved ? pci_restore_state(pci_dev) : 0; +} + static void pci_pm_default_resume_noirq(struct pci_dev *pci_dev) { pci_restore_standard_config(pci_dev); @@ -443,7 +531,6 @@ static void pci_pm_default_suspend(struct pci_dev *pci_dev) /* Disable non-bridge devices without PM support */ if (!pci_is_bridge(pci_dev)) pci_disable_enabled_device(pci_dev); - pci_save_state(pci_dev); } static bool pci_has_legacy_pm_support(struct pci_dev *pci_dev) @@ -493,13 +580,13 @@ static int pci_pm_suspend(struct device *dev) if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_suspend(dev, PMSG_SUSPEND); + pci_dev->state_saved = false; + if (!pm) { pci_pm_default_suspend(pci_dev); goto Fixup; } - pci_dev->state_saved = false; - if (pm->suspend) { pci_power_t prev = pci_dev->current_state; int error; @@ -509,24 +596,14 @@ static int pci_pm_suspend(struct device *dev) if (error) return error; - if (pci_dev->state_saved) - goto Fixup; - - if (pci_dev->current_state != PCI_D0 + if (!pci_dev->state_saved && pci_dev->current_state != PCI_D0 && pci_dev->current_state != PCI_UNKNOWN) { WARN_ONCE(pci_dev->current_state != prev, "PCI PM: State of device not saved by %pF\n", pm->suspend); - goto Fixup; } } - if (!pci_dev->state_saved) { - pci_save_state(pci_dev); - if (!pci_is_bridge(pci_dev)) - pci_prepare_to_sleep(pci_dev); - } - Fixup: pci_fixup_device(pci_fixup_suspend, pci_dev); @@ -536,21 +613,43 @@ 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 device_driver *drv = dev->driver; - int error = 0; + 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); - if (drv && drv->pm && drv->pm->suspend_noirq) { - error = drv->pm->suspend_noirq(dev); - suspend_report_result(drv->pm->suspend_noirq, error); + if (!pm) { + pci_save_state(pci_dev); + return 0; } - if (!error) - pci_pm_set_unknown_state(pci_dev); + if (pm->suspend_noirq) { + pci_power_t prev = pci_dev->current_state; + int error; - return error; + error = pm->suspend_noirq(dev); + suspend_report_result(pm->suspend_noirq, error); + if (error) + return error; + + if (!pci_dev->state_saved && pci_dev->current_state != PCI_D0 + && pci_dev->current_state != PCI_UNKNOWN) { + WARN_ONCE(pci_dev->current_state != prev, + "PCI PM: State of device not saved by %pF\n", + pm->suspend_noirq); + return 0; + } + } + + if (!pci_dev->state_saved) { + pci_save_state(pci_dev); + if (!pci_is_bridge(pci_dev)) + pci_prepare_to_sleep(pci_dev); + } + + pci_pm_set_unknown_state(pci_dev); + + return 0; } static int pci_pm_resume_noirq(struct device *dev) @@ -617,13 +716,13 @@ static int pci_pm_freeze(struct device *dev) if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_suspend(dev, PMSG_FREEZE); + pci_dev->state_saved = false; + if (!pm) { pci_pm_default_suspend(pci_dev); return 0; } - pci_dev->state_saved = false; - if (pm->freeze) { int error; @@ -633,9 +732,6 @@ static int pci_pm_freeze(struct device *dev) return error; } - if (!pci_dev->state_saved) - pci_save_state(pci_dev); - return 0; } @@ -643,20 +739,25 @@ static int pci_pm_freeze_noirq(struct device *dev) { struct pci_dev *pci_dev = to_pci_dev(dev); struct device_driver *drv = dev->driver; - int error = 0; if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_suspend_late(dev, PMSG_FREEZE); if (drv && drv->pm && drv->pm->freeze_noirq) { + int error; + error = drv->pm->freeze_noirq(dev); suspend_report_result(drv->pm->freeze_noirq, error); + if (error) + return error; } - if (!error) - pci_pm_set_unknown_state(pci_dev); + if (!pci_dev->state_saved) + pci_save_state(pci_dev); - return error; + pci_pm_set_unknown_state(pci_dev); + + return 0; } static int pci_pm_thaw_noirq(struct device *dev) @@ -699,46 +800,56 @@ 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; - int error = 0; if (pci_has_legacy_pm_support(pci_dev)) return pci_legacy_suspend(dev, PMSG_HIBERNATE); + pci_dev->state_saved = false; + if (!pm) { pci_pm_default_suspend(pci_dev); goto Fixup; } - pci_dev->state_saved = false; - if (pm->poweroff) { + int error; + error = pm->poweroff(dev); suspend_report_result(pm->poweroff, error); + if (error) + return error; } - if (!pci_dev->state_saved && !pci_is_bridge(pci_dev)) - pci_prepare_to_sleep(pci_dev); - Fixup: pci_fixup_device(pci_fixup_suspend, pci_dev); - return error; + return 0; } static int pci_pm_poweroff_noirq(struct device *dev) { + struct pci_dev *pci_dev = to_pci_dev(dev); struct device_driver *drv = dev->driver; - int error = 0; if (pci_has_legacy_pm_support(to_pci_dev(dev))) return pci_legacy_suspend_late(dev, PMSG_HIBERNATE); - if (drv && drv->pm && drv->pm->poweroff_noirq) { + if (!drv || !drv->pm) + return 0; + + if (drv->pm->poweroff_noirq) { + int error; + error = drv->pm->poweroff_noirq(dev); suspend_report_result(drv->pm->poweroff_noirq, error); + if (error) + return error; } - return error; + if (!pci_dev->state_saved && !pci_is_bridge(pci_dev)) + pci_prepare_to_sleep(pci_dev); + + return 0; } static int pci_pm_restore_noirq(struct device *dev) @@ -852,13 +963,23 @@ int __pci_register_driver(struct pci_driver *drv, struct module *owner, /* register with core */ error = driver_register(&drv->driver); if (error) - return error; + goto out; error = pci_create_newid_file(drv); if (error) - driver_unregister(&drv->driver); + goto out_newid; + error = pci_create_removeid_file(drv); + if (error) + goto out_removeid; +out: return error; + +out_removeid: + pci_remove_newid_file(drv); +out_newid: + driver_unregister(&drv->driver); + goto out; } /** @@ -874,6 +995,7 @@ int __pci_register_driver(struct pci_driver *drv, struct module *owner, void pci_unregister_driver(struct pci_driver *drv) { + pci_remove_removeid_file(drv); pci_remove_newid_file(drv); driver_unregister(&drv->driver); pci_free_dynids(drv); @@ -973,6 +1095,7 @@ struct bus_type pci_bus_type = { .remove = pci_device_remove, .shutdown = pci_device_shutdown, .dev_attrs = pci_dev_attrs, + .bus_attrs = pci_bus_attrs, .pm = PCI_PM_OPS_PTR, }; diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index dfc4e0ddf24..85ebd02a64a 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -148,7 +148,7 @@ static ssize_t is_enabled_store(struct device *dev, return -EPERM; if (!val) { - if (atomic_read(&pdev->enable_cnt) != 0) + if (pci_is_enabled(pdev)) pci_disable_device(pdev); else result = -EIO; @@ -219,6 +219,79 @@ msi_bus_store(struct device *dev, struct device_attribute *attr, return count; } +#ifdef CONFIG_HOTPLUG +static DEFINE_MUTEX(pci_remove_rescan_mutex); +static ssize_t bus_rescan_store(struct bus_type *bus, const char *buf, + size_t count) +{ + unsigned long val; + struct pci_bus *b = NULL; + + if (strict_strtoul(buf, 0, &val) < 0) + return -EINVAL; + + if (val) { + mutex_lock(&pci_remove_rescan_mutex); + while ((b = pci_find_next_bus(b)) != NULL) + pci_rescan_bus(b); + mutex_unlock(&pci_remove_rescan_mutex); + } + return count; +} + +struct bus_attribute pci_bus_attrs[] = { + __ATTR(rescan, (S_IWUSR|S_IWGRP), NULL, bus_rescan_store), + __ATTR_NULL +}; + +static ssize_t +dev_rescan_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + unsigned long val; + struct pci_dev *pdev = to_pci_dev(dev); + + if (strict_strtoul(buf, 0, &val) < 0) + return -EINVAL; + + if (val) { + mutex_lock(&pci_remove_rescan_mutex); + pci_rescan_bus(pdev->bus); + mutex_unlock(&pci_remove_rescan_mutex); + } + return count; +} + +static void remove_callback(struct device *dev) +{ + struct pci_dev *pdev = to_pci_dev(dev); + + mutex_lock(&pci_remove_rescan_mutex); + pci_remove_bus_device(pdev); + mutex_unlock(&pci_remove_rescan_mutex); +} + +static ssize_t +remove_store(struct device *dev, struct device_attribute *dummy, + const char *buf, size_t count) +{ + int ret = 0; + unsigned long val; + + if (strict_strtoul(buf, 0, &val) < 0) + return -EINVAL; + + /* An attribute cannot be unregistered by one of its own methods, + * so we have to use this roundabout approach. + */ + if (val) + ret = device_schedule_callback(dev, remove_callback); + if (ret) + count = ret; + return count; +} +#endif + struct device_attribute pci_dev_attrs[] = { __ATTR_RO(resource), __ATTR_RO(vendor), @@ -237,10 +310,25 @@ struct device_attribute pci_dev_attrs[] = { __ATTR(broken_parity_status,(S_IRUGO|S_IWUSR), broken_parity_status_show,broken_parity_status_store), __ATTR(msi_bus, 0644, msi_bus_show, msi_bus_store), +#ifdef CONFIG_HOTPLUG + __ATTR(remove, (S_IWUSR|S_IWGRP), NULL, remove_store), + __ATTR(rescan, (S_IWUSR|S_IWGRP), NULL, dev_rescan_store), +#endif __ATTR_NULL, }; static ssize_t +boot_vga_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct pci_dev *pdev = to_pci_dev(dev); + + return sprintf(buf, "%u\n", + !!(pdev->resource[PCI_ROM_RESOURCE].flags & + IORESOURCE_ROM_SHADOW)); +} +struct device_attribute vga_attr = __ATTR_RO(boot_vga); + +static ssize_t pci_read_config(struct kobject *kobj, struct bin_attribute *bin_attr, char *buf, loff_t off, size_t count) { @@ -404,6 +492,7 @@ write_vpd_attr(struct kobject *kobj, struct bin_attribute *bin_attr, /** * pci_read_legacy_io - read byte(s) from legacy I/O port space * @kobj: kobject corresponding to file to read from + * @bin_attr: struct bin_attribute for this file * @buf: buffer to store results * @off: offset into legacy I/O port space * @count: number of bytes to read @@ -429,6 +518,7 @@ pci_read_legacy_io(struct kobject *kobj, struct bin_attribute *bin_attr, /** * pci_write_legacy_io - write byte(s) to legacy I/O port space * @kobj: kobject corresponding to file to read from + * @bin_attr: struct bin_attribute for this file * @buf: buffer containing value to be written * @off: offset into legacy I/O port space * @count: number of bytes to write @@ -493,6 +583,19 @@ pci_mmap_legacy_io(struct kobject *kobj, struct bin_attribute *attr, } /** + * pci_adjust_legacy_attr - adjustment of legacy file attributes + * @b: bus to create files under + * @mmap_type: I/O port or memory + * + * Stub implementation. Can be overridden by arch if necessary. + */ +void __weak +pci_adjust_legacy_attr(struct pci_bus *b, enum pci_mmap_state mmap_type) +{ + return; +} + +/** * pci_create_legacy_files - create legacy I/O port and memory files * @b: bus to create files under * @@ -518,6 +621,7 @@ void pci_create_legacy_files(struct pci_bus *b) b->legacy_io->read = pci_read_legacy_io; b->legacy_io->write = pci_write_legacy_io; b->legacy_io->mmap = pci_mmap_legacy_io; + pci_adjust_legacy_attr(b, pci_mmap_io); error = device_create_bin_file(&b->dev, b->legacy_io); if (error) goto legacy_io_err; @@ -528,6 +632,7 @@ void pci_create_legacy_files(struct pci_bus *b) b->legacy_mem->size = 1024*1024; b->legacy_mem->attr.mode = S_IRUSR | S_IWUSR; b->legacy_mem->mmap = pci_mmap_legacy_mem; + pci_adjust_legacy_attr(b, pci_mmap_mem); error = device_create_bin_file(&b->dev, b->legacy_mem); if (error) goto legacy_mem_err; @@ -630,9 +735,9 @@ pci_mmap_resource_wc(struct kobject *kobj, struct bin_attribute *attr, /** * pci_remove_resource_files - cleanup resource files - * @dev: dev to