diff options
Diffstat (limited to 'arch/x86/kernel')
32 files changed, 1958 insertions, 1726 deletions
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 30909a258d0..85eb8e10081 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -809,7 +809,7 @@ void clear_local_APIC(void) u32 v; /* APIC hasn't been mapped yet */ - if (!apic_phys) + if (!x2apic && !apic_phys) return; maxlvt = lapic_get_maxlvt(); @@ -1334,15 +1334,16 @@ void __init enable_IR_x2apic(void) return; } - local_irq_save(flags); - mask_8259A(); - - ret = save_mask_IO_APIC_setup(); + ret = save_IO_APIC_setup(); if (ret) { pr_info("Saving IO-APIC state failed: %d\n", ret); goto end; } + local_irq_save(flags); + mask_IO_APIC_setup(); + mask_8259A(); + ret = enable_intr_remapping(1); if (ret && x2apic_preenabled) { @@ -1367,10 +1368,10 @@ end_restore: else reinit_intr_remapped_IO_APIC(x2apic_preenabled); -end: unmask_8259A(); local_irq_restore(flags); +end: if (!ret) { if (!x2apic_preenabled) pr_info("Enabled x2apic and interrupt-remapping\n"); @@ -1523,12 +1524,10 @@ void __init early_init_lapic_mapping(void) */ void __init init_apic_mappings(void) { -#ifdef CONFIG_X86_X2APIC if (x2apic) { boot_cpu_physical_apicid = read_apic_id(); return; } -#endif /* * If no local APIC can be found then set up a fake all @@ -1972,12 +1971,9 @@ static int lapic_resume(struct sys_device *dev) local_irq_save(flags); -#ifdef CONFIG_X86_X2APIC if (x2apic) enable_x2apic(); - else -#endif - { + else { /* * Make sure the APICBASE points to the right address * diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 00e6071cefc..42cdc78427a 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -389,6 +389,8 @@ struct io_apic { unsigned int index; unsigned int unused[3]; unsigned int data; + unsigned int unused2[11]; + unsigned int eoi; }; static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) @@ -397,6 +399,12 @@ static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) + (mp_ioapics[idx].apicaddr & ~PAGE_MASK); } +static inline void io_apic_eoi(unsigned int apic, unsigned int vector) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(vector, &io_apic->eoi); +} + static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) { struct io_apic __iomem *io_apic = io_apic_base(apic); @@ -546,16 +554,12 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq apic = entry->apic; pin = entry->pin; -#ifdef CONFIG_INTR_REMAP /* * With interrupt-remapping, destination information comes * from interrupt-remapping table entry. */ if (!irq_remapped(irq)) io_apic_write(apic, 0x11 + pin*2, dest); -#else - io_apic_write(apic, 0x11 + pin*2, dest); -#endif reg = io_apic_read(apic, 0x10 + pin*2); reg &= ~IO_APIC_REDIR_VECTOR_MASK; reg |= vector; @@ -849,9 +853,9 @@ __setup("pirq=", ioapic_pirq_setup); static struct IO_APIC_route_entry *early_ioapic_entries[MAX_IO_APICS]; /* - * Saves and masks all the unmasked IO-APIC RTE's + * Saves all the IO-APIC RTE's */ -int save_mask_IO_APIC_setup(void) +int save_IO_APIC_setup(void) { union IO_APIC_reg_01 reg_01; unsigned long flags; @@ -876,16 +880,9 @@ int save_mask_IO_APIC_setup(void) } for (apic = 0; apic < nr_ioapics; apic++) - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { - struct IO_APIC_route_entry entry; - - entry = early_ioapic_entries[apic][pin] = + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) + early_ioapic_entries[apic][pin] = ioapic_read_entry(apic, pin); - if (!entry.mask) { - entry.mask = 1; - ioapic_write_entry(apic, pin, entry); - } - } return 0; @@ -898,6 +895,25 @@ nomem: return -ENOMEM; } +void mask_IO_APIC_setup(void) +{ + int apic, pin; + + for (apic = 0; apic < nr_ioapics; apic++) { + if (!early_ioapic_entries[apic]) + break; + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + struct IO_APIC_route_entry entry; + + entry = early_ioapic_entries[apic][pin]; + if (!entry.mask) { + entry.mask = 1; + ioapic_write_entry(apic, pin, entry); + } + } + } +} + void restore_IO_APIC_setup(void) { int apic, pin; @@ -1411,9 +1427,8 @@ void __setup_vector_irq(int cpu) } static struct irq_chip ioapic_chip; -#ifdef CONFIG_INTR_REMAP static struct irq_chip ir_ioapic_chip; -#endif +static struct irq_chip msi_ir_chip; #define IOAPIC_AUTO -1 #define IOAPIC_EDGE 0 @@ -1452,7 +1467,6 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t else desc->status &= ~IRQ_LEVEL; -#ifdef CONFIG_INTR_REMAP if (irq_remapped(irq)) { desc->status |= IRQ_MOVE_PCNTXT; if (trigger) @@ -1464,7 +1478,7 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t handle_edge_irq, "edge"); return; } -#endif + if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || trigger == IOAPIC_LEVEL) set_irq_chip_and_handler_name(irq, &ioapic_chip, @@ -1478,14 +1492,13 @@ static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long t int setup_ioapic_entry(int apic_id, int irq, struct IO_APIC_route_entry *entry, unsigned int destination, int trigger, - int polarity, int vector) + int polarity, int vector, int pin) { /* * add it to the IO-APIC irq-routing table: */ memset(entry,0,sizeof(*entry)); -#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) { struct intel_iommu *iommu = map_ioapic_to_ir(apic_id); struct irte irte; @@ -1504,7 +1517,14 @@ int setup_ioapic_entry(int apic_id, int irq, irte.present = 1; irte.dst_mode = apic->irq_dest_mode; - irte.trigger_mode = trigger; + /* + * Trigger mode in the IRTE will always be edge, and the + * actual level or edge trigger will be setup in the IO-APIC + * RTE. This will help simplify level triggered irq migration. + * For more details, see the comments above explainig IO-APIC + * irq migration in the presence of interrupt-remapping. + */ + irte.trigger_mode = 0; irte.dlvry_mode = apic->irq_delivery_mode; irte.vector = vector; irte.dest_id = IRTE_DEST(destination); @@ -1515,18 +1535,21 @@ int setup_ioapic_entry(int apic_id, int irq, ir_entry->zero = 0; ir_entry->format = 1; ir_entry->index = (index & 0x7fff); - } else -#endif - { + /* + * IO-APIC RTE will be configured with virtual vector. + * irq handler will do the explicit EOI to the io-apic. + */ + ir_entry->vector = pin; + } else { entry->delivery_mode = apic->irq_delivery_mode; entry->dest_mode = apic->irq_dest_mode; entry->dest = destination; + entry->vector = vector; } entry->mask = 0; /* enable IRQ */ entry->trigger = trigger; entry->polarity = polarity; - entry->vector = vector; /* Mask level triggered irqs. * Use IRQ_DELAYED_DISABLE for edge triggered irqs. @@ -1561,7 +1584,7 @@ static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq if (setup_ioapic_entry(mp_ioapics[apic_id].apicid, irq, &entry, - dest, trigger, polarity, cfg->vector)) { + dest, trigger, polarity, cfg->vector, pin)) { printk("Failed to setup ioapic entry for ioapic %d, pin %d\n", mp_ioapics[apic_id].apicid, pin); __clear_irq_vector(irq, cfg); @@ -1642,10 +1665,8 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic_id, unsigned int pin, { struct IO_APIC_route_entry entry; -#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) return; -#endif memset(&entry, 0, sizeof(entry)); @@ -2040,8 +2061,13 @@ void disable_IO_APIC(void) * If the i8259 is routed through an IOAPIC * Put that IOAPIC in virtual wire mode * so legacy interrupts can be delivered. + * + * With interrupt-remapping, for now we will use virtual wire A mode, + * as virtual wire B is little complex (need to configure both + * IOAPIC RTE aswell as interrupt-remapping table entry). + * As this gets called during crash dump, keep this simple for now. */ - if (ioapic_i8259.pin != -1) { + if (ioapic_i8259.pin != -1 && !intr_remapping_enabled) { struct IO_APIC_route_entry entry; memset(&entry, 0, sizeof(entry)); @@ -2061,7 +2087,10 @@ void disable_IO_APIC(void) ioapic_write_entry(ioapic_i8259.apic, ioapic_i8259.pin, entry); } - disconnect_bsp_APIC(ioapic_i8259.pin != -1); + /* + * Use virtual wire A mode when interrupt remapping is enabled. + */ + disconnect_bsp_APIC(!intr_remapping_enabled && ioapic_i8259.pin != -1); } #ifdef CONFIG_X86_32 @@ -2303,37 +2332,24 @@ static int ioapic_retrigger_irq(unsigned int irq) #ifdef CONFIG_SMP #ifdef CONFIG_INTR_REMAP -static void ir_irq_migration(struct work_struct *work); - -static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration); /* * Migrate the IO-APIC irq in the presence of intr-remapping. * - * For edge triggered, irq migration is a simple atomic update(of vector - * and cpu destination) of IRTE and flush the hardware cache. - * - * For level triggered, we need to modify the io-apic RTE aswell with the update - * vector information, along with modifying IRTE with vector and destination. - * So irq migration for level triggered is little bit more complex compared to - * edge triggered migration. But the good news is, we use the same algorithm - * for level triggered migration as we have today, only difference being, - * we now initiate the irq migration from process context instead of the - * interrupt context. + * For both level and edge triggered, irq migration is a simple atomic + * update(of vector and cpu destination) of IRTE and flush the hardware cache. * - * In future, when we do a directed EOI (combined with cpu EOI broadcast - * suppression) to the IO-APIC, level triggered irq migration will also be - * as simple as edge triggered migration and we can do the irq migration - * with a simple atomic update to IO-APIC RTE. + * For level triggered, we eliminate the io-apic RTE modification (with the + * updated vector information), by using a virtual vector (io-apic pin number). + * Real vector that is used for interrupting cpu will be coming from + * the interrupt-remapping table entry. */ static void migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) { struct irq_cfg *cfg; struct irte irte; - int modify_ioapic_rte; unsigned int dest; - unsigned long flags; unsigned int irq; if (!cpumask_intersects(mask, cpu_online_mask)) @@ -2351,13 +2367,6 @@ migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) dest = apic->cpu_mask_to_apicid_and(cfg->domain, mask); - modify_ioapic_rte = desc->status & IRQ_LEVEL; - if (modify_ioapic_rte) { - spin_lock_irqsave(&ioapic_lock, flags); - __target_IO_APIC_irq(irq, dest, cfg); - spin_unlock_irqrestore(&ioapic_lock, flags); - } - irte.vector = cfg->vector; irte.dest_id = IRTE_DEST(dest); @@ -2372,73 +2381,12 @@ migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) cpumask_copy(desc->affinity, mask); } -static int migrate_irq_remapped_level_desc(struct irq_desc *desc) -{ - int ret = -1; - struct irq_cfg *cfg = desc->chip_data; - - mask_IO_APIC_irq_desc(desc); - - if (io_apic_level_ack_pending(cfg)) { - /* - * Interrupt in progress. Migrating irq now will change the - * vector information in the IO-APIC RTE and that will confuse - * the EOI broadcast performed by cpu. - * So, delay the irq migration to the next instance. - */ - schedule_delayed_work(&ir_migration_work, 1); - goto unmask; - } - - /* everthing is clear. we have right of way */ - migrate_ioapic_irq_desc(desc, desc->pending_mask); - - ret = 0; - desc->status &= ~IRQ_MOVE_PENDING; - cpumask_clear(desc->pending_mask); - -unmask: - unmask_IO_APIC_irq_desc(desc); - - return ret; -} - -static void ir_irq_migration(struct work_struct *work) -{ - unsigned int irq; - struct irq_desc *desc; - - for_each_irq_desc(irq, desc) { - if (desc->status & IRQ_MOVE_PENDING) { - unsigned long flags; - - spin_lock_irqsave(&desc->lock, flags); - if (!desc->chip->set_affinity || - !(desc->status & IRQ_MOVE_PENDING)) { - desc->status &= ~IRQ_MOVE_PENDING; - spin_unlock_irqrestore(&desc->lock, flags); - continue; - } - - desc->chip->set_affinity(irq, desc->pending_mask); - spin_unlock_irqrestore(&desc->lock, flags); - } - } -} - /* * Migrates the IRQ destination in the process context. */ static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask) { - if (desc->status & IRQ_LEVEL) { - desc->status |= IRQ_MOVE_PENDING; - cpumask_copy(desc->pending_mask, mask); - migrate_irq_remapped_level_desc(desc); - return; - } - migrate_ioapic_irq_desc(desc, mask); } static void set_ir_ioapic_affinity_irq(unsigned int irq, @@ -2448,6 +2396,11 @@ static void set_ir_ioapic_affinity_irq(unsigned int irq, set_ir_ioapic_affinity_irq_desc(desc, mask); } +#else +static inline void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, + const struct cpumask *mask) +{ +} #endif asmlinkage void smp_irq_move_cleanup_interrupt(void) @@ -2461,6 +2414,7 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void) me = smp_processor_id(); for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; vector++) { unsigned int irq; + unsigned int irr; struct irq_desc *desc; struct irq_cfg *cfg; irq = __get_cpu_var(vector_irq)[vector]; @@ -2480,6 +2434,18 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void) if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) goto unlock; + irr = apic_read(APIC_IRR + (vector / 32 * 0x10)); + /* + * Check if the vector that needs to be cleanedup is + * registered at the cpu's IRR. If so, then this is not + * the best time to clean it up. Lets clean it up in the + * next attempt by sending another IRQ_MOVE_CLEANUP_VECTOR + * to myself. + */ + if (irr & (1 << (vector % 32))) { + apic->send_IPI_self(IRQ_MOVE_CLEANUP_VECTOR); + goto unlock; + } __get_cpu_var(vector_irq)[vector] = -1; cfg->move_cleanup_count--; unlock: @@ -2529,9 +2495,44 @@ static inline void irq_complete_move(struct irq_desc **descp) {} #endif #ifdef CONFIG_INTR_REMAP +static void __eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) +{ + int apic, pin; + struct irq_pin_list *entry; + + entry = cfg->irq_2_pin; + for (;;) { + + if (!entry) + break; + + apic = entry->apic; + pin = entry->pin; + io_apic_eoi(apic, pin); + entry = entry->next; + } +} + +static void +eoi_ioapic_irq(struct irq_desc *desc) +{ + struct irq_cfg *cfg; + unsigned long flags; + unsigned int irq; + + irq = desc->irq; + cfg = desc->chip_data; + + spin_lock_irqsave(&ioapic_lock, flags); + __eoi_ioapic_irq(irq, cfg); + spin_unlock_irqrestore(&ioapic_lock, flags); +} + static void ack_x2apic_level(unsigned int irq) { + struct irq_desc *desc = irq_to_desc(irq); ack_x2APIC_irq(); + eoi_ioapic_irq(desc); } static void ack_x2apic_edge(unsigned int irq) @@ -2901,10 +2902,8 @@ static inline void __init check_timer(void) * 8259A. */ if (pin1 == -1) { -#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) panic("BIOS bug: timer not connected to IO-APIC"); -#endif pin1 = pin2; apic1 = apic2; no_pin1 = 1; @@ -2940,10 +2939,8 @@ static inline void __init check_timer(void) clear_IO_APIC_pin(0, pin1); goto out; } -#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) panic("timer doesn't work through Interrupt-remapped IO-APIC"); -#endif local_irq_disable(); clear_IO_APIC_pin(apic1, pin1); if (!no_pin1) @@ -3237,9 +3234,7 @@ void destroy_irq(unsigned int irq) if (desc) desc->chip_data = cfg; -#ifdef CONFIG_INTR_REMAP free_irte(irq); -#endif spin_lock_irqsave(&vector_lock, flags); __clear_irq_vector(irq, cfg); spin_unlock_irqrestore(&vector_lock, flags); @@ -3265,7 +3260,6 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus()); -#ifdef CONFIG_INTR_REMAP if (irq_remapped(irq)) { struct irte irte; int ir_index; @@ -3291,10 +3285,13 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms MSI_ADDR_IR_SHV | MSI_ADDR_IR_INDEX1(ir_index) | MSI_ADDR_IR_INDEX2(ir_index); - } else -#endif - { - msg->address_hi = MSI_ADDR_BASE_HI; + } else { + if (x2apic_enabled()) + msg->address_hi = MSI_ADDR_BASE_HI | + MSI_ADDR_EXT_DEST_ID(dest); + else + msg->address_hi = MSI_ADDR_BASE_HI; + msg->address_lo = MSI_ADDR_BASE_LO | ((apic->irq_dest_mode == 0) ? @@ -3405,6 +3402,7 @@ static struct irq_chip msi_ir_chip = { #endif .retrigger = ioapic_retrigger_irq, }; +#endif /* * Map the PCI dev to the corresponding remapping hardware unit @@ -3432,7 +3430,6 @@ static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec) } return index; } -#endif static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) { @@ -3446,7 +3443,6 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) set_irq_msi(irq, msidesc); write_msi_msg(irq, &msg); -#ifdef CONFIG_INTR_REMAP if (irq_remapped(irq)) { struct irq_desc *desc = irq_to_desc(irq); /* @@ -3455,7 +3451,6 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) desc->status |= IRQ_MOVE_PCNTXT; set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge"); } else -#endif set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge"); dev_printk(KERN_DEBUG, &dev->dev, "irq %d for MSI/MSI-X\n", irq); @@ -3469,11 +3464,8 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) int ret, sub_handle; struct msi_desc *msidesc; unsigned int irq_want; - -#ifdef CONFIG_INTR_REMAP struct intel_iommu *iommu = 0; int index = 0; -#endif irq_want = nr_irqs_gsi; sub_handle = 0; @@ -3482,7 +3474,6 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) if (irq == 0) return -1; irq_want = irq + 1; -#ifdef CONFIG_INTR_REMAP if (!intr_remapping_enabled) goto no_ir; @@ -3510,7 +3501,6 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) set_irte_irq(irq, iommu, index, sub_handle); } no_ir: -#endif ret = setup_msi_irq(dev, msidesc, irq); if (ret < 0) goto error; @@ -3528,7 +3518,7 @@ void arch_teardown_msi_irq(unsigned int irq) destroy_irq(irq); } -#ifdef CONFIG_DMAR +#if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP) #ifdef CONFIG_SMP static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) { @@ -4045,11 +4035,9 @@ void __init setup_ioapic_dest(void) else mask = apic->target_cpus(); -#ifdef CONFIG_INTR_REMAP if (intr_remapping_enabled) set_ir_ioapic_affinity_irq_desc(desc, mask); else -#endif set_ioapic_affinity_irq_desc(desc, mask); } diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c index 8d7748efe6a..1783652bb0e 100644 --- a/arch/x86/kernel/apic/probe_64.c +++ b/arch/x86/kernel/apic/probe_64.c @@ -68,6 +68,13 @@ void __init default_setup_apic_routing(void) apic = &apic_physflat; printk(KERN_INFO "Setting APIC routing to %s\n", apic->name); } + + /* + * Now that apic routing model is selected, configure the + * fault handling for intr remapping. + */ + if (intr_remapping_enabled) + enable_drhd_fault_handling(); } /* Same for both flat and physical. */ diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c index 8fb87b6dd63..4a903e2f0d1 100644 --- a/arch/x86/kernel/apic/x2apic_cluster.c +++ b/arch/x86/kernel/apic/x2apic_cluster.c @@ -57,6 +57,8 @@ static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) unsigned long query_cpu; unsigned long flags; + x2apic_wrmsr_fence(); + local_irq_save(flags); for_each_cpu(query_cpu, mask) { __x2apic_send_IPI_dest( @@ -73,6 +75,8 @@ static void unsigned long query_cpu; unsigned long flags; + x2apic_wrmsr_fence(); + local_irq_save(flags); for_each_cpu(query_cpu, mask) { if (query_cpu == this_cpu) @@ -90,6 +94,8 @@ static void x2apic_send_IPI_allbutself(int vector) unsigned long query_cpu; unsigned long flags; + x2apic_wrmsr_fence(); + local_irq_save(flags); for_each_online_cpu(query_cpu) { if (query_cpu == this_cpu) diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c index 23625b9f98b..a284359627e 100644 --- a/arch/x86/kernel/apic/x2apic_phys.c +++ b/arch/x86/kernel/apic/x2apic_phys.c @@ -58,6 +58,8 @@ static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) unsigned long query_cpu; unsigned long flags; + x2apic_wrmsr_fence(); + local_irq_save(flags); for_each_cpu(query_cpu, mask) { __x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu), @@ -73,6 +75,8 @@ static void unsigned long query_cpu; unsigned long flags; + x2apic_wrmsr_fence(); + local_irq_save(flags); for_each_cpu(query_cpu, mask) { if (query_cpu != this_cpu) @@ -89,6 +93,8 @@ static void x2apic_send_IPI_allbutself(int vector) unsigned long query_cpu; unsigned long flags; + x2apic_wrmsr_fence(); + local_irq_save(flags); for_each_online_cpu(query_cpu) { if (query_cpu == this_cpu) diff --git a/arch/x86/kernel/check.c b/arch/x86/kernel/check.c index b617b1164f1..fc999e6fc46 100644 --- a/arch/x86/kernel/check.c +++ b/arch/x86/kernel/check.c @@ -86,12 +86,12 @@ void __init setup_bios_corruption_check(void) if (!(addr + 1)) break; + if (addr >= corruption_check_size) + break; + if ((addr + size) > corruption_check_size) size = corruption_check_size - addr; - if (size == 0) - break; - e820_update_range(addr, size, E820_RAM, E820_RESERVED); scan_areas[num_scan_areas].addr = addr; scan_areas[num_scan_areas].size = size; diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index d4356f8b752..4e242f9a06e 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -19,8 +19,7 @@ obj-$(CONFIG_X86_CPU_DEBUG) += cpu_debug.o obj-$(CONFIG_CPU_SUP_INTEL) += intel.o obj-$(CONFIG_CPU_SUP_AMD) += amd.o obj-$(CONFIG_CPU_SUP_CYRIX_32) += cyrix.o -obj-$(CONFIG_CPU_SUP_CENTAUR_32) += centaur.o -obj-$(CONFIG_CPU_SUP_CENTAUR_64) += centaur_64.o +obj-$(CONFIG_CPU_SUP_CENTAUR) += centaur.o obj-$(CONFIG_CPU_SUP_TRANSMETA_32) += transmeta.o obj-$(CONFIG_CPU_SUP_UMC_32) += umc.o diff --git a/arch/x86/kernel/cpu/centaur.c b/arch/x86/kernel/cpu/centaur.c index 983e0830f0d..c95e831bb09 100644 --- a/arch/x86/kernel/cpu/centaur.c +++ b/arch/x86/kernel/cpu/centaur.c @@ -1,11 +1,11 @@ +#include <linux/bitops.h> #include <linux/kernel.h> #include <linux/init.h> -#include <linux/bitops.h> #include <asm/processor.h> -#include <asm/msr.h> #include <asm/e820.h> #include <asm/mtrr.h> +#include <asm/msr.h> #include "cpu.h" @@ -276,7 +276,7 @@ static void __cpuinit init_c3(struct cpuinfo_x86 *c) */ c->x86_capability[5] = cpuid_edx(0xC0000001); } - +#ifdef CONFIG_X86_32 /* Cyrix III family needs CX8 & PGE explicitly enabled. */ if (c->x86_model >= 6 && c->x86_model <= 9) { rdmsr(MSR_VIA_FCR, lo, hi); @@ -288,6 +288,11 @@ static void __cpuinit init_c3(struct cpuinfo_x86 *c) /* Before Nehemiah, the C3's had 3dNOW! */ if (c->x86_model >= 6 && c->x86_model < 9) set_cpu_cap(c, X86_FEATURE_3DNOW); +#endif + if (c->x86 == 0x6 && c->x86_model >= 0xf) { + c->x86_cache_alignment = c->x86_clflush_size * 2; + set_cpu_cap(c, X86_FEATURE_REP_GOOD); + } display_cacheinfo(c); } @@ -316,16 +321,25 @@ enum { static void __cpuinit early_init_centaur(struct cpuinfo_x86 *c) { switch (c->x86) { +#ifdef CONFIG_X86_32 case 5: /* Emulate MTRRs using Centaur's MCR. */ set_cpu_cap(c, X86_FEATURE_CENTAUR_MCR); break; +#endif + case 6: + if (c->x86_model >= 0xf) + set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); + break; } +#ifdef CONFIG_X86_64 + set_cpu_cap(c, X86_FEATURE_SYSENTER32); +#endif } static void __cpuinit init_centaur(struct cpuinfo_x86 *c) { - +#ifdef CONFIG_X86_32 char *name; u32 fcr_set = 0; u32 fcr_clr = 0; @@ -337,8 +351,10 @@ static void __cpuinit init_centaur(struct cpuinfo_x86 *c) * 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */ clear_cpu_cap(c, 0*32+31); - +#endif + early_init_centaur(c); switch (c->x86) { +#ifdef CONFIG_X86_32 case 5: switch (c->x86_model) { case 4: @@ -442,16 +458,20 @@ static void __cpuinit init_centaur(struct cpuinfo_x86 *c) } sprintf(c->x86_model_id, "WinChip %s", name); break; - +#endif case 6: init_c3(c); break; } +#ifdef CONFIG_X86_64 + set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC); +#endif } static unsigned int __cpuinit centaur_size_cache(struct cpuinfo_x86 *c, unsigned int size) { +#ifdef CONFIG_X86_32 /* VIA C3 CPUs (670-68F) need further shifting. */ if ((c->x86 == 6) && ((c->x86_model == 7) || (c->x86_model == 8))) size >>= 8; @@ -464,7 +484,7 @@ centaur_size_cache(struct cpuinfo_x86 *c, unsigned int size) if ((c->x86 == 6) && (c->x86_model == 9) && (c->x86_mask == 1) && (size == 65)) size -= 1; - +#endif return size; } diff --git a/arch/x86/kernel/cpu/centaur_64.c b/arch/x86/kernel/cpu/centaur_64.c deleted file mode 100644 index 51b09c48c9c..00000000000 --- a/arch/x86/kernel/cpu/centaur_64.c +++ /dev/null @@ -1,37 +0,0 @@ -#include <linux/init.h> -#include <linux/smp.h> - -#include <asm/cpufeature.h> -#include <asm/processor.h> - -#include "cpu.h" - -static void __cpuinit early_init_centaur(struct cpuinfo_x86 *c) -{ - if (c->x86 == 0x6 && c->x86_model >= 0xf) - set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); - - set_cpu_cap(c, X86_FEATURE_SYSENTER32); -} - -static void __cpuinit init_centaur(struct cpuinfo_x86 *c) -{ - early_init_centaur(c); - - if (c->x86 == 0x6 && c->x86_model >= 0xf) { - c->x86_cache_alignment = c->x86_clflush_size * 2; - set_cpu_cap(c, X86_FEATURE_REP_GOOD); - } - set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC); -} - -static const struct cpu_dev centaur_cpu_dev __cpuinitconst = { - .c_vendor = "Centaur", - .c_ident = { "CentaurHauls" }, - .c_early_init = early_init_centaur, - .c_init = init_centaur, - .c_x86_vendor = X86_VENDOR_CENTAUR, -}; - -cpu_dev_register(centaur_cpu_dev); - diff --git a/arch/x86/kernel/cpu/cpu_debug.c b/arch/x86/kernel/cpu/cpu_debug.c index 21c0cf8ced1..46e29ab96c6 100755 --- a/arch/x86/kernel/cpu/cpu_debug.c +++ b/arch/x86/kernel/cpu/cpu_debug.c @@ -64,6 +64,8 @@ static struct cpu_debug_base cpu_base[] = { { "vmx", CPU_VMX, 0 }, { "call", CPU_CALL, 0 }, { "base", CPU_BASE, 0 }, + { "ver", CPU_VER, 0 }, + { "conf", CPU_CONF, 0 }, { "smm", CPU_SMM, 0 }, { "svm", CPU_SVM, 0 }, { "osvm", CPU_OSVM, 0 }, @@ -177,54 +179,59 @@ static struct cpu_debug_range cpu_intel_range[] = { /* AMD Registers Range */ static struct cpu_debug_range cpu_amd_range[] = { - { 0x00000010, 0x00000010, CPU_TIME, CPU_ALL, }, - { 0x0000001B, 0x0000001B, CPU_APIC, CPU_ALL, }, - { 0x000000FE, 0x000000FE, CPU_MTRR, CPU_ALL, }, - - { 0x00000174, 0x00000176, CPU_SYSENTER, CPU_ALL, }, - { 0x00000179, 0x0000017A, CPU_MC, CPU_ALL, }, - { 0x0000017B, 0x0000017B, CPU_MC, CPU_ALL, }, - { 0x000001D9, 0x000001D9, CPU_DEBUG, CPU_ALL, }, - { 0x000001DB, 0x000001DE, CPU_LBRANCH, CPU_ALL, }, - - { 0x00000200, 0x0000020F, CPU_MTRR, CPU_ALL, }, - { 0x00000250, 0x00000250, CPU_MTRR, CPU_ALL, }, - { 0x00000258, 0x00000259, CPU_MTRR, CPU_ALL, }, - { 0x00000268, 0x0000026F, CPU_MTRR, CPU_ALL, }, - { 0x00000277, 0x00000277, CPU_PAT, CPU_ALL, }, - { 0x000002FF, 0x000002FF, CPU_MTRR, CPU_ALL, }, - - { 0x00000400, 0x00000417, CPU_MC, CPU_ALL, }, - - { 0xC0000080, 0xC0000080, CPU_FEATURES, CPU_ALL, }, - { 0xC0000081, 0xC0000084, CPU_CALL, CPU_ALL, }, - { 0xC0000100, 0xC0000102, CPU_BASE, CPU_ALL, }, - { 0xC0000103, 0xC0000103, CPU_TIME, CPU_ALL, }, - - { 0xC0000408, 0xC000040A, CPU_MC, CPU_ALL, }, - - { 0xc0010000, 0xc0010007, CPU_PMC, CPU_ALL, }, - { 0xc0010010, 0xc0010010, CPU_MTRR, CPU_ALL, }, - { 0xc0010016, 0xc001001A, CPU_MTRR, CPU_ALL, }, - { 0xc001001D, 0xc001001D, CPU_MTRR, CPU_ALL, }, - { 0xc0010030, 0xc0010035, CPU_BIOS, CPU_ALL, }, - { 0xc0010056, 0xc0010056, CPU_SMM, CPU_ALL, }, - { 0xc0010061, 0xc0010063, CPU_SMM, CPU_ALL, }, - { 0xc0010074, 0xc0010074, CPU_MC, CPU_ALL, }, - { 0xc0010111, 0xc0010113, CPU_SMM, CPU_ALL, }, - { 0xc0010114, 0xc0010118, CPU_SVM, CPU_ALL, }, - { 0xc0010119, 0xc001011A, CPU_SMM, CPU_ALL, }, - { 0xc0010140, 0xc0010141, CPU_OSVM, CPU_ALL, }, - { 0xc0010156, 0xc0010156, CPU_SMM, CPU_ALL, }, + { 0x00000000, 0x00000001, CPU_MC, CPU_K10_PLUS, }, + { 0x00000010, 0x00000010, CPU_TIME, CPU_K8_PLUS, }, + { 0x0000001B, 0x0000001B, CPU_APIC, CPU_K8_PLUS, }, + { 0x0000002A, 0x0000002A, CPU_POWERON, CPU_K7_PLUS }, + { 0x0000008B, 0x0000008B, CPU_VER, CPU_K8_PLUS }, + { 0x000000FE, 0x000000FE, CPU_MTRR, CPU_K8_PLUS, }, + + { 0x00000174, 0x00000176, CPU_SYSENTER, CPU_K8_PLUS, }, + { 0x00000179, 0x0000017B, CPU_MC, CPU_K8_PLUS, }, + { 0x000001D9, 0x000001D9, CPU_DEBUG, CPU_K8_PLUS, }, + { 0x000001DB, 0x000001DE, CPU_LBRANCH, CPU_K8_PLUS, }, + + { 0x00000200, 0x0000020F, CPU_MTRR, CPU_K8_PLUS, }, + { 0x00000250, 0x00000250, CPU_MTRR, CPU_K8_PLUS, }, + { 0x00000258, 0x00000259, CPU_MTRR, CPU_K8_PLUS, }, + { 0x00000268, 0x0000026F, CPU_MTRR, CPU_K8_PLUS, }, + { 0x00000277, 0x00000277, CPU_PAT, CPU_K8_PLUS, }, + { 0x000002FF, 0x000002FF, CPU_MTRR, CPU_K8_PLUS, }, + + { 0x00000400, 0x00000413, CPU_MC, CPU_K8_PLUS, }, + + { 0xC0000080, 0xC0000080, CPU_FEATURES, CPU_AMD_ALL, }, + { 0xC0000081, 0xC0000084, CPU_CALL, CPU_K8_PLUS, }, + { 0xC0000100, 0xC0000102, CPU_BASE, CPU_K8_PLUS, }, + { 0xC0000103, 0xC0000103, CPU_TIME, CPU_K10_PLUS, }, + + { 0xC0010000, 0xC0010007, CPU_PMC, CPU_K8_PLUS, }, + { 0xC0010010, 0xC0010010, CPU_CONF, CPU_K7_PLUS, }, + { 0xC0010015, 0xC0010015, CPU_CONF, CPU_K7_PLUS, }, + { 0xC0010016, 0xC001001A, CPU_MTRR, CPU_K8_PLUS, }, + { 0xC001001D, 0xC001001D, CPU_MTRR, CPU_K8_PLUS, }, + { 0xC001001F, 0xC001001F, CPU_CONF, CPU_K8_PLUS, }, + { 0xC0010030, 0xC0010035, CPU_BIOS, CPU_K8_PLUS, }, + { 0xC0010044, 0xC0010048, CPU_MC, CPU_K8_PLUS, }, + { 0xC0010050, 0xC0010056, CPU_SMM, CPU_K0F_PLUS, }, + { 0xC0010058, 0xC0010058, CPU_CONF, CPU_K10_PLUS, }, + { 0xC0010060, 0xC0010060, CPU_CACHE, CPU_AMD_11, }, + { 0xC0010061, 0xC0010068, CPU_SMM, CPU_K10_PLUS, }, + { 0xC0010069, 0xC001006B, CPU_SMM, CPU_AMD_11, }, + { 0xC0010070, 0xC0010071, CPU_SMM, CPU_K10_PLUS, }, + { 0xC0010111, 0xC0010113, CPU_SMM, CPU_K8_PLUS, }, + { 0xC0010114, 0xC0010118, CPU_SVM, CPU_K10_PLUS, }, + { 0xC0010140, 0xC0010141, CPU_OSVM, CPU_K10_PLUS, }, + { 0xC0011022, 0xC0011023, CPU_CONF, CPU_K10_PLUS, }, }; -static int get_cpu_modelflag(unsigned cpu) +/* Intel */ +static int get_intel_modelflag(unsigned model) { int flag; - switch (per_cpu(cpu_model, cpu)) { - /* Intel */ + switch (model) { case 0x0501: case 0x0502: case 0x0504: @@ -271,6 +278,59 @@ static int get_cpu_modelflag(unsigned cpu) return flag; } +/* AMD */ +static int get_amd_modelflag(unsigned model) +{ + int flag; + + switch (model >> 8) { + case 0x6: + flag = CPU_AMD_K6; + break; + case 0x7: + flag = CPU_AMD_K7; + break; + case 0x8: + flag = CPU_AMD_K8; + break; + case 0xf: + flag = CPU_AMD_0F; + break; + case 0x10: + flag = CPU_AMD_10; + break; + case 0x11: + flag = CPU_AMD_11; + break; + default: + flag = CPU_NONE; + break; + } + + return flag; +} + +static int get_cpu_modelflag(unsigned cpu) +{ + int flag; + + flag = per_cpu(cpu_model, cpu); + + switch (flag >> 16) { + case X86_VENDOR_INTEL: + flag = get_intel_modelflag(flag); + break; + case X86_VENDOR_AMD: + flag = get_amd_modelflag(flag & 0xffff); + break; + default: + flag = CPU_NONE; + break; + } + + return flag; +} + static int get_cpu_range_count(unsigned cpu) { int index; @@ -311,7 +371,8 @@ static int is_typeflag_valid(unsigned cpu, unsigned flag) return 1; break; case X86_VENDOR_AMD: - if (cpu_amd_range[i].flag & flag) + if ((cpu_amd_range[i].model & modelflag) && + (cpu_amd_range[i].flag & flag)) return 1; break; } @@ -337,7 +398,8 @@ static unsigned get_cpu_range(unsigned cpu, unsigned *min, unsigned *max, } break; case X86_VENDOR_AMD: - if (cpu_amd_range[index].flag & flag) { + if ((cpu_amd_range[index].model & modelflag) && + (cpu_amd_range[index].flag & flag)) { *min = cpu_amd_range[index].min; *max = cpu_amd_range[index].max; } diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c index c5a32f92d07..7d01be86887 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c @@ -92,7 +92,8 @@ struct thresh_restart { }; /* must be called with correct cpu affinity */ -static long threshold_restart_bank(void *_tr) +/* Called via smp_call_function_single() */ +static void threshold_restart_bank(void *_tr) { struct thresh_restart *tr = _tr; u32 mci_misc_hi, mci_misc_lo; @@ -119,7 +120,6 @@ static long threshold_restart_bank(void *_tr) mci_misc_hi |= MASK_COUNT_EN_HI; wrmsr(tr->b->address, mci_misc_lo, mci_misc_hi); - return 0; } /* cpu init entry point, called from mce.c with preempt off */ @@ -279,7 +279,7 @@ static ssize_t store_interrupt_enable(struct threshold_block *b, tr.b = b; tr.reset = 0; tr.old_limit = 0; - work_on_cpu(b->cpu, threshold_restart_bank, &tr); + smp_call_function_single(b->cpu, threshold_restart_bank, &tr, 1); return end - buf; } @@ -301,23 +301,32 @@ static ssize_t store_threshold_limit(struct threshold_block *b, tr.b = b; tr.reset = 0; - work_on_cpu(b->cpu, threshold_restart_bank, &tr); + smp_call_function_single(b->cpu, threshold_restart_bank, &tr, 1); return end - buf; } -static long local_error_count(void *_b) +struct threshold_block_cross_cpu { + struct threshold_block *tb; + long retval; +}; + +static void local_error_count_handler(void *_tbcc) { - struct threshold_block *b = _b; + struct threshold_block_cross_cpu *tbcc = _tbcc; + struct threshold_block *b = tbcc->tb; u32 low, high; rdmsr(b->address, low, high); - return (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit); + tbcc->retval = (high & 0xFFF) - (THRESHOLD_MAX - b->threshold_limit); } static ssize_t show_error_count(struct threshold_block *b, char *buf) { - return sprintf(buf, "%lx\n", work_on_cpu(b->cpu, local_error_count, b)); + struct threshold_block_cross_cpu tbcc = { .tb = b, }; + + smp_call_function_single(b->cpu, local_error_count_handler, &tbcc, 1); + return sprintf(buf, "%lx\n", tbcc.retval); } static ssize_t store_error_count(struct threshold_block *b, @@ -325,7 +334,7 @@ static ssize_t store_error_count(struct threshold_block *b, { struct thresh_restart tr = { .b = b, .reset = 1, .old_limit = 0 }; - work_on_cpu(b->cpu, threshold_restart_bank, &tr); + smp_call_function_single(b->cpu, threshold_restart_bank, &tr, 1); return 1; } @@ -394,7 +403,7 @@ static __cpuinit int allocate_threshold_blocks(unsigned int cpu, if ((bank >= NR_BANKS) || (block >= NR_BLOCKS)) return 0; - if (rdmsr_safe(address, &low, &high)) + if (rdmsr_safe_on_cpu(cpu, address, &low, &high)) return 0; if (!(high & MASK_VALID_HI)) { @@ -458,12 +467,11 @@ out_free: return err; } -static __cpuinit long local_allocate_threshold_blocks(void *_bank) +static __cpuinit long +local_allocate_threshold_blocks(int cpu, unsigned int bank) { - unsigned int *bank = _bank; - - return allocate_threshold_blocks(smp_processor_id(), *bank, 0, - MSR_IA32_MC0_MISC + *bank * 4); + return allocate_threshold_blocks(cpu, bank, 0, + MSR_IA32_MC0_MISC + bank * 4); } /* symlinks sibling shared banks to first core. first core owns dir/files. */ @@ -526,7 +534,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) per_cpu(threshold_banks, cpu)[bank] = b; - err = work_on_cpu(cpu, local_allocate_threshold_blocks, &bank); + err = local_allocate_threshold_blocks(cpu, bank); if (err) goto out_free; diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c index aaa7d973093..57df3d38347 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c @@ -270,7 +270,7 @@ void cmci_reenable(void) cmci_discover(banks, 0); } -static __cpuinit void intel_init_cmci(void) +static void intel_init_cmci(void) { int banks; diff --git a/arch/x86/kernel/cpu/mtrr/Makefile b/arch/x86/kernel/cpu/mtrr/Makefile index 191fc053364..f4361b56f8e 100644 --- a/arch/x86/kernel/cpu/mtrr/Makefile +++ b/arch/x86/kernel/cpu/mtrr/Makefile @@ -1,3 +1,3 @@ -obj-y := main.o if.o generic.o state.o +obj-y := main.o if.o generic.o state.o cleanup.o obj-$(CONFIG_X86_32) += amd.o cyrix.o centaur.o diff --git a/arch/x86/kernel/cpu/mtrr/cleanup.c b/arch/x86/kernel/cpu/mtrr/cleanup.c new file mode 100644 index 00000000000..ce0fe4b5c04 --- /dev/null +++ b/arch/x86/kernel/cpu/mtrr/cleanup.c @@ -0,0 +1,1101 @@ +/* MTRR (Memory Type Range Register) cleanup + + Copyright (C) 2009 Yinghai Lu + + This library is free software; you can redistribute it and/or + modify it under the terms of the GNU Library General Public + License as published by the Free Software Foundation; either + version 2 of the License, or (at your option) any later version. + + This library 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 + Library General Public License for more details. + + You should have received a copy of the GNU Library General Public + License along with this library; if not, write to the Free + Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. +*/ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/pci.h> +#include <linux/smp.h> +#include <linux/cpu.h> +#include <linux/mutex.h> +#include <linux/sort.h> + +#include <asm/e820.h> +#include <asm/mtrr.h> +#include <asm/uaccess.h> +#include <asm/processor.h> +#include <asm/msr.h> +#include <asm/kvm_para.h> +#include "mtrr.h" + +/* should be related to MTRR_VAR_RANGES nums */ +#define RANGE_NUM 256 + +struct res_range { + unsigned long start; + unsigned long end; +}; + +static int __init +add_range(struct res_range *range, int nr_range, unsigned long start, + unsigned long end) +{ + /* out of slots */ + if (nr_range >= RANGE_NUM) + return nr_range; + + range[nr_range].start = start; + range[nr_range].end = end; + + nr_range++; + + return nr_range; +} + +static int __init +add_range_with_merge(struct res_range *range, int nr_range, unsigned long start, + unsigned long end) +{ + int i; + + /* try to merge it with old one */ + for (i = 0; i < nr_range; i++) { + unsigned long final_start, final_end; + unsigned long common_start, common_end; + + if (!range[i].end) + continue; + + common_start = max(range[i].start, start); + common_end = min(range[i].end, end); + if (common_start > common_end + 1) + continue; + + final_start = min(range[i].start, start); + final_end = max(range[i].end, end); + + range[i].start = final_start; + range[i].end = final_end; + return nr_range; + } + + /* need to add that */ + return add_range(range, nr_range, start, end); +} + +static void __init +subtract_range(struct res_range *range, unsigned long start, unsigned long end) +{ + int i, j; + + for (j = 0; j < RANGE_NUM; j++) { + if (!range[j].end) + continue; + + if (start <= range[j].start && end >= range[j].end) { + range[j].start = 0; + range[j].end = 0; + continue; + } + + if (start <= range[j].start && end < range[j].end && + range[j].start < end + 1) { + range[j].start = end + 1; + continue; + } + + + if (start > range[j].start && end >= range[j].end && + range[j].end > start - 1) { + range[j].end = start - 1; + continue; + } + + if (start > range[j].start && end < range[j].end) { + /* find the new spare */ + for (i = 0; i < RANGE_NUM; i++) { + if (range[i].end == 0) + break; + } + if (i < RANGE_NUM) { + range[i].end = range[j].end; + range[i].start = end + 1; + } else { + printk(KERN_ERR "run of slot in ranges\n"); + } + range[j].end = start - 1; + continue; + } + } +} + +static int __init cmp_range(const void *x1, const void *x2) +{ + const struct res_range *r1 = x1; + const struct res_range *r2 = x2; + long start1, start2; + + start1 = r1->start; + start2 = r2->start; + + return start1 - start2; +} + +struct var_mtrr_range_state { + unsigned long base_pfn; + unsigned long size_pfn; + mtrr_type type; +}; + +static struct var_mtrr_range_state __initdata range_state[RANGE_NUM]; +static int __initdata debug_print; + +static int __init +x86_get_mtrr_mem_range(struct res_range *range, int nr_range, + unsigned long extra_remove_base, + unsigned long extra_remove_size) +{ + unsigned long base, size; + mtrr_type type; + int i; + + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + if (type != MTRR_TYPE_WRBACK) + continue; + base = range_state[i].base_pfn; + size = range_state[i].size_pfn; + nr_range = add_range_with_merge(range, nr_range, base, + base + size - 1); + } + if (debug_print) { + printk(KERN_DEBUG "After WB checking\n"); + for (i = 0; i < nr_range; i++) + printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", + range[i].start, range[i].end + 1); + } + + /* take out UC ranges */ + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + if (type != MTRR_TYPE_UNCACHABLE && + type != MTRR_TYPE_WRPROT) + continue; + size = range_state[i].size_pfn; + if (!size) + continue; + base = range_state[i].base_pfn; + if (base < (1<<(20-PAGE_SHIFT)) && mtrr_state.have_fixed && + (mtrr_state.enabled & 1)) { + /* Var MTRR contains UC entry below 1M? Skip it: */ + printk(KERN_WARNING "WARNING: BIOS bug: VAR MTRR %d " + "contains strange UC entry under 1M, check " + "with your system vendor!