cleanup + * @pdev: dev to cleanup * - * If we created resource files for @dev, remove them from sysfs and + * If we created resource files for @pdev, remove them from sysfs and * free their resources. */ static void @@ -690,9 +795,9 @@ static int pci_create_attr(struct pci_dev *pdev, int num, int write_combine) /** * pci_create_resource_files - create resource files in sysfs for @dev - * @dev: dev in question + * @pdev: dev in question * - * Walk the resources in @dev creating files for each resource available. + * Walk the resources in @pdev creating files for each resource available. */ static int pci_create_resource_files(struct pci_dev *pdev) { @@ -719,13 +824,14 @@ static int pci_create_resource_files(struct pci_dev *pdev) return 0; } #else /* !HAVE_PCI_MMAP */ -static inline int pci_create_resource_files(struct pci_dev *dev) { return 0; } -static inline void pci_remove_resource_files(struct pci_dev *dev) { return; } +int __weak pci_create_resource_files(struct pci_dev *dev) { return 0; } +void __weak pci_remove_resource_files(struct pci_dev *dev) { return; } #endif /* HAVE_PCI_MMAP */ /** * pci_write_rom - used to enable access to the PCI ROM display * @kobj: kernel object handle + * @bin_attr: struct bin_attribute for this file * @buf: user input * @off: file offset * @count: number of byte in input @@ -749,6 +855,7 @@ pci_write_rom(struct kobject *kobj, struct bin_attribute *bin_attr, /** * pci_read_rom - read a PCI ROM * @kobj: kernel object handle + * @bin_attr: struct bin_attribute for this file * @buf: where to put the data we read from the ROM * @off: file offset * @count: number of bytes to read @@ -884,18 +991,27 @@ int __must_check pci_create_sysfs_dev_files (struct pci_dev *pdev) pdev->rom_attr = attr; } + if ((pdev->class >> 8) == PCI_CLASS_DISPLAY_VGA) { + retval = device_create_file(&pdev->dev, &vga_attr); + if (retval) + goto err_rom_file; + } + /* add platform-specific attributes */ retval = pcibios_add_platform_entries(pdev); if (retval) - goto err_rom_file; + goto err_vga_file; /* add sysfs entries for various capabilities */ retval = pci_create_capabilities_sysfs(pdev); if (retval) - goto err_rom_file; + goto err_vga_file; return 0; +err_vga_file: + if ((pdev->class >> 8) == PCI_CLASS_DISPLAY_VGA) + device_remove_file(&pdev->dev, &vga_attr); err_rom_file: if (rom_size) { sysfs_remove_bin_file(&pdev->dev.kobj, pdev->rom_attr); diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 6d6120007af..6c93af5ced1 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -20,8 +20,15 @@ #include <linux/pm_wakeup.h> #include <linux/interrupt.h> #include <asm/dma.h> /* isa_dma_bridge_buggy */ +#include <linux/device.h> +#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 @@ -426,7 +433,6 @@ static inline int platform_pci_sleep_wake(struct pci_dev *dev, bool enable) * given PCI device * @dev: PCI device to handle. * @state: PCI power state (D0, D1, D2, D3hot) to put the device into. - * @wait: If 'true', wait for the device to change its power state * * RETURN VALUE: * -EINVAL if the requested state is invalid. @@ -435,12 +441,15 @@ static inline int platform_pci_sleep_wake(struct pci_dev *dev, bool enable) * 0 if device already is in the requested state. * 0 if device's power state has been successfully changed. */ -static int -pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state, bool wait) +static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state) { u16 pmcsr; bool need_restore = false; + /* Check if we're already there */ + if (dev->current_state == state) + return 0; + if (!dev->pm_cap) return -EIO; @@ -451,10 +460,7 @@ pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state, bool wait) * Can enter D0 from any state, but if we can only go deeper * to sleep if we're already in a low power state */ - if (dev->current_state == state) { - /* we're already there */ - return 0; - } else if (state != PCI_D0 && dev->current_state <= PCI_D3cold + if (state != PCI_D0 && dev->current_state <= PCI_D3cold && dev->current_state > state) { dev_err(&dev->dev, "invalid power transition " "(from state %d to %d)\n", dev->current_state, state); @@ -479,12 +485,12 @@ pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state, bool wait) 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)) { + && !(pmcsr & PCI_PM_CTRL_NO_SOFT_RESET)) need_restore = true; - wait = true; - } /* Fall-through: force to D0 */ default: pmcsr = 0; @@ -494,9 +500,6 @@ pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state, bool wait) /* enter specified state */ pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, pmcsr); - if (!wait) - return 0; - /* Mandatory power management transition delays */ /* see PCI PM 1.1 5.6.1 table 18 */ if (state == PCI_D3hot || dev->current_state == PCI_D3hot) @@ -521,7 +524,7 @@ pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state, bool wait) if (need_restore) pci_restore_bars(dev); - if (wait && dev->bus->self) + if (dev->bus->self) pcie_aspm_pm_state_change(dev->bus->self); return 0; @@ -546,11 +549,59 @@ void pci_update_current_state(struct pci_dev *dev, pci_power_t state) } /** + * pci_platform_power_transition - Use platform to change device power state + * @dev: PCI device to handle. + * @state: State to put the device into. + */ +static int pci_platform_power_transition(struct pci_dev *dev, pci_power_t state) +{ + int error; + + if (platform_pci_power_manageable(dev)) { + error = platform_pci_set_power_state(dev, state); + if (!error) + pci_update_current_state(dev, state); + } else { + error = -ENODEV; + /* Fall back to PCI_D0 if native PM is not supported */ + if (!dev->pm_cap) + dev->current_state = PCI_D0; + } + + return error; +} + +/** + * __pci_start_power_transition - Start power transition of a PCI device + * @dev: PCI device to handle. + * @state: State to put the device into. + */ +static void __pci_start_power_transition(struct pci_dev *dev, pci_power_t state) +{ + if (state == PCI_D0) + pci_platform_power_transition(dev, PCI_D0); +} + +/** + * __pci_complete_power_transition - Complete power transition of a PCI device + * @dev: PCI device to handle. + * @state: State to put the device into. + * + * This function should not be called directly by device drivers. + */ +int __pci_complete_power_transition(struct pci_dev *dev, pci_power_t state) +{ + return state > PCI_D0 ? + pci_platform_power_transition(dev, state) : -EINVAL; +} +EXPORT_SYMBOL_GPL(__pci_complete_power_transition); + +/** * pci_set_power_state - Set the power state of a PCI device * @dev: PCI device to handle. * @state: PCI power state (D0, D1, D2, D3hot) to put the device into. * - * Transition a device to a new power state, using the platform formware and/or + * Transition a device to a new power state, using the platform firmware and/or * the device's PCI PM registers. * * RETURN VALUE: @@ -577,30 +628,21 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state) */ return 0; - if (state == PCI_D0 && platform_pci_power_manageable(dev)) { - /* - * Allow the platform to change the state, for example via ACPI - * _PR0, _PS0 and some such, but do not trust it. - */ - int ret = platform_pci_set_power_state(dev, PCI_D0); - if (!ret) - pci_update_current_state(dev, PCI_D0); - } + /* Check if we're already there */ + if (dev->current_state == state) + return 0; + + __pci_start_power_transition(dev, state); + /* This device is quirked not to be put into D3, so don't put it in D3 */ if (state == PCI_D3hot && (dev->dev_flags & PCI_DEV_FLAGS_NO_D3)) return 0; - error = pci_raw_set_power_state(dev, state, true); + error = pci_raw_set_power_state(dev, state); - if (state > PCI_D0 && platform_pci_power_manageable(dev)) { - /* Allow the platform to finalize the transition */ - int ret = platform_pci_set_power_state(dev, state); - if (!ret) { - pci_update_current_state(dev, state); - error = 0; - } - } + if (!__pci_complete_power_transition(dev, state)) + error = 0; return error; } @@ -645,11 +687,36 @@ pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state) EXPORT_SYMBOL(pci_choose_state); +#define PCI_EXP_SAVE_REGS 7 + +#define pcie_cap_has_devctl(type, flags) 1 +#define pcie_cap_has_lnkctl(type, flags) \ + ((flags & PCI_EXP_FLAGS_VERS) > 1 || \ + (type == PCI_EXP_TYPE_ROOT_PORT || \ + type == PCI_EXP_TYPE_ENDPOINT || \ + type == PCI_EXP_TYPE_LEG_END)) +#define pcie_cap_has_sltctl(type, flags) \ + ((flags & PCI_EXP_FLAGS_VERS) > 1 || \ + ((type == PCI_EXP_TYPE_ROOT_PORT) || \ + (type == PCI_EXP_TYPE_DOWNSTREAM && \ + (flags & PCI_EXP_FLAGS_SLOT)))) +#define pcie_cap_has_rtctl(type, flags) \ + ((flags & PCI_EXP_FLAGS_VERS) > 1 || \ + (type == PCI_EXP_TYPE_ROOT_PORT || \ + type == PCI_EXP_TYPE_RC_EC)) +#define pcie_cap_has_devctl2(type, flags) \ + ((flags & PCI_EXP_FLAGS_VERS) > 1) +#define pcie_cap_has_lnkctl2(type, flags) \ + ((flags & PCI_EXP_FLAGS_VERS) > 1) +#define pcie_cap_has_sltctl2(type, flags) \ + ((flags & PCI_EXP_FLAGS_VERS) > 1) + static int pci_save_pcie_state(struct pci_dev *dev) { int pos, i = 0; struct pci_cap_saved_state *save_state; u16 *cap; + u16 flags; pos = pci_find_capability(dev, PCI_CAP_ID_EXP); if (pos <= 0) @@ -657,15 +724,27 @@ static int pci_save_pcie_state(struct pci_dev *dev) save_state = pci_find_saved_cap(dev, PCI_CAP_ID_EXP); if (!save_state) { - dev_err(&dev->dev, "buffer not found in %s\n", __FUNCTION__); + dev_err(&dev->dev, "buffer not found in %s\n", __func__); return -ENOMEM; } cap = (u16 *)&save_state->data[0]; - pci_read_config_word(dev, pos + PCI_EXP_DEVCTL, &cap[i++]); - pci_read_config_word(dev, pos + PCI_EXP_LNKCTL, &cap[i++]); - pci_read_config_word(dev, pos + PCI_EXP_SLTCTL, &cap[i++]); - pci_read_config_word(dev, pos + PCI_EXP_RTCTL, &cap[i++]); + pci_read_config_word(dev, pos + PCI_EXP_FLAGS, &flags); + + if (pcie_cap_has_devctl(dev->pcie_type, flags)) + pci_read_config_word(dev, pos + PCI_EXP_DEVCTL, &cap[i++]); + if (pcie_cap_has_lnkctl(dev->pcie_type, flags)) + pci_read_config_word(dev, pos + PCI_EXP_LNKCTL, &cap[i++]); + if (pcie_cap_has_sltctl(dev->pcie_type, flags)) + pci_read_config_word(dev, pos + PCI_EXP_SLTCTL, &cap[i++]); + if (pcie_cap_has_rtctl(dev->pcie_type, flags)) + pci_read_config_word(dev, pos + PCI_EXP_RTCTL, &cap[i++]); + if (pcie_cap_has_devctl2(dev->pcie_type, flags)) + pci_read_config_word(dev, pos + PCI_EXP_DEVCTL2, &cap[i++]); + if (pcie_cap_has_lnkctl2(dev->pcie_type, flags)) + pci_read_config_word(dev, pos + PCI_EXP_LNKCTL2, &cap[i++]); + if (pcie_cap_has_sltctl2(dev->pcie_type, flags)) + pci_read_config_word(dev, pos + PCI_EXP_SLTCTL2, &cap[i++]); return 0; } @@ -675,6 +754,7 @@ static void pci_restore_pcie_state(struct pci_dev *dev) int i = 0, pos; struct pci_cap_saved_state *save_state; u16 *cap; + u16 flags; save_state = pci_find_saved_cap(dev, PCI_CAP_ID_EXP); pos = pci_find_capability(dev, PCI_CAP_ID_EXP); @@ -682,10 +762,22 @@ static void pci_restore_pcie_state(struct pci_dev *dev) return; cap = (u16 *)&save_state->data[0]; - pci_write_config_word(dev, pos + PCI_EXP_DEVCTL, cap[i++]); - pci_write_config_word(dev, pos + PCI_EXP_LNKCTL, cap[i++]); - pci_write_config_word(dev, pos + PCI_EXP_SLTCTL, cap[i++]); - pci_write_config_word(dev, pos + PCI_EXP_RTCTL, cap[i++]); + pci_read_config_word(dev, pos + PCI_EXP_FLAGS, &flags); + + if (pcie_cap_has_devctl(dev->pcie_type, flags)) + pci_write_config_word(dev, pos + PCI_EXP_DEVCTL, cap[i++]); + if (pcie_cap_has_lnkctl(dev->pcie_type, flags)) + pci_write_config_word(dev, pos + PCI_EXP_LNKCTL, cap[i++]); + if (pcie_cap_has_sltctl(dev->pcie_type, flags)) + pci_write_config_word(dev, pos + PCI_EXP_SLTCTL, cap[i++]); + if (pcie_cap_has_rtctl(dev->pcie_type, flags)) + pci_write_config_word(dev, pos + PCI_EXP_RTCTL, cap[i++]); + if (pcie_cap_has_devctl2(dev->pcie_type, flags)) + pci_write_config_word(dev, pos + PCI_EXP_DEVCTL2, cap[i++]); + if (pcie_cap_has_lnkctl2(dev->pcie_type, flags)) + pci_write_config_word(dev, pos + PCI_EXP_LNKCTL2, cap[i++]); + if (pcie_cap_has_sltctl2(dev->pcie_type, flags)) + pci_write_config_word(dev, pos + PCI_EXP_SLTCTL2, cap[i++]); } @@ -700,7 +792,7 @@ static int pci_save_pcix_state(struct pci_dev *dev) save_state = pci_find_saved_cap(dev, PCI_CAP_ID_PCIX); if (!save_state) { - dev_err(&dev->dev, "buffer not found in %s\n", __FUNCTION__); + dev_err(&dev->dev, "buffer not found in %s\n", __func__); return -ENOMEM; } @@ -773,6 +865,7 @@ pci_restore_state(struct pci_dev *dev) } pci_restore_pcix_state(dev); pci_restore_msi_state(dev); + pci_restore_iov_state(dev); return 0; } @@ -801,7 +894,7 @@ static int do_pci_enable_device(struct pci_dev *dev, int bars) */ int pci_reenable_device(struct pci_dev *dev) { - if (atomic_read(&dev->enable_cnt)) + if (pci_is_enabled(dev)) return do_pci_enable_device(dev, (1 << PCI_NUM_RESOURCES) - 1); return 0; } @@ -999,7 +1092,7 @@ static void do_pci_disable_device(struct pci_dev *dev) */ void pci_disable_enabled_device(struct pci_dev *dev) { - if (atomic_read(&dev->enable_cnt)) + if (pci_is_enabled(dev)) do_pci_disable_device(dev); } @@ -1117,7 +1210,7 @@ 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; @@ -1196,15 +1289,14 @@ 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))) @@ -1231,7 +1323,7 @@ int pci_prepare_to_sleep(struct pci_dev *dev) if (target_state == PCI_POWER_ERROR) return -EIO; - pci_enable_wake(dev, target_state, true); + pci_enable_wake(dev, target_state, device_may_wakeup(&dev->dev)); error = pci_set_power_state(dev, target_state); @@ -1369,7 +1461,8 @@ void pci_allocate_cap_save_buffers(struct pci_dev *dev) { int error; - error = pci_add_cap_save_buffer(dev, PCI_CAP_ID_EXP, 4 * sizeof(u16)); + error = pci_add_cap_save_buffer(dev, PCI_CAP_ID_EXP, + PCI_EXP_SAVE_REGS * sizeof(u16)); if (error) dev_err(&dev->dev, "unable to preallocate PCI Express save buffer\n"); @@ -1381,50 +1474,6 @@ void pci_allocate_cap_save_buffers(struct pci_dev *dev) } /** - * pci_restore_standard_config - restore standard config registers of PCI device - * @dev: PCI device to handle - * - * This function assumes that the device's configuration space is accessible. - * If the device needs to be powered up, the function will wait for it to - * change the state. - */ -int pci_restore_standard_config(struct pci_dev *dev) -{ - pci_power_t prev_state; - int error; - - pci_update_current_state(dev, PCI_D0); - - prev_state = dev->current_state; - if (prev_state == PCI_D0) - goto Restore; - - error = pci_raw_set_power_state(dev, PCI_D0, false); - if (error) - return error; - - /* - * This assumes that we won't get a bus in B2 or B3 from the BIOS, but - * we've made this assumption forever and it appears to be universally - * satisfied. - */ - switch(prev_state) { - case PCI_D3cold: - case PCI_D3hot: - mdelay(pci_pm_d3_delay); - break; - case PCI_D2: - udelay(PCI_PM_D2_DELAY); - break; - } - - pci_update_current_state(dev, PCI_D0); - - Restore: - return dev->state_saved ? pci_restore_state(dev) : 0; -} - -/** * pci_enable_ari - enable ARI forwarding if hardware support it * @dev: the PCI device */ @@ -1484,7 +1533,7 @@ pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge) if (!pin) return -1; - while (dev->bus->self) { + while (!pci_is_root_bus(dev->bus)) { pin = pci_swizzle_interrupt_pin(dev, pin); dev = dev->bus->self; } @@ -1504,7 +1553,7 @@ u8 pci_common_swizzle(struct pci_dev *dev, u8 *pinp) { u8 pin = *pinp; - while (dev->bus->self) { + while (!pci_is_root_bus(dev->bus)) { pin = pci_swizzle_interrupt_pin(dev, pin); dev = dev->bus->self; } @@ -2010,99 +2059,177 @@ 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 */ - msleep(100); - pci_read_config_word(dev, exppos + PCI_EXP_DEVSTA, &status); - if (status & PCI_EXP_DEVSTA_TRPND) { - 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"); + for (i = 0; i < 4; i++) { + if (i) + msleep((1 << (i - 1)) * 100); + + pci_read_config_word(dev, pos + PCI_EXP_DEVSTA, &status); + if (!(status & PCI_EXP_DEVSTA_TRPND)) + goto clear; } - pci_write_config_word(dev, exppos + PCI_EXP_DEVCTL, + 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 */ - msleep(100); - pci_read_config_byte(dev, cappos + PCI_AF_STATUS, &status); - if (status & PCI_AF_STATUS_TP) { - 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"); + 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; } - pci_write_config_byte(dev, cappos + PCI_AF_CTRL, PCI_AF_CTRL_FLR); - mdelay(100); - pci_unblock_user_cfg_access(dev); + 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); + 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; + + pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &csr); + if (csr & PCI_PM_CTRL_NO_SOFT_RESET) + return -ENOTTY; + + if (probe) + return 0; - res = __pcie_flr(pdev, probe); - if (res != -ENOTTY) - return res; + if (dev->current_state != PCI_D0) + return -EINVAL; + + 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); - res = __pci_af_flr(pdev, probe); - if (res != -ENOTTY) - return res; + 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 res; + return 0; +} + +static int pci_parent_bus_reset(struct pci_dev *dev, int probe) +{ + u16 ctrl; + struct pci_dev *pdev; + + if (dev->subordinate) + 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 @@ -2114,18 +2241,18 @@ 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 @@ -2133,32 +2260,33 @@ 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); @@ -2346,18 +2474,140 @@ int pci_select_bars(struct pci_dev *dev, unsigned long flags) */ int pci_resource_bar(struct pci_dev *dev, int resno, enum pci_bar_type *type) { + int reg; + if (resno < PCI_ROM_RESOURCE) { *type = pci_bar_unknown; return PCI_BASE_ADDRESS_0 + 4 * resno; } else if (resno == PCI_ROM_RESOURCE) { *type = pci_bar_mem32; return dev->rom_base_reg; + } else if (resno < PCI_BRIDGE_RESOURCES) { + /* device specific resource */ + reg = pci_iov_resource_bar(dev, resno, type); + if (reg) + return reg; } dev_err(&dev->dev, "BAR: invalid resource #%d\n", resno); return 0; } +#define RESOURCE_ALIGNMENT_PARAM_SIZE COMMAND_LINE_SIZE +static char resource_alignment_param[RESOURCE_ALIGNMENT_PARAM_SIZE] = {0}; +spinlock_t resource_alignment_lock = SPIN_LOCK_UNLOCKED; + +/** + * pci_specified_resource_alignment - get resource alignment specified by user. + * @dev: the PCI device to get + * + * RETURNS: Resource alignment if it is specified. + * Zero if it is not specified. + */ +resource_size_t pci_specified_resource_alignment(struct pci_dev *dev) +{ + int seg, bus, slot, func, align_order, count; + resource_size_t align = 0; + char *p; + + spin_lock(&resource_alignment_lock); + p = resource_alignment_param; + while (*p) { + count = 0; + if (sscanf(p, "%d%n", &align_order, &count) == 1 && + p[count] == '@') { + p += count + 1; + } else { + align_order = -1; + } + if (sscanf(p, "%x:%x:%x.%x%n", + &seg, &bus, &slot, &func, &count) != 4) { + seg = 0; + if (sscanf(p, "%x:%x.%x%n", + &bus, &slot, &func, &count) != 3) { + /* Invalid format */ + printk(KERN_ERR "PCI: Can't parse resource_alignment parameter: %s\n", + p); + break; + } + } + p += count; + if (seg == pci_domain_nr(dev->bus) && + bus == dev->bus->number && + slot == PCI_SLOT(dev->devfn) && + func == PCI_FUNC(dev->devfn)) { + if (align_order == -1) { + align = PAGE_SIZE; + } else { + align = 1 << align_order; + } + /* Found */ + break; + } + if (*p != ';' && *p != ',') { + /* End of param or invalid format */ + break; + } + p++; + } + spin_unlock(&resource_alignment_lock); + return align; +} + +/** + * pci_is_reassigndev - check if specified PCI is target device to reassign + * @dev: the PCI device to check + * + * RETURNS: non-zero for PCI device is a target device to reassign, + * or zero is not. + */ +int pci_is_reassigndev(struct pci_dev *dev) +{ + return (pci_specified_resource_alignment(dev) != 0); +} + +ssize_t pci_set_resource_alignment_param(const char *buf, size_t count) +{ + if (count > RESOURCE_ALIGNMENT_PARAM_SIZE - 1) + count = RESOURCE_ALIGNMENT_PARAM_SIZE - 1; + spin_lock(&resource_alignment_lock); + strncpy(resource_alignment_param, buf, count); + resource_alignment_param[count] = '\0'; + spin_unlock(&resource_alignment_lock); + return count; +} + +ssize_t pci_get_resource_alignment_param(char *buf, size_t size) +{ + size_t count; + spin_lock(&resource_alignment_lock); + count = snprintf(buf, size, "%s", resource_alignment_param); + spin_unlock(&resource_alignment_lock); + return count; +} + +static ssize_t pci_resource_alignment_show(struct bus_type *bus, char *buf) +{ + return pci_get_resource_alignment_param(buf, PAGE_SIZE); +} + +static ssize_t pci_resource_alignment_store(struct bus_type *bus, + const char *buf, size_t count) +{ + return pci_set_resource_alignment_param(buf, count); +} + +BUS_ATTR(resource_alignment, 0644, pci_resource_alignment_show, + pci_resource_alignment_store); + +static int __init pci_resource_alignment_sysfs_init(void) +{ + return bus_create_file(&pci_bus_type, + &bus_attr_resource_alignment); +} + +late_initcall(pci_resource_alignment_sysfs_init); + static void __devinit pci_no_domains(void) { #ifdef CONFIG_PCI_DOMAINS @@ -2406,6 +2656,11 @@ static int __init pci_setup(char *str) pci_cardbus_io_size = memparse(str + 9, &str); } else if (!strncmp(str, "cbmemsize=", 10)) { pci_cardbus_mem_size = memparse(str + 10, &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 07c0aa5275e..d03f6b99f29 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -1,6 +1,8 @@ #ifndef DRIVERS_PCI_H #define DRIVERS_PCI_H +#include <linux/workqueue.h> + #define PCI_CFG_SPACE_SIZE 256 #define PCI_CFG_SPACE_EXP_SIZE 4096 @@ -49,7 +51,6 @@ extern void pci_disable_enabled_device(struct pci_dev *dev); extern void pci_pm_init(struct pci_dev *dev); extern void platform_pci_wakeup_init(struct pci_dev *dev); extern void pci_allocate_cap_save_buffers(struct pci_dev *dev); -extern int pci_restore_standard_config(struct pci_dev *dev); static inline bool pci_is_bridge(struct pci_dev *pci_dev) { @@ -136,6 +137,12 @@ extern int pcie_mch_quirk; extern struct device_attribute pci_dev_attrs[]; extern struct device_attribute dev_attr_cpuaffinity; extern struct device_attribute dev_attr_cpulistaffinity; +#ifdef CONFIG_HOTPLUG +extern struct bus_attribute pci_bus_attrs[]; +#else +#define pci_bus_attrs NULL +#endif + /** * pci_match_one_device - Tell if a PCI device structure has a matching @@ -178,6 +185,7 @@ enum pci_bar_type { pci_bar_mem64, /* A 64-bit memory BAR */ }; +extern int pci_setup_device(struct pci_dev *dev); extern int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type, struct resource *res, unsigned int reg); extern int pci_resource_bar(struct pci_dev *dev, int resno, @@ -195,4 +203,60 @@ static inline int pci_ari_enabled(struct pci_bus *bus) return bus->self && bus->self->ari_enabled; } +#ifdef CONFIG_PCI_QUIRKS +extern int pci_is_reassigndev(struct pci_dev *dev); +resource_size_t pci_specified_resource_alignment(struct pci_dev *dev); +extern void pci_disable_bridge_window(struct pci_dev *dev); +#endif + +/* Single Root I/O Virtualization */ +struct pci_sriov { + int pos; /* capability position */ + int nres; /* number of resources */ + u32 cap; /* SR-IOV Capabilities */ + u16 ctrl; /* SR-IOV Control */ + u16 total; /* total VFs associated with the PF */ + u16 initial; /* initial VFs associated with the PF */ + u16 nr_virtfn; /* number of VFs available */ + u16 offset; /* first VF Routing ID offset */ + u16 stride; /* following VF stride */ + u32 pgsz; /* page size for BAR alignment */ + u8 link; /* Function Dependency Link */ + struct pci_dev *dev; /* lowest numbered PF */ + struct pci_dev *self; /* this PF */ + struct mutex lock; /* lock for VF bus */ + struct work_struct mtask; /* VF Migration task */ + u8 __iomem *mstate; /* VF Migration State Array */ +}; + +#ifdef CONFIG_PCI_IOV +extern int pci_iov_init(struct pci_dev *dev); +extern void pci_iov_release(struct pci_dev *dev); +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); +#else +static inline int pci_iov_init(struct pci_dev *dev) +{ + return -ENODEV; +} +static inline void pci_iov_release(struct pci_dev *dev) + +{ +} +static inline int pci_iov_resource_bar(struct pci_dev *dev, int resno, + enum pci_bar_type *type) +{ + return 0; +} +static inline void pci_restore_iov_state(struct pci_dev *dev) +{ +} +static inline int pci_iov_bus_range(struct pci_bus *bus) +{ + 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 e390707661d..4770f13b3ca 100644 --- a/drivers/pci/pcie/aer/aerdrv.c +++ b/drivers/pci/pcie/aer/aerdrv.c @@ -38,30 +38,13 @@ MODULE_AUTHOR(DRIVER_AUTHOR); MODULE_DESCRIPTION(DRIVER_DESC); MODULE_LICENSE("GPL"); -static int __devinit aer_probe (struct pcie_device *dev, - const struct pcie_port_service_id *id ); +static int __devinit aer_probe (struct pcie_device *dev); static void aer_remove(struct pcie_device *dev); -static int aer_suspend(struct pcie_device *dev, pm_message_t state) -{return 0;} -static int aer_resume(struct pcie_device *dev) {return 0;} static pci_ers_result_t aer_error_detected(struct pci_dev *dev, enum pci_channel_state error); static void aer_error_resume(struct pci_dev *dev); static pci_ers_result_t aer_root_reset(struct pci_dev *dev); -/* - * PCI Express bus's AER Root service driver data structure - */ -static struct pcie_port_service_id aer_id[] = { - { - .vendor = PCI_ANY_ID, - .device = PCI_ANY_ID, - .port_type = PCIE_RC_PORT, - .service_type = PCIE_PORT_SERVICE_AER, - }, - { /* end: all zeroes */ } -}; - static struct pci_error_handlers aer_error_handlers = { .error_detected = aer_error_detected, .resume = aer_error_resume, @@ -69,14 +52,12 @@ static struct pci_error_handlers aer_error_handlers = { static struct pcie_port_service_driver aerdriver = { .name = "aer", - .id_table = &aer_id[0], + .port_type = PCIE_ANY_PORT, + .service = PCIE_PORT_SERVICE_AER, .probe = aer_probe, .remove = aer_remove, - .suspend = aer_suspend, - .resume = aer_resume, - .err_handler = &aer_error_handlers, .reset_link = aer_root_reset, @@ -96,7 +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; @@ -145,6 +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 @@ -207,8 +189,7 @@ static void aer_remove(struct pcie_device *dev) * * Invoked when PCI Express bus loads AER service driver. **/ -static int __devinit aer_probe (struct pcie_device *dev, - const struct pcie_port_service_id *id ) +static int __devinit aer_probe (struct pcie_device *dev) { int status; struct aer_rpc *rpc; diff --git a/drivers/pci/pcie/aer/aerdrv.h b/drivers/pci/pcie/aer/aerdrv.h index c7ad68b6c6d..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 */ @@ -95,6 +100,9 @@ struct aer_broadcast_data { static inline pci_ers_result_t merge_result(enum pci_ers_result orig, enum pci_ers_result new) { + if (new == PCI_ERS_RESULT_NONE) + return orig; + switch (orig) { case PCI_ERS_RESULT_CAN_RECOVER: case PCI_ERS_RESULT_RECOVERED: @@ -117,6 +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_acpi.c b/drivers/pci/pcie/aer/aerdrv_acpi.c index ebce26c3704..8edb2f300e8 100644 --- a/drivers/pci/pcie/aer/aerdrv_acpi.c +++ b/drivers/pci/pcie/aer/aerdrv_acpi.c @@ -38,7 +38,7 @@ int aer_osc_setup(struct pcie_device *pciedev) handle = acpi_find_root_bridge_handle(pdev); if (handle) { - status = pci_osc_control_set(handle, + status = acpi_pci_osc_control_set(handle, OSC_PCI_EXPRESS_AER_CONTROL | OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); } diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c index 38257500738..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; @@ -351,21 +432,21 @@ static int find_aer_service_iter(struct device *device, void *data) { struct device_driver *driver; struct pcie_port_service_driver *service_driver; - struct pcie_device *pcie_dev; struct find_aer_service_data *result; result = (struct find_aer_service_data *) data; if (device->bus == &pcie_port_bus_type) { - pcie_dev = to_pcie_device(device); - if (pcie_dev->id.port_type == PCIE_SW_DOWNSTREAM_PORT) + struct pcie_port_data *port_data; + + port_data = pci_get_drvdata(to_pcie_device(device)->port); + if (port_data->port_type == PCIE_SW_DOWNSTREAM_PORT) result->is_downstream = 1; driver = device->driver; if (driver) { service_driver = to_service_driver(driver); - if (service_driver->id_table->service_type == - PCIE_PORT_SERVICE_AER) { + if (service_driver->service == PCIE_PORT_SERVICE_AER) { result->aer_driver = service_driver; return 1; } @@ -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..ece97df4df6 --- /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.h b/drivers/pci/pcie/portdrv.h index 2529f3f2ea5..17ad53868f9 100644 --- a/drivers/pci/pcie/portdrv.h +++ b/drivers/pci/pcie/portdrv.h @@ -25,19 +25,21 @@ #define PCIE_CAPABILITIES_REG 0x2 #define PCIE_SLOT_CAPABILITIES_REG 0x14 #define PCIE_PORT_DEVICE_MAXSERVICES 4 +#define PCIE_PORT_MSI_VECTOR_MASK 0x1f +/* + * According to the PCI Express Base Specification 2.0, the indices of the MSI-X + * table entires used by port services must not exceed 31 + */ +#define PCIE_PORT_MAX_MSIX_ENTRIES 32 #define get_descriptor_id(type, service) (((type - 4) << 4) | service) -struct pcie_port_device_ext { - int interrupt_mode; /* [0:INTx | 1:MSI | 2:MSI-X] */ -}; - extern struct bus_type pcie_port_bus_type; extern int pcie_port_device_probe(struct pci_dev *dev); extern int pcie_port_device_register(struct pci_dev *dev); #ifdef CONFIG_PM -extern int pcie_port_device_suspend(struct pci_dev *dev, pm_message_t state); -extern int pcie_port_device_resume(struct pci_dev *dev); +extern int pcie_port_device_suspend(struct device *dev); +extern int pcie_port_device_resume(struct device *dev); #endif extern void pcie_port_device_remove(struct pci_dev *dev); extern int __must_check pcie_port_bus_register(void); diff --git a/drivers/pci/pcie/portdrv_bus.c b/drivers/pci/pcie/portdrv_bus.c index eec89b767f9..ef3a4eeaebb 100644 --- a/drivers/pci/pcie/portdrv_bus.c +++ b/drivers/pci/pcie/portdrv_bus.c @@ -26,20 +26,22 @@ EXPORT_SYMBOL_GPL(pcie_port_bus_type); static int pcie_port_bus_match(struct device *dev, struct device_driver *drv) { struct pcie_device *pciedev; + struct pcie_port_data *port_data; struct pcie_port_service_driver *driver; if (drv->bus != &pcie_port_bus_type || dev->bus != &pcie_port_bus_type) return 0; - + pciedev = to_pcie_device(dev); driver = to_service_driver(drv); - if ( (driver->id_table->vendor != PCI_ANY_ID && - driver->id_table->vendor != pciedev->id.vendor) || - (driver->id_table->device != PCI_ANY_ID && - driver->id_table->device != pciedev->id.device) || - (driver->id_table->port_type != PCIE_ANY_PORT && - driver->id_table->port_type != pciedev->id.port_type) || - driver->id_table->service_type != pciedev->id.service_type ) + + if (driver->service != pciedev->service) + return 0; + + port_data = pci_get_drvdata(pciedev->port); + + if (driver->port_type != PCIE_ANY_PORT + && driver->port_type != port_data->port_type) return 0; return 1; diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c index 8b3f8c18032..13ffdc35ea0 100644 --- a/drivers/pci/pcie/portdrv_core.c +++ b/drivers/pci/pcie/portdrv_core.c @@ -15,10 +15,9 @@ #include <linux/slab.h> #include <linux/pcieport_if.h> +#include "../pci.h" #include "portdrv.h" -extern int pcie_mch_quirk; /* MSI-quirk Indicator */ - /** * release_pcie_device - free PCI Express port service device structure * @dev: Port service device to release @@ -31,26 +30,150 @@ static void release_pcie_device(struct device *dev) kfree(to_pcie_device(dev)); } -static int is_msi_quirked(struct pci_dev *dev) +/** + * pcie_port_msix_add_entry - add entry to given array of MSI-X entries + * @entries: Array of MSI-X entries + * @new_entry: Index of the entry to add to the array + * @nr_entries: Number of entries aleady in the array + * + * Return value: Position of the added entry in the array + */ +static int pcie_port_msix_add_entry( + struct msix_entry *entries, int new_entry, int nr_entries) { - int port_type, quirk = 0; + int j; + + for (j = 0; j < nr_entries; j++) + if (entries[j].entry == new_entry) + return j; + + entries[j].entry = new_entry; + return j; +} + +/** + * pcie_port_enable_msix - try to set up MSI-X as interrupt mode for given port + * @dev: PCI Express port to handle + * @vectors: Array of interrupt vectors to populate + * @mask: Bitmask of port capabilities returned by get_port_device_capability() + * + * Return value: 0 on success, error code on failure + */ +static int pcie_port_enable_msix(struct pci_dev *dev, int *vectors, int mask) +{ + struct msix_entry *msix_entries; + int idx[PCIE_PORT_DEVICE_MAXSERVICES]; + int nr_entries, status, pos, i, nvec; u16 reg16; + u32 reg32; - pci_read_config_word(dev, - pci_find_capability(dev, PCI_CAP_ID_EXP) + - PCIE_CAPABILITIES_REG, ®16); - port_type = (reg16 >> 4) & PORT_TYPE_MASK; - switch(port_type) { - case PCIE_RC_PORT: - if (pcie_mch_quirk == 1) - quirk = 1; - break; - case PCIE_SW_UPSTREAM_PORT: - case