\n", i); + if (base + size <= (1<<(20-PAGE_SHIFT))) + continue; + size -= (1<<(20-PAGE_SHIFT)) - base; + base = 1<<(20-PAGE_SHIFT); + } + subtract_range(range, base, base + size - 1); + } + if (extra_remove_size) + subtract_range(range, extra_remove_base, + extra_remove_base + extra_remove_size - 1); + + /* get new range num */ + nr_range = 0; + for (i = 0; i < RANGE_NUM; i++) { + if (!range[i].end) + continue; + nr_range++; + } + if (debug_print) { + printk(KERN_DEBUG "After UC checking\n"); + for (i = 0; i < nr_range; i++) + printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", + range[i].start, range[i].end + 1); + } + + /* sort the ranges */ + sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); + if (debug_print) { + printk(KERN_DEBUG "After sorting\n"); + for (i = 0; i < nr_range; i++) + printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", + range[i].start, range[i].end + 1); + } + + /* clear those is not used */ + for (i = nr_range; i < RANGE_NUM; i++) + memset(&range[i], 0, sizeof(range[i])); + + return nr_range; +} + +static struct res_range __initdata range[RANGE_NUM]; +static int __initdata nr_range; + +#ifdef CONFIG_MTRR_SANITIZER + +static unsigned long __init sum_ranges(struct res_range *range, int nr_range) +{ + unsigned long sum; + int i; + + sum = 0; + for (i = 0; i < nr_range; i++) + sum += range[i].end + 1 - range[i].start; + + return sum; +} + +static int enable_mtrr_cleanup __initdata = + CONFIG_MTRR_SANITIZER_ENABLE_DEFAULT; + +static int __init disable_mtrr_cleanup_setup(char *str) +{ + enable_mtrr_cleanup = 0; + return 0; +} +early_param("disable_mtrr_cleanup", disable_mtrr_cleanup_setup); + +static int __init enable_mtrr_cleanup_setup(char *str) +{ + enable_mtrr_cleanup = 1; + return 0; +} +early_param("enable_mtrr_cleanup", enable_mtrr_cleanup_setup); + +static int __init mtrr_cleanup_debug_setup(char *str) +{ + debug_print = 1; + return 0; +} +early_param("mtrr_cleanup_debug", mtrr_cleanup_debug_setup); + +struct var_mtrr_state { + unsigned long range_startk; + unsigned long range_sizek; + unsigned long chunk_sizek; + unsigned long gran_sizek; + unsigned int reg; +}; + +static void __init +set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, + unsigned char type, unsigned int address_bits) +{ + u32 base_lo, base_hi, mask_lo, mask_hi; + u64 base, mask; + + if (!sizek) { + fill_mtrr_var_range(reg, 0, 0, 0, 0); + return; + } + + mask = (1ULL << address_bits) - 1; + mask &= ~((((u64)sizek) << 10) - 1); + + base = ((u64)basek) << 10; + + base |= type; + mask |= 0x800; + + base_lo = base & ((1ULL<<32) - 1); + base_hi = base >> 32; + + mask_lo = mask & ((1ULL<<32) - 1); + mask_hi = mask >> 32; + + fill_mtrr_var_range(reg, base_lo, base_hi, mask_lo, mask_hi); +} + +static void __init +save_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, + unsigned char type) +{ + range_state[reg].base_pfn = basek >> (PAGE_SHIFT - 10); + range_state[reg].size_pfn = sizek >> (PAGE_SHIFT - 10); + range_state[reg].type = type; +} + +static void __init +set_var_mtrr_all(unsigned int address_bits) +{ + unsigned long basek, sizek; + unsigned char type; + unsigned int reg; + + for (reg = 0; reg < num_var_ranges; reg++) { + basek = range_state[reg].base_pfn << (PAGE_SHIFT - 10); + sizek = range_state[reg].size_pfn << (PAGE_SHIFT - 10); + type = range_state[reg].type; + + set_var_mtrr(reg, basek, sizek, type, address_bits); + } +} + +static unsigned long to_size_factor(unsigned long sizek, char *factorp) +{ + char factor; + unsigned long base = sizek; + + if (base & ((1<<10) - 1)) { + /* not MB alignment */ + factor = 'K'; + } else if (base & ((1<<20) - 1)) { + factor = 'M'; + base >>= 10; + } else { + factor = 'G'; + base >>= 20; + } + + *factorp = factor; + + return base; +} + +static unsigned int __init +range_to_mtrr(unsigned int reg, unsigned long range_startk, + unsigned long range_sizek, unsigned char type) +{ + if (!range_sizek || (reg >= num_var_ranges)) + return reg; + + while (range_sizek) { + unsigned long max_align, align; + unsigned long sizek; + + /* Compute the maximum size I can make a range */ + if (range_startk) + max_align = ffs(range_startk) - 1; + else + max_align = 32; + align = fls(range_sizek) - 1; + if (align > max_align) + align = max_align; + + sizek = 1 << align; + if (debug_print) { + char start_factor = 'K', size_factor = 'K'; + unsigned long start_base, size_base; + + start_base = to_size_factor(range_startk, + &start_factor), + size_base = to_size_factor(sizek, &size_factor), + + printk(KERN_DEBUG "Setting variable MTRR %d, " + "base: %ld%cB, range: %ld%cB, type %s\n", + reg, start_base, start_factor, + size_base, size_factor, + (type == MTRR_TYPE_UNCACHABLE) ? "UC" : + ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other") + ); + } + save_var_mtrr(reg++, range_startk, sizek, type); + range_startk += sizek; + range_sizek -= sizek; + if (reg >= num_var_ranges) + break; + } + return reg; +} + +static unsigned __init +range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek, + unsigned long sizek) +{ + unsigned long hole_basek, hole_sizek; + unsigned long second_basek, second_sizek; + unsigned long range0_basek, range0_sizek; + unsigned long range_basek, range_sizek; + unsigned long chunk_sizek; + unsigned long gran_sizek; + + hole_basek = 0; + hole_sizek = 0; + second_basek = 0; + second_sizek = 0; + chunk_sizek = state->chunk_sizek; + gran_sizek = state->gran_sizek; + + /* align with gran size, prevent small block used up MTRRs */ + range_basek = ALIGN(state->range_startk, gran_sizek); + if ((range_basek > basek) && basek) + return second_sizek; + state->range_sizek -= (range_basek - state->range_startk); + range_sizek = ALIGN(state->range_sizek, gran_sizek); + + while (range_sizek > state->range_sizek) { + range_sizek -= gran_sizek; + if (!range_sizek) + return 0; + } + state->range_sizek = range_sizek; + + /* try to append some small hole */ + range0_basek = state->range_startk; + range0_sizek = ALIGN(state->range_sizek, chunk_sizek); + + /* no increase */ + if (range0_sizek == state->range_sizek) { + if (debug_print) + printk(KERN_DEBUG "rangeX: %016lx - %016lx\n", + range0_basek<<10, + (range0_basek + state->range_sizek)<<10); + state->reg = range_to_mtrr(state->reg, range0_basek, + state->range_sizek, MTRR_TYPE_WRBACK); + return 0; + } + + /* only cut back, when it is not the last */ + if (sizek) { + while (range0_basek + range0_sizek > (basek + sizek)) { + if (range0_sizek >= chunk_sizek) + range0_sizek -= chunk_sizek; + else + range0_sizek = 0; + + if (!range0_sizek) + break; + } + } + +second_try: + range_basek = range0_basek + range0_sizek; + + /* one hole in the middle */ + if (range_basek > basek && range_basek <= (basek + sizek)) + second_sizek = range_basek - basek; + + if (range0_sizek > state->range_sizek) { + + /* one hole in middle or at end */ + hole_sizek = range0_sizek - state->range_sizek - second_sizek; + + /* hole size should be less than half of range0 size */ + if (hole_sizek >= (range0_sizek >> 1) && + range0_sizek >= chunk_sizek) { + range0_sizek -= chunk_sizek; + second_sizek = 0; + hole_sizek = 0; + + goto second_try; + } + } + + if (range0_sizek) { + if (debug_print) + printk(KERN_DEBUG "range0: %016lx - %016lx\n", + range0_basek<<10, + (range0_basek + range0_sizek)<<10); + state->reg = range_to_mtrr(state->reg, range0_basek, + range0_sizek, MTRR_TYPE_WRBACK); + } + + if (range0_sizek < state->range_sizek) { + /* need to handle left over */ + range_sizek = state->range_sizek - range0_sizek; + + if (debug_print) + printk(KERN_DEBUG "range: %016lx - %016lx\n", + range_basek<<10, + (range_basek + range_sizek)<<10); + state->reg = range_to_mtrr(state->reg, range_basek, + range_sizek, MTRR_TYPE_WRBACK); + } + + if (hole_sizek) { + hole_basek = range_basek - hole_sizek - second_sizek; + if (debug_print) + printk(KERN_DEBUG "hole: %016lx - %016lx\n", + hole_basek<<10, + (hole_basek + hole_sizek)<<10); + state->reg = range_to_mtrr(state->reg, hole_basek, + hole_sizek, MTRR_TYPE_UNCACHABLE); + } + + return second_sizek; +} + +static void __init +set_var_mtrr_range(struct var_mtrr_state *state, unsigned long base_pfn, + unsigned long size_pfn) +{ + unsigned long basek, sizek; + unsigned long second_sizek = 0; + + if (state->reg >= num_var_ranges) + return; + + basek = base_pfn << (PAGE_SHIFT - 10); + sizek = size_pfn << (PAGE_SHIFT - 10); + + /* See if I can merge with the last range */ + if ((basek <= 1024) || + (state->range_startk + state->range_sizek == basek)) { + unsigned long endk = basek + sizek; + state->range_sizek = endk - state->range_startk; + return; + } + /* Write the range mtrrs */ + if (state->range_sizek != 0) + second_sizek = range_to_mtrr_with_hole(state, basek, sizek); + + /* Allocate an msr */ + state->range_startk = basek + second_sizek; + state->range_sizek = sizek - second_sizek; +} + +/* mininum size of mtrr block that can take hole */ +static u64 mtrr_chunk_size __initdata = (256ULL<<20); + +static int __init parse_mtrr_chunk_size_opt(char *p) +{ + if (!p) + return -EINVAL; + mtrr_chunk_size = memparse(p, &p); + return 0; +} +early_param("mtrr_chunk_size", parse_mtrr_chunk_size_opt); + +/* granity of mtrr of block */ +static u64 mtrr_gran_size __initdata; + +static int __init parse_mtrr_gran_size_opt(char *p) +{ + if (!p) + return -EINVAL; + mtrr_gran_size = memparse(p, &p); + return 0; +} +early_param("mtrr_gran_size", parse_mtrr_gran_size_opt); + +static int nr_mtrr_spare_reg __initdata = + CONFIG_MTRR_SANITIZER_SPARE_REG_NR_DEFAULT; + +static int __init parse_mtrr_spare_reg(char *arg) +{ + if (arg) + nr_mtrr_spare_reg = simple_strtoul(arg, NULL, 0); + return 0; +} + +early_param("mtrr_spare_reg_nr", parse_mtrr_spare_reg); + +static int __init +x86_setup_var_mtrrs(struct res_range *range, int nr_range, + u64 chunk_size, u64 gran_size) +{ + struct var_mtrr_state var_state; + int i; + int num_reg; + + var_state.range_startk = 0; + var_state.range_sizek = 0; + var_state.reg = 0; + var_state.chunk_sizek = chunk_size >> 10; + var_state.gran_sizek = gran_size >> 10; + + memset(range_state, 0, sizeof(range_state)); + + /* Write the range etc */ + for (i = 0; i < nr_range; i++) + set_var_mtrr_range(&var_state, range[i].start, + range[i].end - range[i].start + 1); + + /* Write the last range */ + if (var_state.range_sizek != 0) + range_to_mtrr_with_hole(&var_state, 0, 0); + + num_reg = var_state.reg; + /* Clear out the extra MTRR's */ + while (var_state.reg < num_var_ranges) { + save_var_mtrr(var_state.reg, 0, 0, 0); + var_state.reg++; + } + + return num_reg; +} + +struct mtrr_cleanup_result { + unsigned long gran_sizek; + unsigned long chunk_sizek; + unsigned long lose_cover_sizek; + unsigned int num_reg; + int bad; +}; + +/* + * gran_size: 64K, 128K, 256K, 512K, 1M, 2M, ..., 2G + * chunk size: gran_size, ..., 2G + * so we need (1+16)*8 + */ +#define NUM_RESULT 136 +#define PSHIFT (PAGE_SHIFT - 10) + +static struct mtrr_cleanup_result __initdata result[NUM_RESULT]; +static unsigned long __initdata min_loss_pfn[RANGE_NUM]; + +static void __init print_out_mtrr_range_state(void) +{ + int i; + char start_factor = 'K', size_factor = 'K'; + unsigned long start_base, size_base; + mtrr_type type; + + for (i = 0; i < num_var_ranges; i++) { + + size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10); + if (!size_base) + continue; + + size_base = to_size_factor(size_base, &size_factor), + start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10); + start_base = to_size_factor(start_base, &start_factor), + type = range_state[i].type; + + printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n", + i, start_base, start_factor, + size_base, size_factor, + (type == MTRR_TYPE_UNCACHABLE) ? "UC" : + ((type == MTRR_TYPE_WRPROT) ? "WP" : + ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other")) + ); + } +} + +static int __init mtrr_need_cleanup(void) +{ + int i; + mtrr_type type; + unsigned long size; + /* extra one for all 0 */ + int num[MTRR_NUM_TYPES + 1]; + + /* check entries number */ + memset(num, 0, sizeof(num)); + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + size = range_state[i].size_pfn; + if (type >= MTRR_NUM_TYPES) + continue; + if (!size) + type = MTRR_NUM_TYPES; + if (type == MTRR_TYPE_WRPROT) + type = MTRR_TYPE_UNCACHABLE; + num[type]++; + } + + /* check if we got UC entries */ + if (!num[MTRR_TYPE_UNCACHABLE]) + return 0; + + /* check if we only had WB and UC */ + if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != + num_var_ranges - num[MTRR_NUM_TYPES]) + return 0; + + return 1; +} + +static unsigned long __initdata range_sums; +static void __init mtrr_calc_range_state(u64 chunk_size, u64 gran_size, + unsigned long extra_remove_base, + unsigned long extra_remove_size, + int i) +{ + int num_reg; + static struct res_range range_new[RANGE_NUM]; + static int nr_range_new; + unsigned long range_sums_new; + + /* convert ranges to var ranges state */ + num_reg = x86_setup_var_mtrrs(range, nr_range, + chunk_size, gran_size); + + /* we got new setting in range_state, check it */ + memset(range_new, 0, sizeof(range_new)); + nr_range_new = x86_get_mtrr_mem_range(range_new, 0, + extra_remove_base, extra_remove_size); + range_sums_new = sum_ranges(range_new, nr_range_new); + + result[i].chunk_sizek = chunk_size >> 10; + result[i].gran_sizek = gran_size >> 10; + result[i].num_reg = num_reg; + if (range_sums < range_sums_new) { + result[i].lose_cover_sizek = + (range_sums_new - range_sums) << PSHIFT; + result[i].bad = 1; + } else + result[i].lose_cover_sizek = + (range_sums - range_sums_new) << PSHIFT; + + /* double check it */ + if (!result[i].bad && !result[i].lose_cover_sizek) { + if (nr_range_new != nr_range || + memcmp(range, range_new, sizeof(range))) + result[i].bad = 1; + } + + if (!result[i].bad && (range_sums - range_sums_new < + min_loss_pfn[num_reg])) { + min_loss_pfn[num_reg] = + range_sums - range_sums_new; + } +} + +static void __init mtrr_print_out_one_result(int i) +{ + char gran_factor, chunk_factor, lose_factor; + unsigned long gran_base, chunk_base, lose_base; + + gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), + chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), + lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), + printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t", + result[i].bad ? "*BAD*" : " ", + gran_base, gran_factor, chunk_base, chunk_factor); + printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ld%c\n", + result[i].num_reg, result[i].bad ? "-" : "", + lose_base, lose_factor); +} + +static int __init mtrr_search_optimal_index(void) +{ + int i; + int num_reg_good; + int index_good; + + if (nr_mtrr_spare_reg >= num_var_ranges) + nr_mtrr_spare_reg = num_var_ranges - 1; + num_reg_good = -1; + for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) { + if (!min_loss_pfn[i]) + num_reg_good = i; + } + + index_good = -1; + if (num_reg_good != -1) { + for (i = 0; i < NUM_RESULT; i++) { + if (!result[i].bad && + result[i].num_reg == num_reg_good && + !result[i].lose_cover_sizek) { + index_good = i; + break; + } + } + } + + return index_good; +} + + +int __init mtrr_cleanup(unsigned address_bits) +{ + unsigned long extra_remove_base, extra_remove_size; + unsigned long base, size, def, dummy; + mtrr_type type; + u64 chunk_size, gran_size; + int index_good; + int i; + + if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1) + return 0; + rdmsr(MTRRdefType_MSR, def, dummy); + def &= 0xff; + if (def != MTRR_TYPE_UNCACHABLE) + return 0; + + /* get it and store it aside */ + memset(range_state, 0, sizeof(range_state)); + for (i = 0; i < num_var_ranges; i++) { + mtrr_if->get(i, &base, &size, &type); + range_state[i].base_pfn = base; + range_state[i].size_pfn = size; + range_state[i].type = type; + } + + /* check if we need handle it and can handle it */ + if (!mtrr_need_cleanup()) + return 0; + + /* print original var MTRRs at first, for debugging: */ + printk(KERN_DEBUG "original variable MTRRs\n"); + print_out_mtrr_range_state(); + + memset(range, 0, sizeof(range)); + extra_remove_size = 0; + extra_remove_base = 1 << (32 - PAGE_SHIFT); + if (mtrr_tom2) + extra_remove_size = + (mtrr_tom2 >> PAGE_SHIFT) - extra_remove_base; + nr_range = x86_get_mtrr_mem_range(range, 0, extra_remove_base, + extra_remove_size); + /* + * [0, 1M) should always be coverred by var mtrr with WB + * and fixed mtrrs should take effective before var mtrr for it + */ + nr_range = add_range_with_merge(range, nr_range, 0, + (1ULL<<(20 - PAGE_SHIFT)) - 1); + /* sort the ranges */ + sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); + + range_sums = sum_ranges(range, nr_range); + printk(KERN_INFO "total RAM coverred: %ldM\n", + range_sums >> (20 - PAGE_SHIFT)); + + if (mtrr_chunk_size && mtrr_gran_size) { + i = 0; + mtrr_calc_range_state(mtrr_chunk_size, mtrr_gran_size, + extra_remove_base, extra_remove_size, i); + + mtrr_print_out_one_result(i); + + if (!result[i].bad) { + set_var_mtrr_all(address_bits); + printk(KERN_DEBUG "New variable MTRRs\n"); + print_out_mtrr_range_state(); + return 1; + } + printk(KERN_INFO "invalid mtrr_gran_size or mtrr_chunk_size, " + "will find optimal one\n"); + } + + i = 0; + memset(min_loss_pfn, 0xff, sizeof(min_loss_pfn)); + memset(result, 0, sizeof(result)); + for (gran_size = (1ULL<<16); gran_size < (1ULL<<32); gran_size <<= 1) { + + for (chunk_size = gran_size; chunk_size < (1ULL<<32); + chunk_size <<= 1) { + + if (i >= NUM_RESULT) + continue; + + mtrr_calc_range_state(chunk_size, gran_size, + extra_remove_base, extra_remove_size, i); + if (debug_print) { + mtrr_print_out_one_result(i); + printk(KERN_INFO "\n"); + } + + i++; + } + } + + /* try to find the optimal index */ + index_good = mtrr_search_optimal_index(); + + if (index_good != -1) { + printk(KERN_INFO "Found optimal setting for mtrr clean up\n"); + i = index_good; + mtrr_print_out_one_result(i); + + /* convert ranges to var ranges state */ + chunk_size = result[i].chunk_sizek; + chunk_size <<= 10; + gran_size = result[i].gran_sizek; + gran_size <<= 10; + x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size); + set_var_mtrr_all(address_bits); + printk(KERN_DEBUG "New variable MTRRs\n"); + print_out_mtrr_range_state(); + return 1; + } else { + /* print out all */ + for (i = 0; i < NUM_RESULT; i++) + mtrr_print_out_one_result(i); + } + + printk(KERN_INFO "mtrr_cleanup: can not find optimal value\n"); + printk(KERN_INFO "please specify mtrr_gran_size/mtrr_chunk_size\n"); + + return 0; +} +#else +int __init mtrr_cleanup(unsigned address_bits) +{ + return 0; +} +#endif + +static int disable_mtrr_trim; + +static int __init disable_mtrr_trim_setup(char *str) +{ + disable_mtrr_trim = 1; + return 0; +} +early_param("disable_mtrr_trim", disable_mtrr_trim_setup); + +/* + * Newer AMD K8s and later CPUs have a special magic MSR way to force WB + * for memory >4GB. Check for that here. + * Note this won't check if the MTRRs < 4GB where the magic bit doesn't + * apply to are wrong, but so far we don't know of any such case in the wild. + */ +#define Tom2Enabled (1U << 21) +#define Tom2ForceMemTypeWB (1U << 22) + +int __init amd_special_default_mtrr(void) +{ + u32 l, h; + + if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) + return 0; + if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11) + return 0; + /* In case some hypervisor doesn't pass SYSCFG through */ + if (rdmsr_safe(MSR_K8_SYSCFG, &l, &h) < 0) + return 0; + /* + * Memory between 4GB and top of mem is forced WB by this magic bit. + * Reserved before K8RevF, but should be zero there. + */ + if ((l & (Tom2Enabled | Tom2ForceMemTypeWB)) == + (Tom2Enabled | Tom2ForceMemTypeWB)) + return 1; + return 0; +} + +static u64 __init real_trim_memory(unsigned long start_pfn, + unsigned long limit_pfn) +{ + u64 trim_start, trim_size; + trim_start = start_pfn; + trim_start <<= PAGE_SHIFT; + trim_size = limit_pfn; + trim_size <<= PAGE_SHIFT; + trim_size -= trim_start; + + return e820_update_range(trim_start, trim_size, E820_RAM, + E820_RESERVED); +} +/** + * mtrr_trim_uncached_memory - trim RAM not covered by MTRRs + * @end_pfn: ending page frame number + * + * Some buggy BIOSes don't setup the MTRRs properly for systems with certain + * memory configurations. This routine checks that the highest MTRR matches + * the end of memory, to make sure the MTRRs having a write back type cover + * all of the memory the kernel is intending to use. If not, it'll trim any + * memory off the end by adjusting end_pfn, removing it from the kernel's + * allocation pools, warning the user with an obnoxious message. + */ +int __init mtrr_trim_uncached_memory(unsigned long end_pfn) +{ + unsigned long i, base, size, highest_pfn = 0, def, dummy; + mtrr_type type; + u64 total_trim_size; + + /* extra one for all 0 */ + int num[MTRR_NUM_TYPES + 1]; + /* + * Make sure we only trim uncachable memory on machines that + * support the Intel MTRR architecture: + */ + if (!is_cpu(INTEL) || disable_mtrr_trim) + return 0; + rdmsr(MTRRdefType_MSR, def, dummy); + def &= 0xff; + if (def != MTRR_TYPE_UNCACHABLE) + return 0; + + /* get it and store it aside */ + memset(range_state, 0, sizeof(range_state)); + for (i = 0; i < num_var_ranges; i++) { + mtrr_if->get(i, &base, &size, &type); + range_state[i].base_pfn = base; + range_state[i].size_pfn = size; + range_state[i].type = type; + } + + /* Find highest cached pfn */ + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + if (type != MTRR_TYPE_WRBACK) + continue; + base = range_state[i].base_pfn; + size = range_state[i].size_pfn; + if (highest_pfn < base + size) + highest_pfn = base + size; + } + + /* kvm/qemu doesn't have mtrr set right, don't trim them all */ + if (!highest_pfn) { + printk(KERN_INFO "CPU MTRRs all blank - virtualized system.\n"); + return 0; + } + + /* check entries number */ + memset(num, 0, sizeof(num)); + for (i = 0; i < num_var_ranges; i++) { + type = range_state[i].type; + if (type >= MTRR_NUM_TYPES) + continue; + size = range_state[i].size_pfn; + if (!size) + type = MTRR_NUM_TYPES; + num[type]++; + } + + /* no entry for WB? */ + if (!num[MTRR_TYPE_WRBACK]) + return 0; + + /* check if we only had WB and UC */ + if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != + num_var_ranges - num[MTRR_NUM_TYPES]) + return 0; + + memset(range, 0, sizeof(range)); + nr_range = 0; + if (mtrr_tom2) { + range[nr_range].start = (1ULL<<(32 - PAGE_SHIFT)); + range[nr_range].end = (mtrr_tom2 >> PAGE_SHIFT) - 1; + if (highest_pfn < range[nr_range].end + 1) + highest_pfn = range[nr_range].end + 1; + nr_range++; + } + nr_range = x86_get_mtrr_mem_range(range, nr_range, 0, 0); + + total_trim_size = 0; + /* check the head */ + if (range[0].start) + total_trim_size += real_trim_memory(0, range[0].start); + /* check the holes */ + for (i = 0; i < nr_range - 1; i++) { + if (range[i].end + 1 < range[i+1].start) + total_trim_size += real_trim_memory(range[i].end + 1, + range[i+1].start); + } + /* check the top */ + i = nr_range - 1; + if (range[i].end + 1 < end_pfn) + total_trim_size += real_trim_memory(range[i].end + 1, + end_pfn); + + if (total_trim_size) { + printk(KERN_WARNING "WARNING: BIOS bug: CPU MTRRs don't cover" + " all of memory, losing %lluMB of RAM.\n", + total_trim_size >> 20); + + if (!changed_by_mtrr_cleanup) + WARN_ON(1); + + printk(KERN_INFO "update e820 for mtrr\n"); + update_e820(); + + return 1; + } + + return 0; +} + diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index 0c0a455fe95..37f28fc7cf9 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -33,13 +33,31 @@ u64 mtrr_tom2; struct mtrr_state_type mtrr_state = {}; EXPORT_SYMBOL_GPL(mtrr_state); -static int __initdata mtrr_show; -static int __init mtrr_debug(char *opt) +/** + * BIOS is expected to clear MtrrFixDramModEn bit, see for example + * "BIOS and Kernel Developer's Guide for the AMD Athlon 64 and AMD + * Opteron Processors" (26094 Rev. 3.30 February 2006), section + * "13.2.1.2 SYSCFG Register": "The MtrrFixDramModEn bit should be set + * to 1 during BIOS initalization of the fixed MTRRs, then cleared to + * 0 for operation." + */ +static inline void k8_check_syscfg_dram_mod_en(void) { - mtrr_show = 1; - return 0; + u32 lo, hi; + + if (!((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && + (boot_cpu_data.x86 >= 0x0f))) + return; + + rdmsr(MSR_K8_SYSCFG, lo, hi); + if (lo & K8_MTRRFIXRANGE_DRAM_MODIFY) { + printk(KERN_ERR FW_WARN "MTRR: CPU %u: SYSCFG[MtrrFixDramModEn]" + " not cleared by BIOS, clearing this bit\n", + smp_processor_id()); + lo &= ~K8_MTRRFIXRANGE_DRAM_MODIFY; + mtrr_wrmsr(MSR_K8_SYSCFG, lo, hi); + } } -early_param("mtrr.show", mtrr_debug); /* * Returns the effective MTRR type for the region @@ -174,6 +192,8 @@ get_fixed_ranges(mtrr_type * frs) unsigned int *p = (unsigned int *) frs; int i; + k8_check_syscfg_dram_mod_en(); + rdmsr(MTRRfix64K_00000_MSR, p[0], p[1]); for (i = 0; i < 2; i++) @@ -188,18 +208,94 @@ void mtrr_save_fixed_ranges(void *info) get_fixed_ranges(mtrr_state.fixed_ranges); } -static void print_fixed(unsigned base, unsigned step, const mtrr_type*types) +static unsigned __initdata last_fixed_start; +static unsigned __initdata last_fixed_end; +static mtrr_type __initdata last_fixed_type; + +static void __init print_fixed_last(void) +{ + if (!last_fixed_end) + return; + + printk(KERN_DEBUG " %05X-%05X %s\n", last_fixed_start, + last_fixed_end - 1, mtrr_attrib_to_str(last_fixed_type)); + + last_fixed_end = 0; +} + +static void __init update_fixed_last(unsigned base, unsigned end, + mtrr_type type) +{ + last_fixed_start = base; + last_fixed_end = end; + last_fixed_type = type; +} + +static void __init print_fixed(unsigned base, unsigned step, + const mtrr_type *types) { unsigned i; - for (i = 0; i < 8; ++i, ++types, base += step) - printk(KERN_INFO "MTRR %05X-%05X %s\n", - base, base + step - 1, mtrr_attrib_to_str(*types)); + for (i = 0; i < 8; ++i, ++types, base += step) { + if (last_fixed_end == 0) { + update_fixed_last(base, base + step, *types); + continue; + } + if (last_fixed_end == base && last_fixed_type == *types) { + last_fixed_end = base + step; + continue; + } + /* new segments: gap or different type */ + print_fixed_last(); + update_fixed_last(base, base + step, *types); + } } static void prepare_set(void); static void post_set(void); +static void __init print_mtrr_state(void) +{ + unsigned int i; + int high_width; + + printk(KERN_DEBUG "MTRR default type: %s\n", + mtrr_attrib_to_str(mtrr_state.def_type)); + if (mtrr_state.have_fixed) { + printk(KERN_DEBUG "MTRR fixed ranges %sabled:\n", + mtrr_state.enabled & 1 ? "en" : "dis"); + print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0); + for (i = 0; i < 2; ++i) + print_fixed(0x80000 + i * 0x20000, 0x04000, mtrr_state.fixed_ranges + (i + 1) * 8); + for (i = 0; i < 8; ++i) + print_fixed(0xC0000 + i * 0x08000, 0x01000, mtrr_state.fixed_ranges + (i + 3) * 8); + + /* tail */ + print_fixed_last(); + } + printk(KERN_DEBUG "MTRR variable ranges %sabled:\n", + mtrr_state.enabled & 2 ? "en" : "dis"); + high_width = ((size_or_mask ? ffs(size_or_mask) - 1 : 32) - (32 - PAGE_SHIFT) + 3) / 4; + for (i = 0; i < num_var_ranges; ++i) { + if (mtrr_state.var_ranges[i].mask_lo & (1 << 11)) + printk(KERN_DEBUG " %u base %0*X%05X000 mask %0*X%05X000 %s\n", + i, + high_width, + mtrr_state.var_ranges[i].base_hi, + mtrr_state.var_ranges[i].base_lo >> 12, + high_width, + mtrr_state.var_ranges[i].mask_hi, + mtrr_state.var_ranges[i].mask_lo >> 12, + mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff)); + else + printk(KERN_DEBUG " %u disabled\n", i); + } + if (mtrr_tom2) { + printk(KERN_DEBUG "TOM2: %016llx aka %lldM\n", + mtrr_tom2, mtrr_tom2>>20); + } +} + /* Grab all of the MTRR state for this CPU into *state */ void __init get_mtrr_state(void) { @@ -231,41 +327,9 @@ void __init get_mtrr_state(void) mtrr_tom2 |= low; mtrr_tom2 &= 0xffffff800000ULL; } - if (mtrr_show) { - int high_width; - - printk(KERN_INFO "MTRR default type: %s\n", mtrr_attrib_to_str(mtrr_state.def_type)); - if (mtrr_state.have_fixed) { - printk(KERN_INFO "MTRR fixed ranges %sabled:\n", - mtrr_state.enabled & 1 ? "en" : "dis"); - print_fixed(0x00000, 0x10000, mtrr_state.fixed_ranges + 0); - for (i = 0; i < 2; ++i) - print_fixed(0x80000 + i * 0x20000, 0x04000, mtrr_state.fixed_ranges + (i + 1) * 8); - for (i = 0; i < 8; ++i) - print_fixed(0xC0000 + i * 0x08000, 0x01000, mtrr_state.fixed_ranges + (i + 3) * 8); - } - printk(KERN_INFO "MTRR variable ranges %sabled:\n", - mtrr_state.enabled & 2 ? "en" : "dis"); - high_width = ((size_or_mask ? ffs(size_or_mask) - 1 : 32) - (32 - PAGE_SHIFT) + 3) / 4; - for (i = 0; i < num_var_ranges; ++i) { - if (mtrr_state.var_ranges[i].mask_lo & (1 << 11)) - printk(KERN_INFO "MTRR %u base %0*X%05X000 mask %0*X%05X000 %s\n", - i, - high_width, - mtrr_state.var_ranges[i].base_hi, - mtrr_state.var_ranges[i].base_lo >> 12, - high_width, - mtrr_state.var_ranges[i].mask_hi, - mtrr_state.var_ranges[i].mask_lo >> 12, - mtrr_attrib_to_str(mtrr_state.var_ranges[i].base_lo & 0xff)); - else - printk(KERN_INFO "MTRR %u disabled\n", i); - } - if (mtrr_tom2) { - printk(KERN_INFO "TOM2: %016llx aka %lldM\n", - mtrr_tom2, mtrr_tom2>>20); - } - } + + print_mtrr_state(); + mtrr_state_set = 1; /* PAT setup for BP. We need to go through sync steps here */ @@ -308,27 +372,10 @@ void mtrr_wrmsr(unsigned msr, unsigned a, unsigned b) } /** - * Enable and allow read/write of extended fixed-range MTRR bits on K8 CPUs - * see AMD publication no. 24593, chapter 3.2.1 for more information - */ -static inline void k8_enable_fixed_iorrs(void) -{ - unsigned lo, hi; - - rdmsr(MSR_K8_SYSCFG, lo, hi); - mtrr_wrmsr(MSR_K8_SYSCFG, lo - | K8_MTRRFIXRANGE_DRAM_ENABLE - | K8_MTRRFIXRANGE_DRAM_MODIFY, hi); -} - -/** * set_fixed_range - checks & updates a fixed-range MTRR if it differs from the value it should have * @msr: MSR address of the MTTR which should be checked and updated * @changed: pointer which indicates whether the MTRR needed to be changed * @msrwords: pointer to the MSR values which the MSR should have - * - * If K8 extentions are wanted, update the K8 SYSCFG MSR also. - * See AMD publication no. 24593, chapter 7.8.1, page 233 for more information. */ static void set_fixed_range(int msr, bool *changed, unsigned int *msrwords) { @@ -337,10 +384,6 @@ static void set_fixed_range(int msr, bool *changed, unsigned int *msrwords) rdmsr(msr, lo, hi); if (lo != msrwords[0] || hi != msrwords[1]) { - if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD && - (boot_cpu_data.x86 >= 0x0f && boot_cpu_data.x86 <= 0x11) && - ((msrwords[0] | msrwords[1]) & K8_MTRR_RDMEM_WRMEM_MASK)) - k8_enable_fixed_iorrs(); mtrr_wrmsr(msr, msrwords[0], msrwords[1]); *changed = true; } @@ -376,22 +419,31 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base, { unsigned int mask_lo, mask_hi, base_lo, base_hi; unsigned int tmp, hi; + int cpu; + + /* + * get_mtrr doesn't need to update mtrr_state, also it could be called + * from any cpu, so try to print it out directly. + */ + cpu = get_cpu(); rdmsr(MTRRphysMask_MSR(reg), mask_lo, mask_hi); + if ((mask_lo & 0x800) == 0) { /* Invalid (i.e. free) range */ *base = 0; *size = 0; *type = 0; - return; + goto out_put_cpu; } rdmsr(MTRRphysBase_MSR(reg), base_lo, base_hi); - /* Work out the shifted address mask. */ + /* Work out the shifted address mask: */ tmp = mask_hi << (32 - PAGE_SHIFT) | mask_lo >> PAGE_SHIFT; mask_lo = size_or_mask | tmp; - /* Expand tmp with high bits to all 1s*/ + + /* Expand tmp with high bits to all 1s: */ hi = fls(tmp); if (hi > 0) { tmp |= ~((1<<(hi - 1)) - 1); @@ -402,11 +454,19 @@ static void generic_get_mtrr(unsigned int reg, unsigned long *base, } } - /* This works correctly if size is a power of two, i.e. a - contiguous range. */ + /* + * This works correctly if size is a power of two, i.e. a + * contiguous range: + */ *size = -mask_lo; *base = base_hi << (32 - PAGE_SHIFT) | base_lo >> PAGE_SHIFT; *type = base_lo & 0xff; + + printk(KERN_DEBUG " get_mtrr: cpu%d reg%02d base=%010lx size=%010lx %s\n", + cpu, reg, *base, *size, + mtrr_attrib_to_str(*type & 0xff)); +out_put_cpu: + put_cpu(); } /** @@ -419,6 +479,8 @@ static int set_fixed_ranges(mtrr_type * frs) bool changed = false; int block=-1, range; + k8_check_syscfg_dram_mod_en(); + while (fixed_range_blocks[++block].ranges) for (range=0; range < fixed_range_blocks[block].ranges; range++) set_fixed_range(fixed_range_blocks[block].base_msr + range, diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c index 236a401b825..03cda01f57c 100644 --- a/arch/x86/kernel/cpu/mtrr/main.c +++ b/arch/x86/kernel/cpu/mtrr/main.c @@ -574,7 +574,7 @@ struct mtrr_value { unsigned long lsize; }; -static struct mtrr_value mtrr_state[MTRR_MAX_VAR_RANGES]; +static struct mtrr_value mtrr_value[MTRR_MAX_VAR_RANGES]; static int mtrr_save(struct sys_device * sysdev, pm_message_t state) { @@ -582,9 +582,9 @@ static int mtrr_save(struct sys_device * sysdev, pm_message_t state) for (i = 0; i < num_var_ranges; i++) { mtrr_if->get(i, - &mtrr_state[i].lbase, - &mtrr_state[i].lsize, - &mtrr_state[i].ltype); + &mtrr_value[i].lbase, + &mtrr_value[i].lsize, + &mtrr_value[i].ltype); } return 0; } @@ -594,11 +594,11 @@ static int mtrr_restore(struct sys_device * sysdev) int i; for (i = 0; i < num_var_ranges; i++) { - if (mtrr_state[i].lsize) + if (mtrr_value[i].lsize) set_mtrr(i, - mtrr_state[i].lbase, - mtrr_state[i].lsize, - mtrr_state[i].ltype); + mtrr_value[i].lbase, + mtrr_value[i].lsize, + mtrr_value[i].ltype); } return 0; } @@ -610,1058 +610,7 @@ static struct sysdev_driver mtrr_sysdev_driver = { .resume = mtrr_restore, }; -/* should be related to MTRR_VAR_RANGES nums */ -#define RANGE_NUM 256 - -struct res_range { - unsigned long start; - unsigned long end; -}; - -static int __init -add_range(struct res_range *range, int nr_range, unsigned long start, - unsigned long end) -{ - /* out of slots */ - if (nr_range >= RANGE_NUM) - return nr_range; - - range[nr_range].start = start; - range[nr_range].end = end; - - nr_range++; - - return nr_range; -} - -static int __init -add_range_with_merge(struct res_range *range, int nr_range, unsigned long start, - unsigned long end) -{ - int i; - - /* try to merge it with old one */ - for (i = 0; i < nr_range; i++) { - unsigned long final_start, final_end; - unsigned long common_start, common_end; - - if (!range[i].end) - continue; - - common_start = max(range[i].start, start); - common_end = min(range[i].end, end); - if (common_start > common_end + 1) - continue; - - final_start = min(range[i].start, start); - final_end = max(range[i].end, end); - - range[i].start = final_start; - range[i].end = final_end; - return nr_range; - } - - /* need to add that */ - return add_range(range, nr_range, start, end); -} - -static void __init -subtract_range(struct res_range *range, unsigned long start, unsigned long end) -{ - int i, j; - - for (j = 0; j < RANGE_NUM; j++) { - if (!range[j].end) - continue; - - if (start <= range[j].start && end >= range[j].end) { - range[j].start = 0; - range[j].end = 0; - continue; - } - - if (start <= range[j].start && end < range[j].end && - range[j].start < end + 1) { - range[j].start = end + 1; - continue; - } - - - if (start > range[j].start && end >= range[j].end && - range[j].end > start - 1) { - range[j].end = start - 1; - continue; - } - - if (start > range[j].start && end < range[j].end) { - /* find the new spare */ - for (i = 0; i < RANGE_NUM; i++) { - if (range[i].end == 0) - break; - } - if (i < RANGE_NUM) { - range[i].end = range[j].end; - range[i].start = end + 1; - } else { - printk(KERN_ERR "run of slot in ranges\n"); - } - range[j].end = start - 1; - continue; - } - } -} - -static int __init cmp_range(const void *x1, const void *x2) -{ - const struct res_range *r1 = x1; - const struct res_range *r2 = x2; - long start1, start2; - - start1 = r1->start; - start2 = r2->start; - - return start1 - start2; -} - -struct var_mtrr_range_state { - unsigned long base_pfn; - unsigned long size_pfn; - mtrr_type type; -}; - -static struct var_mtrr_range_state __initdata range_state[RANGE_NUM]; -static int __initdata debug_print; - -static int __init -x86_get_mtrr_mem_range(struct res_range *range, int nr_range, - unsigned long extra_remove_base, - unsigned long extra_remove_size) -{ - unsigned long i, base, size; - mtrr_type type; - - for (i = 0; i < num_var_ranges; i++) { - type = range_state[i].type; - if (type != MTRR_TYPE_WRBACK) - continue; - base = range_state[i].base_pfn; - size = range_state[i].size_pfn; - nr_range = add_range_with_merge(range, nr_range, base, - base + size - 1); - } - if (debug_print) { - printk(KERN_DEBUG "After WB checking\n"); - for (i = 0; i < nr_range; i++) - printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", - range[i].start, range[i].end + 1); - } - - /* take out UC ranges */ - for (i = 0; i < num_var_ranges; i++) { - type = range_state[i].type; - if (type != MTRR_TYPE_UNCACHABLE && - type != MTRR_TYPE_WRPROT) - continue; - size = range_state[i].size_pfn; - if (!size) - continue; - base = range_state[i].base_pfn; - subtract_range(range, base, base + size - 1); - } - if (extra_remove_size) - subtract_range(range, extra_remove_base, - extra_remove_base + extra_remove_size - 1); - - /* get new range num */ - nr_range = 0; - for (i = 0; i < RANGE_NUM; i++) { - if (!range[i].end) - continue; - nr_range++; - } - if (debug_print) { - printk(KERN_DEBUG "After UC checking\n"); - for (i = 0; i < nr_range; i++) - printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", - range[i].start, range[i].end + 1); - } - - /* sort the ranges */ - sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); - if (debug_print) { - printk(KERN_DEBUG "After sorting\n"); - for (i = 0; i < nr_range; i++) - printk(KERN_DEBUG "MTRR MAP PFN: %016lx - %016lx\n", - range[i].start, range[i].end + 1); - } - - /* clear those is not used */ - for (i = nr_range; i < RANGE_NUM; i++) - memset(&range[i], 0, sizeof(range[i])); - - return nr_range; -} - -static struct res_range __initdata range[RANGE_NUM]; -static int __initdata nr_range; - -#ifdef CONFIG_MTRR_SANITIZER - -static unsigned long __init sum_ranges(struct res_range *range, int nr_range) -{ - unsigned long sum; - int i; - - sum = 0; - for (i = 0; i < nr_range; i++) - sum += range[i].end + 1 - range[i].start; - - return sum; -} - -static int enable_mtrr_cleanup __initdata = - CONFIG_MTRR_SANITIZER_ENABLE_DEFAULT; - -static int __init disable_mtrr_cleanup_setup(char *str) -{ - enable_mtrr_cleanup = 0; - return 0; -} -early_param("disable_mtrr_cleanup", disable_mtrr_cleanup_setup); - -static int __init enable_mtrr_cleanup_setup(char *str) -{ - enable_mtrr_cleanup = 1; - return 0; -} -early_param("enable_mtrr_cleanup", enable_mtrr_cleanup_setup); - -static int __init mtrr_cleanup_debug_setup(char *str) -{ - debug_print = 1; - return 0; -} -early_param("mtrr_cleanup_debug", mtrr_cleanup_debug_setup); - -struct var_mtrr_state { - unsigned long range_startk; - unsigned long range_sizek; - unsigned long chunk_sizek; - unsigned long gran_sizek; - unsigned int reg; -}; - -static void __init -set_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, - unsigned char type, unsigned int address_bits) -{ - u32 base_lo, base_hi, mask_lo, mask_hi; - u64 base, mask; - - if (!sizek) { - fill_mtrr_var_range(reg, 0, 0, 0, 0); - return; - } - - mask = (1ULL << address_bits) - 1; - mask &= ~((((u64)sizek) << 10) - 1); - - base = ((u64)basek) << 10; - - base |= type; - mask |= 0x800; - - base_lo = base & ((1ULL<<32) - 1); - base_hi = base >> 32; - - mask_lo = mask & ((1ULL<<32) - 1); - mask_hi = mask >> 32; - - fill_mtrr_var_range(reg, base_lo, base_hi, mask_lo, mask_hi); -} - -static void __init -save_var_mtrr(unsigned int reg, unsigned long basek, unsigned long sizek, - unsigned char type) -{ - range_state[reg].base_pfn = basek >> (PAGE_SHIFT - 10); - range_state[reg].size_pfn = sizek >> (PAGE_SHIFT - 10); - range_state[reg].type = type; -} - -static void __init -set_var_mtrr_all(unsigned int address_bits) -{ - unsigned long basek, sizek; - unsigned char type; - unsigned int reg; - - for (reg = 0; reg < num_var_ranges; reg++) { - basek = range_state[reg].base_pfn << (PAGE_SHIFT - 10); - sizek = range_state[reg].size_pfn << (PAGE_SHIFT - 10); - type = range_state[reg].type; - - set_var_mtrr(reg, basek, sizek, type, address_bits); - } -} - -static unsigned long to_size_factor(unsigned long sizek, char *factorp) -{ - char factor; - unsigned long base = sizek; - - if (base & ((1<<10) - 1)) { - /* not MB alignment */ - factor = 'K'; - } else if (base & ((1<<20) - 1)){ - factor = 'M'; - base >>= 10; - } else { - factor = 'G'; - base >>= 20; - } - - *factorp = factor; - - return base; -} - -static unsigned int __init -range_to_mtrr(unsigned int reg, unsigned long range_startk, - unsigned long range_sizek, unsigned char type) -{ - if (!range_sizek || (reg >= num_var_ranges)) - return reg; - - while (range_sizek) { - unsigned long max_align, align; - unsigned long sizek; - - /* Compute the maximum size I can make a range */ - if (range_startk) - max_align = ffs(range_startk) - 1; - else - max_align = 32; - align = fls(range_sizek) - 1; - if (align > max_align) - align = max_align; - - sizek = 1 << align; - if (debug_print) { - char start_factor = 'K', size_factor = 'K'; - unsigned long start_base, size_base; - - start_base = to_size_factor(range_startk, &start_factor), - size_base = to_size_factor(sizek, &size_factor), - - printk(KERN_DEBUG "Setting variable MTRR %d, " - "base: %ld%cB, range: %ld%cB, type %s\n", - reg, start_base, start_factor, - size_base, size_factor, - (type == MTRR_TYPE_UNCACHABLE)?"UC": - ((type == MTRR_TYPE_WRBACK)?"WB":"Other") - ); - } - save_var_mtrr(reg++, range_startk, sizek, type); - range_startk += sizek; - range_sizek -= sizek; - if (reg >= num_var_ranges) - break; - } - return reg; -} - -static unsigned __init -range_to_mtrr_with_hole(struct var_mtrr_state *state, unsigned long basek, - unsigned long sizek) -{ - unsigned long hole_basek, hole_sizek; - unsigned long second_basek, second_sizek; - unsigned long range0_basek, range0_sizek; - unsigned long range_basek, range_sizek; - unsigned long chunk_sizek; - unsigned long gran_sizek; - - hole_basek = 0; - hole_sizek = 0; - second_basek = 0; - second_sizek = 0; - chunk_sizek = state->chunk_sizek; - gran_sizek = state->gran_sizek; - - /* align with gran size, prevent small block used up MTRRs */ - range_basek = ALIGN(state->range_startk, gran_sizek); - if ((range_basek > basek) && basek) - return second_sizek; - state->range_sizek -= (range_basek - state->range_startk); - range_sizek = ALIGN(state->range_sizek, gran_sizek); - - while (range_sizek > state->range_sizek) { - range_sizek -= gran_sizek; - if (!range_sizek) - return 0; - } - state->range_sizek = range_sizek; - - /* try to append some small hole */ - range0_basek = state->range_startk; - range0_sizek = ALIGN(state->range_sizek, chunk_sizek); - - /* no increase */ - if (range0_sizek == state->range_sizek) { - if (debug_print) - printk(KERN_DEBUG "rangeX: %016lx - %016lx\n", - range0_basek<<10, - (range0_basek + state->range_sizek)<<10); - state->reg = range_to_mtrr(state->reg, range0_basek, - state->range_sizek, MTRR_TYPE_WRBACK); - return 0; - } - - /* only cut back, when it is not the last */ - if (sizek) { - while (range0_basek + range0_sizek > (basek + sizek)) { - if (range0_sizek >= chunk_sizek) - range0_sizek -= chunk_sizek; - else - range0_sizek = 0; - - if (!range0_sizek) - break; - } - } - -second_try: - range_basek = range0_basek + range0_sizek; - - /* one hole in the middle */ - if (range_basek > basek && range_basek <= (basek + sizek)) - second_sizek = range_basek - basek; - - if (range0_sizek > state->range_sizek) { - - /* one hole in middle or at end */ - hole_sizek = range0_sizek - state->range_sizek - second_sizek; - - /* hole size should be less than half of range0 size */ - if (hole_sizek >= (range0_sizek >> 1) && - range0_sizek >= chunk_sizek) { - range0_sizek -= chunk_sizek; - second_sizek = 0; - hole_sizek = 0; - - goto second_try; - } - } - - if (range0_sizek) { - if (debug_print) - printk(KERN_DEBUG "range0: %016lx - %016lx\n", - range0_basek<<10, - (range0_basek + range0_sizek)<<10); - state->reg = range_to_mtrr(state->reg, range0_basek, - range0_sizek, MTRR_TYPE_WRBACK); - } - - if (range0_sizek < state->range_sizek) { - /* need to handle left over */ - range_sizek = state->range_sizek - range0_sizek; - - if (debug_print) - printk(KERN_DEBUG "range: %016lx - %016lx\n", - range_basek<<10, - (range_basek + range_sizek)<<10); - state->reg = range_to_mtrr(state->reg, range_basek, - range_sizek, MTRR_TYPE_WRBACK); - } - - if (hole_sizek) { - hole_basek = range_basek - hole_sizek - second_sizek; - if (debug_print) - printk(KERN_DEBUG "hole: %016lx - %016lx\n", - hole_basek<<10, - (hole_basek + hole_sizek)<<10); - state->reg = range_to_mtrr(state->reg, hole_basek, - hole_sizek, MTRR_TYPE_UNCACHABLE); - } - - return second_sizek; -} - -static void __init -set_var_mtrr_range(struct var_mtrr_state *state, unsigned long base_pfn, - unsigned long size_pfn) -{ - unsigned long basek, sizek; - unsigned long second_sizek = 0; - - if (state->reg >= num_var_ranges) - return; - - basek = base_pfn << (PAGE_SHIFT - 10); - sizek = size_pfn << (PAGE_SHIFT - 10); - - /* See if I can merge with the last range */ - if ((basek <= 1024) || - (state->range_startk + state->range_sizek == basek)) { - unsigned long endk = basek + sizek; - state->range_sizek = endk - state->range_startk; - return; - } - /* Write the range mtrrs */ - if (state->range_sizek != 0) - second_sizek = range_to_mtrr_with_hole(state, basek, sizek); - - /* Allocate an msr */ - state->range_startk = basek + second_sizek; - state->range_sizek = sizek - second_sizek; -} - -/* mininum size of mtrr block that can take hole */ -static u64 mtrr_chunk_size __initdata = (256ULL<<20); - -static int __init parse_mtrr_chunk_size_opt(char *p) -{ - if (!p) - return -EINVAL; - mtrr_chunk_size = memparse(p, &p); - return 0; -} -early_param("mtrr_chunk_size", parse_mtrr_chunk_size_opt); - -/* granity of mtrr of block */ -static u64 mtrr_gran_size __initdata; - -static int __init parse_mtrr_gran_size_opt(char *p) -{ - if (!p) - return -EINVAL; - mtrr_gran_size = memparse(p, &p); - return 0; -} -early_param("mtrr_gran_size", parse_mtrr_gran_size_opt); - -static int nr_mtrr_spare_reg __initdata = - CONFIG_MTRR_SANITIZER_SPARE_REG_NR_DEFAULT; - -static int __init parse_mtrr_spare_reg(char *arg) -{ - if (arg) - nr_mtrr_spare_reg = simple_strtoul(arg, NULL, 0); - return 0; -} - -early_param("mtrr_spare_reg_nr", parse_mtrr_spare_reg); - -static int __init -x86_setup_var_mtrrs(struct res_range *range, int nr_range, - u64 chunk_size, u64 gran_size) -{ - struct var_mtrr_state var_state; - int i; - int num_reg; - - var_state.range_startk = 0; - var_state.range_sizek = 0; - var_state.reg = 0; - var_state.chunk_sizek = chunk_size >> 10; - var_state.gran_sizek = gran_size >> 10; - - memset(range_state, 0, sizeof(range_state)); - - /* Write the range etc */ - for (i = 0; i < nr_range; i++) - set_var_mtrr_range(&var_state, range[i].start, - range[i].end - range[i].start + 1); - - /* Write the last range */ - if (var_state.range_sizek != 0) - range_to_mtrr_with_hole(&var_state, 0, 0); - - num_reg = var_state.reg; - /* Clear out the extra MTRR's */ - while (var_state.reg < num_var_ranges) { - save_var_mtrr(var_state.reg, 0, 0, 0); - var_state.reg++; - } - - return num_reg; -} - -struct mtrr_cleanup_result { - unsigned long gran_sizek; - unsigned long chunk_sizek; - unsigned long lose_cover_sizek; - unsigned int num_reg; - int bad; -}; - -/* - * gran_size: 64K, 128K, 256K, 512K, 1M, 2M, ..., 2G - * chunk size: gran_size, ..., 2G - * so we need (1+16)*8 - */ -#define NUM_RESULT 136 -#define PSHIFT (PAGE_SHIFT - 10) - -static struct mtrr_cleanup_result __initdata result[NUM_RESULT]; -static unsigned long __initdata min_loss_pfn[RANGE_NUM]; - -static void __init print_out_mtrr_range_state(void) -{ - int i; - char start_factor = 'K', size_factor = 'K'; - unsigned long start_base, size_base; - mtrr_type type; - - for (i = 0; i < num_var_ranges; i++) { - - size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10); - if (!size_base) - continue; - - size_base = to_size_factor(size_base, &size_factor), - start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10); - start_base = to_size_factor(start_base, &start_factor), - type = range_state[i].type; - - printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n", - i, start_base, start_factor, - size_base, size_factor, - (type == MTRR_TYPE_UNCACHABLE) ? "UC" : - ((type == MTRR_TYPE_WRPROT) ? "WP" : - ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other")) - ); - } -} - -static int __init mtrr_need_cleanup(void) -{ - int i; - mtrr_type type; - unsigned long size; - /* extra one for all 0 */ - int num[MTRR_NUM_TYPES + 1]; - - /* check entries number */ - memset(num, 0, sizeof(num)); - for (i = 0; i < num_var_ranges; i++) { - type = range_state[i].type; - size = range_state[i].size_pfn; - if (type >= MTRR_NUM_TYPES) - continue; - if (!size) - type = MTRR_NUM_TYPES; - if (type == MTRR_TYPE_WRPROT) - type = MTRR_TYPE_UNCACHABLE; - num[type]++; - } - - /* check if we got UC entries */ - if (!num[MTRR_TYPE_UNCACHABLE]) - return 0; - - /* check if we only had WB and UC */ - if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != - num_var_ranges - num[MTRR_NUM_TYPES]) - return 0; - - return 1; -} - -static unsigned long __initdata range_sums; -static void __init mtrr_calc_range_state(u64 chunk_size, u64 gran_size, - unsigned long extra_remove_base, - unsigned long extra_remove_size, - int i) -{ - int num_reg; - static struct res_range range_new[RANGE_NUM]; - static int nr_range_new; - unsigned long range_sums_new; - - /* convert ranges to var ranges state */ - num_reg = x86_setup_var_mtrrs(range, nr_range, - chunk_size, gran_size); - - /* we got new setting in range_state, check it */ - memset(range_new, 0, sizeof(range_new)); - nr_range_new = x86_get_mtrr_mem_range(range_new, 0, - extra_remove_base, extra_remove_size); - range_sums_new = sum_ranges(range_new, nr_range_new); - - result[i].chunk_sizek = chunk_size >> 10; - result[i].gran_sizek = gran_size >> 10; - result[i].num_reg = num_reg; - if (range_sums < range_sums_new) { - result[i].lose_cover_sizek = - (range_sums_new - range_sums) << PSHIFT; - result[i].bad = 1; - } else - result[i].lose_cover_sizek = - (range_sums - range_sums_new) << PSHIFT; - - /* double check it */ - if (!result[i].bad && !result[i].lose_cover_sizek) { - if (nr_range_new != nr_range || - memcmp(range, range_new, sizeof(range))) - result[i].bad = 1; - } - - if (!result[i].bad && (range_sums - range_sums_new < - min_loss_pfn[num_reg])) { - min_loss_pfn[num_reg] = - range_sums - range_sums_new; - } -} - -static void __init mtrr_print_out_one_result(int i) -{ - char gran_factor, chunk_factor, lose_factor; - unsigned long gran_base, chunk_base, lose_base; - - gran_base = to_size_factor(result[i].gran_sizek, &gran_factor), - chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor), - lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor), - printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t", - result[i].bad ? "*BAD*" : " ", - gran_base, gran_factor, chunk_base, chunk_factor); - printk(KERN_CONT "num_reg: %d \tlose cover RAM: %s%ld%c\n", - result[i].num_reg, result[i].bad ? "-" : "", - lose_base, lose_factor); -} - -static int __init mtrr_search_optimal_index(void) -{ - int i; - int num_reg_good; - int index_good; - - if (nr_mtrr_spare_reg >= num_var_ranges) - nr_mtrr_spare_reg = num_var_ranges - 1; - num_reg_good = -1; - for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) { - if (!min_loss_pfn[i]) - num_reg_good = i; - } - - index_good = -1; - if (num_reg_good != -1) { - for (i = 0; i < NUM_RESULT; i++) { - if (!result[i].bad && - result[i].num_reg == num_reg_good && - !result[i].lose_cover_sizek) { - index_good = i; - break; - } - } - } - - return index_good; -} - - -static int __init mtrr_cleanup(unsigned address_bits) -{ - unsigned long extra_remove_base, extra_remove_size; - unsigned long base, size, def, dummy; - mtrr_type type; - u64 chunk_size, gran_size; - int index_good; - int i; - - if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1) - return 0; - rdmsr(MTRRdefType_MSR, def, dummy); - def &= 0xff; - if (def != MTRR_TYPE_UNCACHABLE) - return 0; - - /* get it and store it aside */ - memset(range_state, 0, sizeof(range_state)); - for (i = 0; i < num_var_ranges; i++) { - mtrr_if->get(i, &base, &size, &type); - range_state[i].base_pfn = base; - range_state[i].size_pfn = size; - range_state[i].type = type; - } - - /* check if we need handle it and can handle it */ - if (!mtrr_need_cleanup()) - return 0; - - /* print original var MTRRs at first, for debugging: */ - printk(KERN_DEBUG "original variable MTRRs\n"); - print_out_mtrr_range_state(); - - memset(range, 0, sizeof(range)); - extra_remove_size = 0; - extra_remove_base = 1 << (32 - PAGE_SHIFT); - if (mtrr_tom2) - extra_remove_size = - (mtrr_tom2 >> PAGE_SHIFT) - extra_remove_base; - nr_range = x86_get_mtrr_mem_range(range, 0, extra_remove_base, - extra_remove_size); - /* - * [0, 1M) should always be coverred by var mtrr with WB - * and fixed mtrrs should take effective before var mtrr for it - */ - nr_range = add_range_with_merge(range, nr_range, 0, - (1ULL<<(20 - PAGE_SHIFT)) - 1); - /* sort the ranges */ - sort(range, nr_range, sizeof(struct res_range), cmp_range, NULL); - - range_sums = sum_ranges(range, nr_range); - printk(KERN_INFO "total RAM coverred: %ldM\n", - range_sums >> (20 - PAGE_SHIFT)); - - if (mtrr_chunk_size && mtrr_gran_size) { - i = 0; - mtrr_calc_range_state(mtrr_chunk_size, mtrr_gran_size, - extra_remove_base, extra_remove_size, i); - - mtrr_print_out_one_result(i); - - if (!result[i].bad) { - set_var_mtrr_all(address_bits); - return 1; - } - printk(KERN_INFO "invalid mtrr_gran_size or mtrr_chunk_size, " - "will find optimal one\n"); - } - - i = 0; - memset(min_loss_pfn, 0xff, sizeof(min_loss_pfn)); - memset(result, 0, sizeof(result)); - for (gran_size = (1ULL<<16); gran_size < (1ULL<<32); gran_size <<= 1) { - - for (chunk_size = gran_size; chunk_size < (1ULL<<32); - chunk_size <<= 1) { - - if (i >= NUM_RESULT) - continue; - - mtrr_calc_range_state(chunk_size, gran_size, - extra_remove_base, extra_remove_size, i); - if (debug_print) { - mtrr_print_out_one_result(i); - printk(KERN_INFO "\n"); - } - - i++; - } - } - - /* try to find the optimal index */ - index_good = mtrr_search_optimal_index(); - - if (index_good != -1) { - printk(KERN_INFO "Found optimal setting for mtrr clean up\n"); - i = index_good; - mtrr_print_out_one_result(i); - - /* convert ranges to var ranges state */ - chunk_size = result[i].chunk_sizek; - chunk_size <<= 10; - gran_size = result[i].gran_sizek; - gran_size <<= 10; - x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size); - set_var_mtrr_all(address_bits); - printk(KERN_DEBUG "New variable MTRRs\n"); - print_out_mtrr_range_state(); - return 1; - } else { - /* print out all */ - for (i = 0; i < NUM_RESULT; i++) - mtrr_print_out_one_result(i); - } - - printk(KERN_INFO "mtrr_cleanup: can not find optimal value\n"); - printk(KERN_INFO "please specify mtrr_gran_size/mtrr_chunk_size\n"); - - return 0; -} -#else -static int __init mtrr_cleanup(unsigned address_bits) -{ - return 0; -} -#endif - -static int __initdata changed_by_mtrr_cleanup; - -static int disable_mtrr_trim; - -static int __init disable_mtrr_trim_setup(char *str) -{ - disable_mtrr_trim = 1; - return 0; -} -early_param("disable_mtrr_trim", disable_mtrr_trim_setup); - -/* - * Newer AMD K8s and later CPUs have a special magic MSR way to force WB - * for memory >4GB. Check for that