PCIE_SW_DOWNSTREAM_PORT: - default: - break; + nr_entries = pci_msix_table_size(dev); + if (!nr_entries) + return -EINVAL; + if (nr_entries > PCIE_PORT_MAX_MSIX_ENTRIES) + nr_entries = PCIE_PORT_MAX_MSIX_ENTRIES; + + msix_entries = kzalloc(sizeof(*msix_entries) * nr_entries, GFP_KERNEL); + if (!msix_entries) + return -ENOMEM; + + /* + * Allocate as many entries as the port wants, so that we can check + * which of them will be useful. Moreover, if nr_entries is correctly + * equal to the number of entries this port actually uses, we'll happily + * go through without any tricks. + */ + for (i = 0; i < nr_entries; i++) + msix_entries[i].entry = i; + + status = pci_enable_msix(dev, msix_entries, nr_entries); + if (status) + goto Exit; + + for (i = 0; i < PCIE_PORT_DEVICE_MAXSERVICES; i++) + idx[i] = -1; + status = -EIO; + nvec = 0; + + if (mask & (PCIE_PORT_SERVICE_PME | PCIE_PORT_SERVICE_HP)) { + int entry; + + /* + * The code below follows the PCI Express Base Specification 2.0 + * stating in Section 6.1.6 that "PME and Hot-Plug Event + * interrupts (when both are implemented) always share the same + * MSI or MSI-X vector, as indicated by the Interrupt Message + * Number field in the PCI Express Capabilities register", where + * according to Section 7.8.2 of the specification "For MSI-X, + * the value in this field indicates which MSI-X Table entry is + * used to generate the interrupt message." + */ + pos = pci_find_capability(dev, PCI_CAP_ID_EXP); + pci_read_config_word(dev, pos + PCIE_CAPABILITIES_REG, ®16); + entry = (reg16 >> 9) & PCIE_PORT_MSI_VECTOR_MASK; + if (entry >= nr_entries) + goto Error; + + i = pcie_port_msix_add_entry(msix_entries, entry, nvec); + if (i == nvec) + nvec++; + + idx[PCIE_PORT_SERVICE_PME_SHIFT] = i; + idx[PCIE_PORT_SERVICE_HP_SHIFT] = i; + } + + if (mask & PCIE_PORT_SERVICE_AER) { + int entry; + + /* + * The code below follows Section 7.10.10 of the PCI Express + * Base Specification 2.0 stating that bits 31-27 of the Root + * Error Status Register contain a value indicating which of the + * MSI/MSI-X vectors assigned to the port is going to be used + * for AER, where "For MSI-X, the value in this register + * indicates which MSI-X Table entry is used to generate the + * interrupt message." + */ + pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR); + pci_read_config_dword(dev, pos + PCI_ERR_ROOT_STATUS, ®32); + entry = reg32 >> 27; + if (entry >= nr_entries) + goto Error; + + i = pcie_port_msix_add_entry(msix_entries, entry, nvec); + if (i == nvec) + nvec++; + + idx[PCIE_PORT_SERVICE_AER_SHIFT] = i; } - return quirk; + + /* + * If nvec is equal to the allocated number of entries, we can just use + * what we have. Otherwise, the port has some extra entries not for the + * services we know and we need to work around that. + */ + if (nvec == nr_entries) { + status = 0; + } else { + /* Drop the temporary MSI-X setup */ + pci_disable_msix(dev); + + /* Now allocate the MSI-X vectors for real */ + status = pci_enable_msix(dev, msix_entries, nvec); + if (status) + goto Exit; + } + + for (i = 0; i < PCIE_PORT_DEVICE_MAXSERVICES; i++) + vectors[i] = idx[i] >= 0 ? msix_entries[idx[i]].vector : -1; + + Exit: + kfree(msix_entries); + return status; + + Error: + pci_disable_msix(dev); + goto Exit; } /** @@ -64,47 +187,32 @@ static int is_msi_quirked(struct pci_dev *dev) */ static int assign_interrupt_mode(struct pci_dev *dev, int *vectors, int mask) { - int i, pos, nvec, status = -EINVAL; - int interrupt_mode = PCIE_PORT_INTx_MODE; + struct pcie_port_data *port_data = pci_get_drvdata(dev); + int irq, interrupt_mode = PCIE_PORT_NO_IRQ; + int i; - /* Set INTx as default */ - for (i = 0, nvec = 0; i < PCIE_PORT_DEVICE_MAXSERVICES; i++) { - if (mask & (1 << i)) - nvec++; - vectors[i] = dev->irq; - } - /* Check MSI quirk */ - if (is_msi_quirked(dev)) - return interrupt_mode; - - /* Select MSI-X over MSI if supported */ - pos = pci_find_capability(dev, PCI_CAP_ID_MSIX); - if (pos) { - struct msix_entry msix_entries[PCIE_PORT_DEVICE_MAXSERVICES] = - {{0, 0}, {0, 1}, {0, 2}, {0, 3}}; - status = pci_enable_msix(dev, msix_entries, nvec); - if (!status) { - int j = 0; - - interrupt_mode = PCIE_PORT_MSIX_MODE; - for (i = 0; i < PCIE_PORT_DEVICE_MAXSERVICES; i++) { - if (mask & (1 << i)) - vectors[i] = msix_entries[j++].vector; - } - } - } - if (status) { - pos = pci_find_capability(dev, PCI_CAP_ID_MSI); - if (pos) { - status = pci_enable_msi(dev); - if (!status) { - interrupt_mode = PCIE_PORT_MSI_MODE; - for (i = 0;i < PCIE_PORT_DEVICE_MAXSERVICES;i++) - vectors[i] = dev->irq; - } - } - } + if (port_data->port_type == PCIE_RC_PORT && pcie_mch_quirk) + goto Fallback; + + /* Try to use MSI-X if supported */ + if (!pcie_port_enable_msix(dev, vectors, mask)) + return PCIE_PORT_MSIX_MODE; + + /* We're not going to use MSI-X, so try MSI and fall back to INTx */ + if (!pci_enable_msi(dev)) + interrupt_mode = PCIE_PORT_MSI_MODE; + + Fallback: + if (interrupt_mode == PCIE_PORT_NO_IRQ && dev->pin) + interrupt_mode = PCIE_PORT_INTx_MODE; + + irq = interrupt_mode != PCIE_PORT_NO_IRQ ? dev->irq : -1; + for (i = 0; i < PCIE_PORT_DEVICE_MAXSERVICES; i++) + vectors[i] = irq; + + vectors[PCIE_PORT_SERVICE_VC_SHIFT] = -1; + return interrupt_mode; } @@ -132,13 +240,11 @@ static int get_port_device_capability(struct pci_dev *dev) pos + PCIE_SLOT_CAPABILITIES_REG, ®32); if (reg32 & SLOT_HP_CAPABLE_MASK) services |= PCIE_PORT_SERVICE_HP; - } - /* PME Capable - root port capability */ - if (((reg16 >> 4) & PORT_TYPE_MASK) == PCIE_RC_PORT) - services |= PCIE_PORT_SERVICE_PME; - + } + /* AER capable */ if (pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR)) services |= PCIE_PORT_SERVICE_AER; + /* VC support */ if (pci_find_ext_capability(dev, PCI_EXT_CAP_ID_VC)) services |= PCIE_PORT_SERVICE_VC; @@ -152,27 +258,24 @@ static int get_port_device_capability(struct pci_dev *dev) * @port_type: Type of the port * @service_type: Type of service to associate with the service device * @irq: Interrupt vector to associate with the service device - * @irq_mode: Interrupt mode of the service (INTx, MSI-X, MSI) */ static void pcie_device_init(struct pci_dev *parent, struct pcie_device *dev, - int port_type, int service_type, int irq, int irq_mode) + int service_type, int irq) { + struct pcie_port_data *port_data = pci_get_drvdata(parent); struct device *device; + int port_type = port_data->port_type; dev->port = parent; - dev->interrupt_mode = irq_mode; dev->irq = irq; - dev->id.vendor = parent->vendor; - dev->id.device = parent->device; - dev->id.port_type = port_type; - dev->id.service_type = (1 << service_type); + dev->service = service_type; /* Initialize generic device interface */ device = &dev->device; 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)); @@ -185,10 +288,9 @@ static void pcie_device_init(struct pci_dev *parent, struct pcie_device *dev, * @port_type: Type of the port * @service_type: Type of service to associate with the service device * @irq: Interrupt vector to associate with the service device - * @irq_mode: Interrupt mode of the service (INTx, MSI-X, MSI) */ static struct pcie_device* alloc_pcie_device(struct pci_dev *parent, - int port_type, int service_type, int irq, int irq_mode) + int service_type, int irq) { struct pcie_device *device; @@ -196,7 +298,7 @@ static struct pcie_device* alloc_pcie_device(struct pci_dev *parent, if (!device) return NULL; - pcie_device_init(parent, device, port_type, service_type, irq,irq_mode); + pcie_device_init(parent, device, service_type, irq); return device; } @@ -230,63 +332,90 @@ int pcie_port_device_probe(struct pci_dev *dev) */ int pcie_port_device_register(struct pci_dev *dev) { - struct pcie_port_device_ext *p_ext; - int status, type, capabilities, irq_mode, i; + struct pcie_port_data *port_data; + int status, capabilities, irq_mode, i, nr_serv; int vectors[PCIE_PORT_DEVICE_MAXSERVICES]; u16 reg16; - /* Allocate port device extension */ - if (!(p_ext = kmalloc(sizeof(struct pcie_port_device_ext), GFP_KERNEL))) + port_data = kzalloc(sizeof(*port_data), GFP_KERNEL); + if (!port_data) return -ENOMEM; - - pci_set_drvdata(dev, p_ext); + pci_set_drvdata(dev, port_data); /* Get port type */ pci_read_config_word(dev, pci_find_capability(dev, PCI_CAP_ID_EXP) + PCIE_CAPABILITIES_REG, ®16); - type = (reg16 >> 4) & PORT_TYPE_MASK; + port_data->port_type = (reg16 >> 4) & PORT_TYPE_MASK; - /* Now get port services */ capabilities = get_port_device_capability(dev); + /* Root ports are capable of generating PME too */ + if (port_data->port_type == PCIE_RC_PORT) + capabilities |= PCIE_PORT_SERVICE_PME; + irq_mode = assign_interrupt_mode(dev, vectors, capabilities); - p_ext->interrupt_mode = irq_mode; + if (irq_mode == PCIE_PORT_NO_IRQ) { + /* + * Don't use service devices that require interrupts if there is + * no way to generate them. + */ + if (!(capabilities & PCIE_PORT_SERVICE_VC)) { + status = -ENODEV; + goto Error; + } + capabilities = PCIE_PORT_SERVICE_VC; + } + port_data->port_irq_mode = irq_mode; + + status = pci_enable_device(dev); + if (status) + goto Error; + pci_set_master(dev); /* Allocate child services if any */ - for (i = 0; i < PCIE_PORT_DEVICE_MAXSERVICES; i++) { + for (i = 0, nr_serv = 0; i < PCIE_PORT_DEVICE_MAXSERVICES; i++) { struct pcie_device *child; + int service = 1 << i; + + if (!(capabilities & service)) + continue; - if (capabilities & (1 << i)) { - child = alloc_pcie_device( - dev, /* parent */ - type, /* port type */ - i, /* service type */ - vectors[i], /* irq */ - irq_mode /* interrupt mode */); - if (child) { - status = device_register(&child->device); - if (status) { - kfree(child); - continue; - } - get_device(&child->device); - } + child = alloc_pcie_device(dev, service, vectors[i]); + if (!child) + continue; + + status = device_register(&child->device); + if (status) { + kfree(child); + continue; } + + get_device(&child->device); + nr_serv++; + } + if (!nr_serv) { + pci_disable_device(dev); + status = -ENODEV; + goto Error; } + return 0; + + Error: + kfree(port_data); + return status; } #ifdef CONFIG_PM static int suspend_iter(struct device *dev, void *data) { struct pcie_port_service_driver *service_driver; - pm_message_t state = * (pm_message_t *) data; if ((dev->bus == &pcie_port_bus_type) && (dev->driver)) { service_driver = to_service_driver(dev->driver); if (service_driver->suspend) - service_driver->suspend(to_pcie_device(dev), state); + service_driver->suspend(to_pcie_device(dev)); } return 0; } @@ -294,11 +423,10 @@ static int suspend_iter(struct device *dev, void *data) /** * pcie_port_device_suspend - suspend port services associated with a PCIe port * @dev: PCI Express port to handle - * @state: Representation of system power management transition in progress */ -int pcie_port_device_suspend(struct pci_dev *dev, pm_message_t state) +int pcie_port_device_suspend(struct device *dev) { - return device_for_each_child(&dev->dev, &state, suspend_iter); + return device_for_each_child(dev, NULL, suspend_iter); } static int resume_iter(struct device *dev, void *data) @@ -318,24 +446,17 @@ static int resume_iter(struct device *dev, void *data) * pcie_port_device_suspend - resume port services associated with a PCIe port * @dev: PCI Express port to handle */ -int pcie_port_device_resume(struct pci_dev *dev) +int pcie_port_device_resume(struct device *dev) { - return device_for_each_child(&dev->dev, NULL, resume_iter); + return device_for_each_child(dev, NULL, resume_iter); } -#endif +#endif /* PM */ static int remove_iter(struct device *dev, void *data) { - struct pcie_port_service_driver *service_driver; - if (dev->bus == &pcie_port_bus_type) { - if (dev->driver) { - service_driver = to_service_driver(dev->driver); - if (service_driver->remove) - service_driver->remove(to_pcie_device(dev)); - } - *(unsigned long*)data = (unsigned long)dev; - return 1; + put_device(dev); + device_unregister(dev); } return 0; } @@ -349,25 +470,21 @@ static int remove_iter(struct device *dev, void *data) */ void pcie_port_device_remove(struct pci_dev *dev) { - struct device *device; - unsigned long device_addr; - int interrupt_mode = PCIE_PORT_INTx_MODE; - int status; + struct pcie_port_data *port_data = pci_get_drvdata(dev); - do { - status = device_for_each_child(&dev->dev, &device_addr, remove_iter); - if (status) { - device = (struct device*)device_addr; - interrupt_mode = (to_pcie_device(device))->interrupt_mode; - put_device(device); - device_unregister(device); - } - } while (status); - /* Switch to INTx by default if MSI enabled */ - if (interrupt_mode == PCIE_PORT_MSIX_MODE) + device_for_each_child(&dev->dev, NULL, remove_iter); + pci_disable_device(dev); + + switch (port_data->port_irq_mode) { + case PCIE_PORT_MSIX_MODE: pci_disable_msix(dev); - else if (interrupt_mode == PCIE_PORT_MSI_MODE) + break; + case PCIE_PORT_MSI_MODE: pci_disable_msi(dev); + break; + } + + kfree(port_data); } /** @@ -392,7 +509,7 @@ static int pcie_port_probe_service(struct device *dev) return -ENODEV; pciedev = to_pcie_device(dev); - status = driver->probe(pciedev, driver->id_table); + status = driver->probe(pciedev); if (!status) { dev_printk(KERN_DEBUG, dev, "service driver %s loaded\n", driver->name); diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c index 5ea566e20b3..091ce70051e 100644 --- a/drivers/pci/pcie/portdrv_pci.c +++ b/drivers/pci/pcie/portdrv_pci.c @@ -32,11 +32,6 @@ MODULE_LICENSE("GPL"); /* global data */ static const char device_name[] = "pcieport-driver"; -static int pcie_portdrv_save_config(struct pci_dev *dev) -{ - return pci_save_state(dev); -} - static int pcie_portdrv_restore_config(struct pci_dev *dev) { int retval; @@ -49,21 +44,21 @@ static int pcie_portdrv_restore_config(struct pci_dev *dev) } #ifdef CONFIG_PM -static int pcie_portdrv_suspend(struct pci_dev *dev, pm_message_t state) -{ - return pcie_port_device_suspend(dev, state); +static struct dev_pm_ops pcie_portdrv_pm_ops = { + .suspend = pcie_port_device_suspend, + .resume = pcie_port_device_resume, + .freeze = pcie_port_device_suspend, + .thaw = pcie_port_device_resume, + .poweroff = pcie_port_device_suspend, + .restore = pcie_port_device_resume, +}; -} +#define PCIE_PORTDRV_PM_OPS (&pcie_portdrv_pm_ops) -static int pcie_portdrv_resume(struct pci_dev *dev) -{ - pci_set_master(dev); - return pcie_port_device_resume(dev); -} -#else -#define pcie_portdrv_suspend NULL -#define pcie_portdrv_resume NULL -#endif +#else /* !PM */ + +#define PCIE_PORTDRV_PM_OPS NULL +#endif /* !PM */ /* * pcie_portdrv_probe - Probe PCI-Express port devices @@ -82,20 +77,15 @@ static int __devinit pcie_portdrv_probe (struct pci_dev *dev, if (status) return status; - if (pci_enable_device(dev) < 0) - return -ENODEV; - - pci_set_master(dev); if (!dev->irq && dev->pin) { dev_warn(&dev->dev, "device [%04x:%04x] has invalid IRQ; " "check vendor BIOS\n", dev->vendor, dev->device); } - if (pcie_port_device_register(dev)) { - pci_disable_device(dev); - return -ENOMEM; - } + status = pcie_port_device_register(dev); + if (status) + return status; - pcie_portdrv_save_config(dev); + pci_save_state(dev); return 0; } @@ -104,7 +94,6 @@ static void pcie_portdrv_remove (struct pci_dev *dev) { pcie_port_device_remove(dev); pci_disable_device(dev); - kfree(pci_get_drvdata(dev)); } static int error_detected_iter(struct device *device, void *data) @@ -211,7 +200,7 @@ static int slot_reset_iter(struct device *device, void *data) static pci_ers_result_t pcie_portdrv_slot_reset(struct pci_dev *dev) { - pci_ers_result_t status = PCI_ERS_RESULT_NONE; + pci_ers_result_t status = PCI_ERS_RESULT_RECOVERED; int retval; /* If fatal, restore cfg space for possible link reset at upstream */ @@ -278,10 +267,9 @@ static struct pci_driver pcie_portdriver = { .probe = pcie_portdrv_probe, .remove = pcie_portdrv_remove, - .suspend = pcie_portdrv_suspend, - .resume = pcie_portdrv_resume, - .err_handler = &pcie_portdrv_err_handler, + + .driver.pm = PCIE_PORTDRV_PM_OPS, }; static int __init pcie_portdrv_init(void) diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 55ec44a27e8..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 (!dev) /* 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", @@ -511,21 +516,21 @@ int __devinit pci_scan_bridge(struct pci_bus *bus, struct pci_dev *dev, int max, /* * If we already got to this bus through a different bridge, - * ignore it. This can happen with the i450NX chipset. + * don't re-add it. This can happen with the i450NX chipset. + * + * However, we continue to descend down the hierarchy and + * scan remaining child buses. */ - if (pci_find_bus(pci_domain_nr(bus), busnr)) { - dev_info(&dev->dev, "bus %04x:%02x already known\n", - pci_domain_nr(bus), busnr); - goto out; + child = pci_find_bus(pci_domain_nr(bus), busnr); + if (!child) { + child = pci_add_new_bus(bus, dev, busnr); + if (!child) + goto out; + child->primary = buses & 0xFF; + child->subordinate = (buses >> 16) & 0xFF; + child->bridge_ctl = bctl; } - child = pci_add_new_bus(bus, dev, busnr); - if (!child) - goto out; - child->primary = buses & 0xFF; - child->subordinate = (buses >> 16) & 0xFF; - child->bridge_ctl = bctl; - cmax = pci_scan_child_bus(child); if (cmax > max) max = cmax; @@ -674,6 +679,19 @@ static void pci_read_irq(struct pci_dev *dev) dev->irq = irq; } +static void set_pcie_port_type(struct pci_dev *pdev) +{ + int pos; + u16 reg16; + + pos = pci_find_capability(pdev, PCI_CAP_ID_EXP); + if (!pos) + return; + pdev->is_pcie = 1; + pci_read_config_word(pdev, pos + PCI_EXP_FLAGS, ®16); + pdev->pcie_type = (reg16 & PCI_EXP_FLAGS_TYPE) >> 4; +} + #define LEGACY_IO_RESOURCE (IORESOURCE_IO | IORESOURCE_PCI_FIXED) /** @@ -683,12 +701,33 @@ static void pci_read_irq(struct pci_dev *dev) * Initialize the device structure with information about the device's * vendor,class,memory and IO-space addresses,IRQ lines etc. * Called at initialisation of the PCI subsystem and by CardBus services. - * Returns 0 on success and -1 if unknown type of device (not normal, bridge - * or CardBus). + * Returns 0 on success and negative if unknown type of device (not normal, + * bridge or CardBus). */ -static int pci_setup_device(struct pci_dev * dev) +int pci_setup_device(struct pci_dev *dev) { u32 class; + u8 hdr_type; + struct pci_slot *slot; + + if (pci_read_config_byte(dev, PCI_HEADER_TYPE, &hdr_type)) + return -EIO; + + dev->sysdata = dev->bus->sysdata; + dev->dev.parent = dev->bus->bridge; + dev->dev.bus = &pci_bus_type; + dev->hdr_type = hdr_type & 0x7f; + dev->multifunction = !!(hdr_type & 0x80); + dev->error_state = pci_channel_io_normal; + set_pcie_port_type(dev); + + list_for_each_entry(slot, &dev->bus->slots, list) + if (PCI_SLOT(dev->devfn) == slot->number) + dev->slot = slot; + + /* Assume 32-bit PCI; let 64-bit PCI cards (which are far rarer) + set this higher, assuming the system even supports it. */ + dev->dma_mask = 0xffffffff; dev_set_name(&dev->dev, "%04x:%02x:%02x.%d", pci_domain_nr(dev->bus), dev->bus->number, PCI_SLOT(dev->devfn), @@ -703,11 +742,15 @@ static int pci_setup_device(struct pci_dev * dev) dev_dbg(&dev->dev, "found [%04x:%04x] class %06x header type %02x\n", dev->vendor, dev->device, class, dev->hdr_type); + /* need to have dev->class ready */ + dev->cfg_size = pci_cfg_space_size(dev); + /* "Unknown power state" */ dev->current_state = PCI_UNKNOWN; /* Early fixups, before probing the BARs */ pci_fixup_device(pci_fixup_early, dev); + /* device class may be changed after fixup */ class = dev->class >> 8; switch (dev->hdr_type) { /* header type */ @@ -770,7 +813,7 @@ static int pci_setup_device(struct pci_dev * dev) default: /* unknown header */ dev_err(&dev->dev, "unknown header type %02x, " "ignoring device\n", dev->hdr_type); - return -1; + return -EIO; bad: dev_err(&dev->dev, "ignoring class %02x (doesn't match header " @@ -785,6 +828,7 @@ static int pci_setup_device(struct pci_dev * dev) static void pci_release_capabilities(struct pci_dev *dev) { pci_vpd_release(dev); + pci_iov_release(dev); } /** @@ -803,19 +847,6 @@ static void pci_release_dev(struct device *dev) kfree(pci_dev); } -static void set_pcie_port_type(struct pci_dev *pdev) -{ - int pos; - u16 reg16; - - pos = pci_find_capability(pdev, PCI_CAP_ID_EXP); - if (!pos) - return; - pdev->is_pcie = 1; - pci_read_config_word(pdev, pos + PCI_EXP_FLAGS, ®16); - pdev->pcie_type = (reg16 & PCI_EXP_FLAGS_TYPE) >> 4; -} - /** * pci_cfg_space_size - get the configuration space size of the PCI device. * @dev: PCI device @@ -847,6 +878,11 @@ int pci_cfg_space_size(struct pci_dev *dev) { int pos; u32 status; + u16 class; + + class = dev->class >> 8; + if (class == PCI_CLASS_BRIDGE_HOST) + return pci_cfg_space_size_ext(dev); pos = pci_find_capability(dev, PCI_CAP_ID_EXP); if (!pos) { @@ -891,9 +927,7 @@ EXPORT_SYMBOL(alloc_pci_dev); static struct pci_dev *pci_scan_device(struct pci_bus *bus, int devfn) { struct pci_dev *dev; - struct pci_slot *slot; u32 l; - u8 hdr_type; int delay = 1; if (pci_bus_read_config_dword(bus, devfn, PCI_VENDOR_ID, &l)) @@ -920,34 +954,16 @@ static struct pci_dev *pci_scan_device(struct pci_bus *bus, int devfn) } } - if (pci_bus_read_config_byte(bus, devfn, PCI_HEADER_TYPE, &hdr_type)) - return NULL; - dev = alloc_pci_dev(); if (!dev) return NULL; dev->bus = bus; - dev->sysdata = bus->sysdata; - dev->dev.parent = bus->bridge; - dev->dev.bus = &pci_bus_type; dev->devfn = devfn; - dev->hdr_type = hdr_type & 0x7f; - dev->multifunction = !!(hdr_type & 0x80); dev->vendor = l & 0xffff; dev->device = (l >> 16) & 0xffff; - dev->cfg_size = pci_cfg_space_size(dev); - dev->error_state = pci_channel_io_normal; - set_pcie_port_type(dev); - - list_for_each_entry(slot, &bus->slots, list) - if (PCI_SLOT(devfn) == slot->number) - dev->slot = slot; - /* Assume 32-bit PCI; let 64-bit PCI cards (which are far rarer) - set this higher, assuming the system even supports it. */ - dev->dma_mask = 0xffffffff; - if (pci_setup_device(dev) < 0) { + if (pci_setup_device(dev)) { kfree(dev); return NULL; } @@ -972,6 +988,9 @@ static void pci_init_capabilities(struct pci_dev *dev) /* Alternative Routing-ID Forwarding */ pci_enable_ari(dev); + + /* Single Root I/O Virtualization */ + pci_iov_init(dev); } void pci_device_add(struct pci_dev *dev, struct pci_bus *bus) @@ -1006,6 +1025,12 @@ struct pci_dev *__ref pci_scan_single_device(struct pci_bus *bus, int devfn) { struct pci_dev *dev; + dev = pci_get_slot(bus, devfn); + if (dev) { + pci_dev_put(dev); + return dev; + } + dev = pci_scan_device(bus, devfn); if (!dev) return NULL; @@ -1024,35 +1049,27 @@ EXPORT_SYMBOL(pci_scan_single_device); * Scan a PCI slot on the specified PCI bus for devices, adding * discovered devices to the @bus->devices list. New devices * will not have is_added set. + * + * Returns the number of new devices found. */ int pci_scan_slot(struct pci_bus *bus, int devfn) { - int func, nr = 0; - int scan_all_fns; - - scan_all_fns = pcibios_scan_all_fns(bus, devfn); - - for (func = 0; func < 8; func++, devfn++) { - struct pci_dev *dev; - - dev = pci_scan_single_device(bus, devfn); - if (dev) { - nr++; + int fn, nr = 0; + struct pci_dev *dev; - /* - * If this is a single function device, - * don't scan past the first function. - */ - if (!dev->multifunction) { - if (func > 0) { - dev->multifunction = 1; - } else { - break; - } + dev = pci_scan_single_device(bus, devfn); + if (dev && !dev->is_added) /* new device? */ + nr++; + + if ((dev && dev->multifunction) || + (!dev && pcibios_scan_all_fns(bus, devfn))) { + for (fn = 1; fn < 8; fn++) { + dev = pci_scan_single_device(bus, devfn + fn); + if (dev) { + if (!dev->is_added) + nr++; + dev->multifunction = 1; } - } else { - if (func == 0 && !scan_all_fns) - break; } } @@ -1074,12 +1091,21 @@ unsigned int __devinit pci_scan_child_bus(struct pci_bus *bus) for (devfn = 0; devfn < 0x100; devfn += 8) pci_scan_slot(bus, devfn); + /* Reserve buses for SR-IOV capability. */ + max += pci_iov_bus_range(bus); + /* * After performing arch-dependent fixup of the bus, look behind * all PCI-to-PCI bridges on this bus. */ - pr_debug("PCI: Fixups for bus %04x:%02x\n", pci_domain_nr(bus), bus->number); - pcibios_fixup_bus(bus); + if (!bus->is_added) { + pr_debug("PCI: Fixups for bus %04x:%02x\n", + pci_domain_nr(bus), bus->number); + pcibios_fixup_bus(bus); + if (pci_is_root_bus(bus)) + bus->is_added = 1; + } + for (pass=0; pass < 2; pass++) list_for_each_entry(dev, &bus->devices, bus_list) { if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE || @@ -1099,10 +1125,6 @@ unsigned int __devinit pci_scan_child_bus(struct pci_bus *bus) return max; } -void __attribute__((weak)) set_pci_bus_resources_arch_default(struct pci_bus *b) -{ -} - struct pci_bus * pci_create_bus(struct device *parent, int bus, struct pci_ops *ops, void *sysdata) { @@ -1114,7 +1136,7 @@ struct pci_bus * pci_create_bus(struct device *parent, if (!b) return NULL; - dev = kmalloc(sizeof(*dev), GFP_KERNEL); + dev = kzalloc(sizeof(*dev), GFP_KERNEL); if (!dev){ kfree(b); return NULL; @@ -1133,7 +1155,6 @@ struct pci_bus * pci_create_bus(struct device *parent, list_add_tail(&b->node, &pci_root_buses); up_write(&pci_bus_sem); - memset(dev, 0, sizeof(*dev)); dev->parent = parent; dev->release = pci_release_bus_bridge_dev; dev_set_name(dev, "pci%04x:%02x", pci_domain_nr(b), bus); @@ -1162,8 +1183,6 @@ struct pci_bus * pci_create_bus(struct device *parent, b->resource[0] = &ioport_resource; b->resource[1] = &iomem_resource; - set_pci_bus_resources_arch_default(b); - return b; dev_create_file_err: @@ -1193,6 +1212,38 @@ struct pci_bus * __devinit pci_scan_bus_parented(struct device *parent, EXPORT_SYMBOL(pci_scan_bus_parented); #ifdef CONFIG_HOTPLUG +/** + * pci_rescan_bus - scan a PCI bus for devices. + * @bus: PCI bus to scan + * + * Scan a PCI bus and child buses for new devices, adds them, + * and enables them. + * + * Returns the max number of subordinate bus discovered. + */ +unsigned int __ref pci_rescan_bus(struct pci_bus *bus) +{ + unsigned int max; + struct pci_dev *dev; + + max = pci_scan_child_bus(bus); + + down_read(&pci_bus_sem); + list_for_each_entry(dev, &bus->devices, bus_list) + if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE || + dev->hdr_type == PCI_HEADER_TYPE_CARDBUS) + if (dev->subordinate) + pci_bus_size_bridges(dev->subordinate); + up_read(&pci_bus_sem); + + pci_bus_assign_resources(bus); + pci_enable_bridges(bus); + pci_bus_add_devices(bus); + + return max; +} +EXPORT_SYMBOL_GPL(pci_rescan_bus); + EXPORT_SYMBOL(pci_add_new_bus); EXPORT_SYMBOL(pci_scan_slot); EXPORT_SYMBOL(pci_scan_bridge); diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 92b9efe9bca..56552d74abe 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -24,6 +24,7 @@ #include <linux/kallsyms.h> #include <linux/dmi.h> #include <linux/pci-aspm.h> +#include <linux/ioport.h> #include "pci.h" int isa_dma_bridge_buggy; @@ -34,6 +35,69 @@ int pcie_mch_quirk; EXPORT_SYMBOL(pcie_mch_quirk); #ifdef CONFIG_PCI_QUIRKS +/* + * This quirk function disables memory decoding and releases memory resources + * of the device specified by kernel's boot parameter 'pci=resource_alignment='. + * It also rounds up size to specified alignment. + * Later on, the kernel will assign page-aligned memory resource back + * to the device. + */ +static void __devinit quirk_resource_alignment(struct pci_dev *dev) +{ + int i; + struct resource *r; + resource_size_t align, size; + u16 command; + + if (!pci_is_reassigndev(dev)) + return; + + if (dev->hdr_type == PCI_HEADER_TYPE_NORMAL && + (dev->class >> 8) == PCI_CLASS_BRIDGE_HOST) { + dev_warn(&dev->dev, + "Can't reassign resources to host bridge.\n"); + return; + } + + dev_info(&dev->dev, + "Disabling memory decoding and releasing memory resources.\n"); + pci_read_config_word(dev, PCI_COMMAND, &command); + command &= ~PCI_COMMAND_MEMORY; + pci_write_config_word(dev, PCI_COMMAND, command); + + align = pci_specified_resource_alignment(dev); + for (i=0; i < PCI_BRIDGE_RESOURCES; i++) { + r = &dev->resource[i]; + if (!(r->flags & IORESOURCE_MEM)) + continue; + size = resource_size(r); + if (size < align) { + size = align; + dev_info(&dev->dev, + "Rounding up size of resource #%d to %#llx.\n", + i, (unsigned long long)size); + } + r->end = size - 1; + r->start = 0; + } + /* Need to disable bridge's resource window, + * to enable the kernel to reassign new resource + * window later on. + */ + if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE && + (dev->class >> 8) == PCI_CLASS_BRIDGE_PCI) { + for (i = PCI_BRIDGE_RESOURCES; i < PCI_NUM_RESOURCES; i++) { + r = &dev->resource[i]; + if (!(r->flags & IORESOURCE_MEM)) + continue; + r->end = resource_size(r) - 1; + r->start = 0; + } + pci_disable_bridge_window(dev); + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, quirk_resource_alignment); + /* The Mellanox Tavor device gives false positive parity errors * Mark this device with a broken_parity_status, to allow * PCI scanning code to "skip" this now blacklisted device. @@ -1069,6 +1133,7 @@ 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) @@ -1099,6 +1164,7 @@ 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) @@ -1126,10 +1192,15 @@ static void __init asus_hides_smbus_hostbridge(struct pci_dev *dev) * its on-board VGA controller */ asus_hides_smbus = 1; } - else if (dev->device == PCI_DEVICE_ID_INTEL_82845G_IG) + else if (dev->device == PCI_DEVICE_ID_INTEL_82801DB_2) switch(dev->subsystem_device) { case 0x00b8: /* Compaq Evo D510 CMT */ case 0x00b9: /* Compaq Evo D510 SFF */ + /* Motherboard doesn't have Host bridge + * subvendor/subdevice IDs and on-board VGA + * controller is disabled if an AGP card is + * inserted, therefore checking USB UHCI + * Controller #1 */ asus_hides_smbus = 1; } else if (dev->device == PCI_DEVICE_ID_INTEL_82815_CGC) @@ -1154,7 +1225,7 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82855GM_HB, as DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82915GM_HB, asus_hides_smbus_hostbridge); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82810_IG3, asus_hides_smbus_hostbridge); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82845G_IG, asus_hides_smbus_hostbridge); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801DB_2, asus_hides_smbus_hostbridge); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82815_CGC, asus_hides_smbus_hostbridge); static void asus_hides_smbus_lpc(struct pci_dev *dev) @@ -1664,9 +1735,13 @@ static void __devinit quirk_netmos(struct pci_dev *dev) * of parallel ports and <S> is the number of serial ports. */ switch (dev->device) { + case PCI_DEVICE_ID_NETMOS_9835: + /* Well, this rule doesn't hold for the following 9835 device */ + if (dev->subsystem_vendor == PCI_VENDOR_ID_IBM && + dev->subsystem_device == 0x0299) + return; case PCI_DEVICE_ID_NETMOS_9735: case PCI_DEVICE_ID_NETMOS_9745: - case PCI_DEVICE_ID_NETMOS_9835: case PCI_DEVICE_ID_NETMOS_9845: case PCI_DEVICE_ID_NETMOS_9855: if ((dev->class >> 8) == PCI_CLASS_COMMUNICATION_SERIAL && @@ -1943,6 +2018,28 @@ 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 @@ -1960,6 +2057,7 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_RS400_200, quirk_di DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_RS480, quirk_disable_all_msi); DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_VT3336, quirk_disable_all_msi); DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_VT3351, quirk_disable_all_msi); +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_VT3364, quirk_disable_all_msi); /* Disable MSI on chipsets that are known to not support it */ static void __devinit quirk_disable_msi(struct pci_dev *dev) @@ -2078,6 +2176,92 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NVENET_15, nvenet_msi_disable); +static int __devinit ht_check_msi_mapping(struct pci_dev *dev) +{ + int pos, ttl = 48; + int found = 0; + + /* check if there is HT MSI cap or enabled on this device */ + pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); + while (pos && ttl--) { + u8 flags; + + if (found < 1) + found = 1; + if (pci_read_config_byte(dev, pos + HT_MSI_FLAGS, + &flags) == 0) { + if (flags & HT_MSI_FLAGS_ENABLE) { + if (found < 2) { + found = 2; + break; + } + } + } + pos = pci_find_next_ht_capability(dev, pos, + HT_CAPTYPE_MSI_MAPPING); + } + + return found; +} + +static int __devinit host_bridge_with_leaf(struct pci_dev *host_bridge) +{ + struct pci_dev *dev; + int pos; + int i, dev_no; + int found = 0; + + dev_no = host_bridge->devfn >> 3; + for (i = dev_no + 1; i < 0x20; i++) { + dev = pci_get_slot(host_bridge->bus, PCI_DEVFN(i, 0)); + if (!dev) + continue; + + /* found next host bridge ?*/ + pos = pci_find_ht_capability(dev, HT_CAPTYPE_SLAVE); + if (pos != 0) { + pci_dev_put(dev); + break; + } + + if (ht_check_msi_mapping(dev)) { + found = 1; + pci_dev_put(dev); + break; + } + pci_dev_put(dev); + } + + return found; +} + +#define PCI_HT_CAP_SLAVE_CTRL0 4 /* link control */ +#define PCI_HT_CAP_SLAVE_CTRL1 8 /* link control to */ + +static int __devinit is_end_of_ht_chain(struct pci_dev *dev) +{ + int pos, ctrl_off; + int end = 0; + u16 flags, ctrl; + + pos = pci_find_ht_capability(dev, HT_CAPTYPE_SLAVE); + + if (!pos) + goto out; + + pci_read_config_word(dev, pos + PCI_CAP_FLAGS, &flags); + + ctrl_off = ((flags >> 10) & 1) ? + PCI_HT_CAP_SLAVE_CTRL0 : PCI_HT_CAP_SLAVE_CTRL1; + pci_read_config_word(dev, pos + ctrl_off, &ctrl); + + if (ctrl & (1 << 6)) + end = 1; + +out: + return end; +} + static void __devinit nv_ht_enable_msi_mapping(struct pci_dev *dev) { struct pci_dev *host_bridge; @@ -2102,6 +2286,11 @@ static void __devinit nv_ht_enable_msi_mapping(struct pci_dev *dev) if (!found) return; + /* don't enable end_device/host_bridge with leaf directly here */ + if (host_bridge == dev && is_end_of_ht_chain(host_bridge) && + host_bridge_with_leaf(host_bridge)) + goto out; + /* root did that ! */ if (msi_ht_cap_enabled(host_bridge)) goto out; @@ -2132,44 +2321,12 @@ static void __devinit ht_disable_msi_mapping(struct pci_dev *dev) } } -static int __devinit ht_check_msi_mapping(struct pci_dev *dev) -{ - int pos, ttl = 48; - int found = 0; - - /* check if there is HT MSI cap or enabled on this device */ - pos = pci_find_ht_capability(dev, HT_CAPTYPE_MSI_MAPPING); - while (pos && ttl--) { - u8 flags; - - if (found < 1) - found = 1; - if (pci_read_config_byte(dev, pos + HT_MSI_FLAGS, - &flags) == 0) { - if (flags & HT_MSI_FLAGS_ENABLE) { - if (found < 2) { - found = 2; - break; - } - } - } - pos = pci_find_next_ht_capability(dev, pos, - HT_CAPTYPE_MSI_MAPPING); - } - - return found; -} - -static void __devinit nv_msi_ht_cap_quirk(struct pci_dev *dev) +static void __devinit __nv_msi_ht_cap_quirk(struct pci_dev *dev, int all) { struct pci_dev *host_bridge; int pos; int found; - /* Enabling HT MSI mapping on this device breaks MCP51 */ - if (dev->device == 0x270) - return; - /* check if there is HT MSI cap or enabled on this device */ found = ht_check_msi_mapping(dev); @@ -2193,7 +2350,10 @@ static void __devinit nv_msi_ht_cap_quirk(struct pci_dev *dev) /* Host bridge is to HT */ if (found == 1) { /* it is not enabled, try to enable it */ - nv_ht_enable_msi_mapping(dev); + if (all) + ht_enable_msi_mapping(dev); + else + nv_ht_enable_msi_mapping(dev); } return; } @@ -2205,8 +2365,20 @@ static void __devinit nv_msi_ht_cap_quirk(struct pci_dev *dev) /* Host bridge is not to HT, disable HT MSI mapping on this device */ ht_disable_msi_mapping(dev); } -DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA, PCI_ANY_ID, nv_msi_ht_cap_quirk); -DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AL, PCI_ANY_ID, nv_msi_ht_cap_quirk); + +static void __devinit nv_msi_ht_cap_quirk_all(struct pci_dev *dev) +{ + return __nv_msi_ht_cap_quirk(dev, 1); +} + +static void __devinit nv_msi_ht_cap_quirk_leaf(struct pci_dev *dev) +{ + return __nv_msi_ht_cap_quirk(dev, 0); +} + +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA, PCI_ANY_ID, nv_msi_ht_cap_quirk_leaf); + +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AL, PCI_ANY_ID, nv_msi_ht_cap_quirk_all); static void __devinit quirk_msi_intx_disable_bug(struct pci_dev *dev) { @@ -2268,6 +2440,56 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x4375, #endif /* CONFIG_PCI_MSI */ +#ifdef CONFIG_PCI_IOV + +/* + * For Intel 82576 SR-IOV NIC, if BIOS doesn't allocate resources for the + * SR-IOV BARs, zero the Flash BAR and program the SR-IOV BARs to use the + * old Flash Memory Space. + */ +static void __devinit quirk_i82576_sriov(struct pci_dev *dev) +{ + int pos, flags; + u32 bar, start, size; + + if (PAGE_SIZE > 0x10000) + return; + + flags = pci_resource_flags(dev, 0); + if ((flags & PCI_BASE_ADDRESS_SPACE) != + PCI_BASE_ADDRESS_SPACE_MEMORY || + (flags & PCI_BASE_ADDRESS_MEM_TYPE_MASK) != + PCI_BASE_ADDRESS_MEM_TYPE_32) + return; + + pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_SRIOV); + if (!pos) + return; + + pci_read_config_dword(dev, pos + PCI_SRIOV_BAR, &bar); + if (bar & PCI_BASE_ADDRESS_MEM_MASK) + return; + + start = pci_resource_start(dev, 1); + size = pci_resource_len(dev, 1); + if (!start || size != 0x400000 || start & (size - 1)) + return; + + pci_resource_flags(dev, 1) = 0; + pci_write_config_dword(dev, PCI_BASE_ADDRESS_1, 0); + pci_write_config_dword(dev, pos + PCI_SRIOV_BAR, start); + pci_write_config_dword(dev, pos + PCI_SRIOV_BAR + 12, start + size / 2); + + dev_info(&dev->dev, "use Flash Memory Space for SR-IOV BARs\n"); +} +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 */ + static void pci_do_fixups(struct pci_dev *dev, struct pci_fixup *f, struct pci_fixup *end) { diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c index 042e0892442..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); @@ -71,6 +69,9 @@ void pci_remove_bus(struct pci_bus *pci_bus) down_write(&pci_bus_sem); list_del(&pci_bus->node); up_write(&pci_bus_sem); + if (!pci_bus->is_added) + return; + pci_remove_legacy_files(pci_bus); device_remove_file(&pci_bus->dev, &dev_attr_cpuaffinity); device_remove_file(&pci_bus->dev, &dev_attr_cpulistaffinity); @@ -92,6 +93,7 @@ EXPORT_SYMBOL(pci_remove_bus); */ void pci_remove_bus_device(struct pci_dev *dev) { + pci_stop_bus_device(dev); if (dev->subordinate) { struct pci_bus *b = dev->subordinate; diff --git a/drivers/pci/search.c b/drivers/pci/search.c index 5af8bd53814..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->self) + 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 70460894578..b636e245445 100644 --- a/drivers/pci/setup-bus.c +++ b/drivers/pci/setup-bus.c @@ -27,7 +27,7 @@ #include <linux/slab.h> -static void pbus_assign_resources_sorted(struct pci_bus *bus) +static void pbus_assign_resources_sorted(const struct pci_bus *bus) { struct pci_dev *dev; struct resource *res; @@ -58,7 +58,6 @@ static void pbus_assign_resources_sorted(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 +142,10 @@ 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; dev_info(&bridge->dev, "PCI bridge, secondary bus %04x:%02x\n", pci_domain_nr(bus), bus->number); @@ -195,16 +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; @@ -212,9 +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); } @@ -252,8 +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 @@ -333,6 +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; @@ -341,9 +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; @@ -369,6 +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; } } @@ -393,6 +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; } @@ -495,7 +529,7 @@ void __ref pci_bus_size_bridges(struct pci_bus *bus) } EXPORT_SYMBOL(pci_bus_size_bridges); -void __ref pci_bus_assign_resources(struct pci_bus *bus) +void __ref pci_bus_assign_resources(const struct pci_bus *bus) { struct pci_bus *b; struct pci_dev *dev; @@ -533,11 +567,13 @@ static void pci_bus_dump_res(struct pci_bus *bus) for (i = 0; i < PCI_BUS_NUM_RESOURCES; i++) { struct resource *res = bus->resource[i]; - if (!res) + if (!res || !res->end) continue; dev_printk(KERN_DEBUG, &bus->dev, "resource %d %s %pR\n", i, - (res->flags & IORESOURCE_IO) ? "io: " : "mem:", res); + (res->flags & IORESOURCE_IO) ? "io: " : + ((res->flags & IORESOURCE_PREFETCH)? "pref mem":"mem:"), + res); } } diff --git a/drivers/pci/setup-res.c b/drivers/pci/setup-res.c index 32e8d88a461..b711fb7181e 100644 --- a/drivers/pci/setup-res.c +++ b/drivers/pci/setup-res.c @@ -99,11 +99,11 @@ 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; + struct resource *root; char *dtype = resource < PCI_BRIDGE_RESOURCES ? "device" : "bridge"; int err; - root = pcibios_select_root(dev, res); + root = pci_find_parent_resource(dev, res); err = -EINVAL; if (root != NULL) @@ -120,23 +120,31 @@ int pci_claim_resource(struct pci_dev *dev, int resource) return err; } -int pci_assign_resource(struct pci_dev *dev, int resno) +#ifdef CONFIG_PCI_QUIRKS +void pci_disable_bridge_window(struct pci_dev *dev) +{ + dev_dbg(&dev->dev, "Disabling bridge window.\n"); + + /* MMIO Base/Limit */ + pci_write_config_dword(dev, PCI_MEMORY_BASE, 0x0000fff0); + + /* Prefetchable MMIO Base/Limit */ + pci_write_config_dword(dev, PCI_PREF_LIMIT_UPPER32, 0); + pci_write_config_dword(dev, PCI_PREF_MEMORY_BASE, 0x0000fff0); + pci_write_config_dword(dev, PCI_PREF_BASE_UPPER32, 0xffffffff); +} +#endif /* CONFIG_PCI_QUIRKS */ + +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, @@ -154,10 +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); @@ -166,6 +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 5a8ccb4f604..eddb0748b0e 100644 --- a/drivers/pci/slot.c +++ b/drivers/pci/slot.c @@ -1,8 +1,8 @@ /* * drivers/pci/slot.c * Copyright (C) 2006 Matthew Wilcox <matthew@wil.cx> - * Copyright (C) 2006-2008 Hewlett-Packard Development Company, L.P. - * Alex Chiang <achiang@hp.com> + * Copyright (C) 2006-2009 Hewlett-Packard Development Company, L.P. + * Alex Chiang <achiang@hp.com> */ #include <linux/kobject.h> @@ -52,8 +52,8 @@ static void pci_slot_release(struct kobject *kobj) struct pci_dev *dev; struct pci_slot *slot = to_pci_slot(kobj); - pr_debug("%s: releasing pci_slot on %x:%d\n", __func__, - slot->bus->number, slot->number); + dev_dbg(&slot->bus->dev, "dev %02x, released physical slot %s\n", + slot->number, pci_slot_name(slot)); list_for_each_entry(dev, &slot->bus->devices, bus_list) if (PCI_SLOT(dev->devfn) == slot->number) @@ -248,9 +248,8 @@ placeholder: if (PCI_SLOT(dev->devfn) == slot_nr) dev->slot = slot; - /* Don't care if debug printk has a -1 for slot_nr */ - pr_debug("%s: created pci_slot on %04x:%02x:%02x\n", - __func__, pci_domain_nr(parent), parent->number, slot_nr); + dev_dbg(&parent->dev, "dev %02x, created physical slot %s\n", + slot_nr, pci_slot_name(slot)); out: kfree(slot_name); @@ -265,8 +264,8 @@ EXPORT_SYMBOL_GPL(pci_create_slot); /** * pci_renumber_slot - update %struct pci_slot -> number - * @slot - %struct pci_slot to update - * @slot_nr - new number for slot + * @slot: &struct pci_slot to update + * @slot_nr: new number for slot * * The primary purpose of this interface is to allow callers who earlier * created a placeholder slot in pci_create_slot() by passing a -1 as @@ -299,9 +298,8 @@ EXPORT_SYMBOL_GPL(pci_renumber_slot); */ void pci_destroy_slot(struct pci_slot *slot) { - pr_debug("%s: dec refcount to %d on %04x:%02x:%02x\n", __func__, - atomic_read(&slot->kobj.kref.refcount) - 1, - pci_domain_nr(slot->bus), slot->bus->number, slot->number); + dev_dbg(&slot->bus->dev, "dev %02x, dec refcount to %d\n", + slot->number, atomic_read(&slot->kobj.kref.refcount) - 1); down_write(&pci_bus_sem); kobject_put(&slot->kobj); @@ -309,6 +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. + * @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. + * @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; |