here. - * Note this won't check if the MTRRs < 4GB where the magic bit doesn't - * apply to are wrong, but so far we don't know of any such case in the wild. - */ -#define Tom2Enabled (1U << 21) -#define Tom2ForceMemTypeWB (1U << 22) - -int __init amd_special_default_mtrr(void) -{ - u32 l, h; - - if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) - return 0; - if (boot_cpu_data.x86 < 0xf || boot_cpu_data.x86 > 0x11) - return 0; - /* In case some hypervisor doesn't pass SYSCFG through */ - if (rdmsr_safe(MSR_K8_SYSCFG, &l, &h) < 0) - return 0; - /* - * Memory between 4GB and top of mem is forced WB by this magic bit. - * Reserved before K8RevF, but should be zero there. - */ - if ((l & (Tom2Enabled | Tom2ForceMemTypeWB)) == - (Tom2Enabled | Tom2ForceMemTypeWB)) - return 1; - return 0; -} - -static u64 __init real_trim_memory(unsigned long start_pfn, - unsigned long limit_pfn) -{ - u64 trim_start, trim_size; - trim_start = start_pfn; - trim_start <<= PAGE_SHIFT; - trim_size = limit_pfn; - trim_size <<= PAGE_SHIFT; - trim_size -= trim_start; - - return e820_update_range(trim_start, trim_size, E820_RAM, - E820_RESERVED); -} -/** - * mtrr_trim_uncached_memory - trim RAM not covered by MTRRs - * @end_pfn: ending page frame number - * - * Some buggy BIOSes don't setup the MTRRs properly for systems with certain - * memory configurations. This routine checks that the highest MTRR matches - * the end of memory, to make sure the MTRRs having a write back type cover - * all of the memory the kernel is intending to use. If not, it'll trim any - * memory off the end by adjusting end_pfn, removing it from the kernel's - * allocation pools, warning the user with an obnoxious message. - */ -int __init mtrr_trim_uncached_memory(unsigned long end_pfn) -{ - unsigned long i, base, size, highest_pfn = 0, def, dummy; - mtrr_type type; - u64 total_trim_size; - - /* extra one for all 0 */ - int num[MTRR_NUM_TYPES + 1]; - /* - * Make sure we only trim uncachable memory on machines that - * support the Intel MTRR architecture: - */ - if (!is_cpu(INTEL) || disable_mtrr_trim) - return 0; - rdmsr(MTRRdefType_MSR, def, dummy); - def &= 0xff; - if (def != MTRR_TYPE_UNCACHABLE) - return 0; - - /* get it and store it aside */ - memset(range_state, 0, sizeof(range_state)); - for (i = 0; i < num_var_ranges; i++) { - mtrr_if->get(i, &base, &size, &type); - range_state[i].base_pfn = base; - range_state[i].size_pfn = size; - range_state[i].type = type; - } - - /* Find highest cached pfn */ - for (i = 0; i < num_var_ranges; i++) { - type = range_state[i].type; - if (type != MTRR_TYPE_WRBACK) - continue; - base = range_state[i].base_pfn; - size = range_state[i].size_pfn; - if (highest_pfn < base + size) - highest_pfn = base + size; - } - - /* kvm/qemu doesn't have mtrr set right, don't trim them all */ - if (!highest_pfn) { - printk(KERN_INFO "CPU MTRRs all blank - virtualized system.\n"); - return 0; - } - - /* check entries number */ - memset(num, 0, sizeof(num)); - for (i = 0; i < num_var_ranges; i++) { - type = range_state[i].type; - if (type >= MTRR_NUM_TYPES) - continue; - size = range_state[i].size_pfn; - if (!size) - type = MTRR_NUM_TYPES; - num[type]++; - } - - /* no entry for WB? */ - if (!num[MTRR_TYPE_WRBACK]) - return 0; - - /* check if we only had WB and UC */ - if (num[MTRR_TYPE_WRBACK] + num[MTRR_TYPE_UNCACHABLE] != - num_var_ranges - num[MTRR_NUM_TYPES]) - return 0; - - memset(range, 0, sizeof(range)); - nr_range = 0; - if (mtrr_tom2) { - range[nr_range].start = (1ULL<<(32 - PAGE_SHIFT)); - range[nr_range].end = (mtrr_tom2 >> PAGE_SHIFT) - 1; - if (highest_pfn < range[nr_range].end + 1) - highest_pfn = range[nr_range].end + 1; - nr_range++; - } - nr_range = x86_get_mtrr_mem_range(range, nr_range, 0, 0); - - total_trim_size = 0; - /* check the head */ - if (range[0].start) - total_trim_size += real_trim_memory(0, range[0].start); - /* check the holes */ - for (i = 0; i < nr_range - 1; i++) { - if (range[i].end + 1 < range[i+1].start) - total_trim_size += real_trim_memory(range[i].end + 1, - range[i+1].start); - } - /* check the top */ - i = nr_range - 1; - if (range[i].end + 1 < end_pfn) - total_trim_size += real_trim_memory(range[i].end + 1, - end_pfn); - - if (total_trim_size) { - printk(KERN_WARNING "WARNING: BIOS bug: CPU MTRRs don't cover" - " all of memory, losing %lluMB of RAM.\n", - total_trim_size >> 20); - - if (!changed_by_mtrr_cleanup) - WARN_ON(1); - - printk(KERN_INFO "update e820 for mtrr\n"); - update_e820(); - - return 1; - } - - return 0; -} +int __initdata changed_by_mtrr_cleanup; /** * mtrr_bp_init - initialize mtrrs on the boot CPU diff --git a/arch/x86/kernel/cpu/mtrr/mtrr.h b/arch/x86/kernel/cpu/mtrr/mtrr.h index ffd60409cc6..77f67f7b347 100644 --- a/arch/x86/kernel/cpu/mtrr/mtrr.h +++ b/arch/x86/kernel/cpu/mtrr/mtrr.h @@ -79,6 +79,7 @@ extern struct mtrr_ops * mtrr_if; extern unsigned int num_var_ranges; extern u64 mtrr_tom2; +extern struct mtrr_state_type mtrr_state; void mtrr_state_warn(void); const char *mtrr_attrib_to_str(int x); @@ -88,3 +89,6 @@ void mtrr_wrmsr(unsigned, unsigned, unsigned); int amd_init_mtrr(void); int cyrix_init_mtrr(void); int centaur_init_mtrr(void); + +extern int changed_by_mtrr_cleanup; +extern int mtrr_cleanup(unsigned address_bits); diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 95b81c18b6b..fb638d9ce6d 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -131,6 +131,31 @@ void __init e820_add_region(u64 start, u64 size, int type) __e820_add_region(&e820, start, size, type); } +static void __init e820_print_type(u32 type) +{ + switch (type) { + case E820_RAM: + case E820_RESERVED_KERN: + printk(KERN_CONT "(usable)"); + break; + case E820_RESERVED: + printk(KERN_CONT "(reserved)"); + break; + case E820_ACPI: + printk(KERN_CONT "(ACPI data)"); + break; + case E820_NVS: + printk(KERN_CONT "(ACPI NVS)"); + break; + case E820_UNUSABLE: + printk(KERN_CONT "(unusable)"); + break; + default: + printk(KERN_CONT "type %u", type); + break; + } +} + void __init e820_print_map(char *who) { int i; @@ -140,27 +165,8 @@ void __init e820_print_map(char *who) (unsigned long long) e820.map[i].addr, (unsigned long long) (e820.map[i].addr + e820.map[i].size)); - switch (e820.map[i].type) { - case E820_RAM: - case E820_RESERVED_KERN: - printk(KERN_CONT "(usable)\n"); - break; - case E820_RESERVED: - printk(KERN_CONT "(reserved)\n"); - break; - case E820_ACPI: - printk(KERN_CONT "(ACPI data)\n"); - break; - case E820_NVS: - printk(KERN_CONT "(ACPI NVS)\n"); - break; - case E820_UNUSABLE: - printk("(unusable)\n"); - break; - default: - printk(KERN_CONT "type %u\n", e820.map[i].type); - break; - } + e820_print_type(e820.map[i].type); + printk(KERN_CONT "\n"); } } @@ -427,6 +433,7 @@ static u64 __init __e820_update_range(struct e820map *e820x, u64 start, u64 size, unsigned old_type, unsigned new_type) { + u64 end; unsigned int i; u64 real_updated_size = 0; @@ -435,21 +442,43 @@ static u64 __init __e820_update_range(struct e820map *e820x, u64 start, if (size > (ULLONG_MAX - start)) size = ULLONG_MAX - start; + end = start + size; + printk(KERN_DEBUG "e820 update range: %016Lx - %016Lx ", + (unsigned long long) start, + (unsigned long long) end); + e820_print_type(old_type); + printk(KERN_CONT " ==> "); + e820_print_type(new_type); + printk(KERN_CONT "\n"); + for (i = 0; i < e820x->nr_map; i++) { struct e820entry *ei = &e820x->map[i]; u64 final_start, final_end; + u64 ei_end; + if (ei->type != old_type) continue; - /* totally covered? */ - if (ei->addr >= start && - (ei->addr + ei->size) <= (start + size)) { + + ei_end = ei->addr + ei->size; + /* totally covered by new range? */ + if (ei->addr >= start && ei_end <= end) { ei->type = new_type; real_updated_size += ei->size; continue; } + + /* new range is totally covered? */ + if (ei->addr < start && ei_end > end) { + __e820_add_region(e820x, start, size, new_type); + __e820_add_region(e820x, end, ei_end - end, ei->type); + ei->size = start - ei->addr; + real_updated_size += size; + continue; + } + /* partially covered */ final_start = max(start, ei->addr); - final_end = min(start + size, ei->addr + ei->size); + final_end = min(end, ei_end); if (final_start >= final_end) continue; diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index ac108d1fe18..3f8579f8d42 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -18,7 +18,7 @@ void __init i386_start_kernel(void) { reserve_trampoline_memory(); - reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS"); + reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD /* Reserve INITRD */ @@ -29,9 +29,6 @@ void __init i386_start_kernel(void) reserve_early(ramdisk_image, ramdisk_end, "RAMDISK"); } #endif - reserve_early(init_pg_tables_start, init_pg_tables_end, - "INIT_PG_TABLE"); - reserve_ebda_region(); /* diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index f5b27224769..70eaa852c73 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -100,7 +100,7 @@ void __init x86_64_start_reservations(char *real_mode_data) reserve_trampoline_memory(); - reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS"); + reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD /* Reserve INITRD */ diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S index c32ca19d591..30683883e0c 100644 --- a/arch/x86/kernel/head_32.S +++ b/arch/x86/kernel/head_32.S @@ -38,42 +38,40 @@ #define X86_VENDOR_ID new_cpu_data+CPUINFO_x86_vendor_id /* - * This is how much memory *in addition to the memory covered up to - * and including _end* we need mapped initially. + * This is how much memory in addition to the memory covered up to + * and including _end we need mapped initially. * We need: - * - one bit for each possible page, but only in low memory, which means - * 2^32/4096/8 = 128K worst case (4G/4G split.) - * - enough space to map all low memory, which means - * (2^32/4096) / 1024 pages (worst case, non PAE) - * (2^32/4096) / 512 + 4 pages (worst case for PAE) - * - a few pages for allocator use before the kernel pagetable has - * been set up + * (KERNEL_IMAGE_SIZE/4096) / 1024 pages (worst case, non PAE) + * (KERNEL_IMAGE_SIZE/4096) / 512 + 4 pages (worst case for PAE) * * Modulo rounding, each megabyte assigned here requires a kilobyte of * memory, which is currently unreclaimed. * * This should be a multiple of a page. + * + * KERNEL_IMAGE_SIZE should be greater than pa(_end) + * and small than max_low_pfn, otherwise will waste some page table entries */ -LOW_PAGES = 1<<(32-PAGE_SHIFT_asm) - -/* - * To preserve the DMA pool in PAGEALLOC kernels, we'll allocate - * pagetables from above the 16MB DMA limit, so we'll have to set - * up pagetables 16MB more (worst-case): - */ -#ifdef CONFIG_DEBUG_PAGEALLOC -LOW_PAGES = LOW_PAGES + 0x1000000 -#endif #if PTRS_PER_PMD > 1 -PAGE_TABLE_SIZE = (LOW_PAGES / PTRS_PER_PMD) + PTRS_PER_PGD +#define PAGE_TABLE_SIZE(pages) (((pages) / PTRS_PER_PMD) + PTRS_PER_PGD) #else -PAGE_TABLE_SIZE = (LOW_PAGES / PTRS_PER_PGD) +#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PGD) #endif -BOOTBITMAP_SIZE = LOW_PAGES / 8 -ALLOCATOR_SLOP = 4 -INIT_MAP_BEYOND_END = BOOTBITMAP_SIZE + (PAGE_TABLE_SIZE + ALLOCATOR_SLOP)*PAGE_SIZE_asm +/* Enough space to fit pagetables for the low memory linear map */ +MAPPING_BEYOND_END = \ + PAGE_TABLE_SIZE(((1<<32) - __PAGE_OFFSET) >> PAGE_SHIFT) << PAGE_SHIFT + +/* + * Worst-case size of the kernel mapping we need to make: + * the worst-case size of the kernel itself, plus the extra we need + * to map for the linear map. + */ +KERNEL_PAGES = (KERNEL_IMAGE_SIZE + MAPPING_BEYOND_END)>>PAGE_SHIFT + +INIT_MAP_SIZE = PAGE_TABLE_SIZE(KERNEL_PAGES) * PAGE_SIZE_asm +RESERVE_BRK(pagetables, INIT_MAP_SIZE) /* * 32-bit kernel entrypoint; only used by the boot CPU. On entry, @@ -166,10 +164,10 @@ num_subarch_entries = (. - subarch_entries) / 4 /* * Initialize page tables. This creates a PDE and a set of page - * tables, which are located immediately beyond _end. The variable - * init_pg_tables_end is set up to point to the first "safe" location. + * tables, which are located immediately beyond __brk_base. The variable + * _brk_end is set up to point to the first "safe" location. * Mappings are created both at virtual address 0 (identity mapping) - * and PAGE_OFFSET for up to _end+sizeof(page tables)+INIT_MAP_BEYOND_END. + * and PAGE_OFFSET for up to _end. * * Note that the stack is not yet set up! */ @@ -190,8 +188,7 @@ default_entry: xorl %ebx,%ebx /* %ebx is kept at zero */ - movl $pa(pg0), %edi - movl %edi, pa(init_pg_tables_start) + movl $pa(__brk_base), %edi movl $pa(swapper_pg_pmd), %edx movl $PTE_IDENT_ATTR, %eax 10: @@ -209,14 +206,14 @@ default_entry: loop 11b /* - * End condition: we must map up to and including INIT_MAP_BEYOND_END - * bytes beyond the end of our own page tables. + * End condition: we must map up to the end + MAPPING_BEYOND_END. */ - leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp + movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp cmpl %ebp,%eax jb 10b 1: - movl %edi,pa(init_pg_tables_end) + addl $__PAGE_OFFSET, %edi + movl %edi, pa(_brk_end) shrl $12, %eax movl %eax, pa(max_pfn_mapped) @@ -227,8 +224,7 @@ default_entry: page_pde_offset = (__PAGE_OFFSET >> 20); - movl $pa(pg0), %edi - movl %edi, pa(init_pg_tables_start) + movl $pa(__brk_base), %edi movl $pa(swapper_pg_dir), %edx movl $PTE_IDENT_ATTR, %eax 10: @@ -242,14 +238,13 @@ page_pde_offset = (__PAGE_OFFSET >> 20); addl $0x1000,%eax loop 11b /* - * End condition: we must map up to and including INIT_MAP_BEYOND_END - * bytes beyond the end of our own page tables; the +0x007 is - * the attribute bits + * End condition: we must map up to the end + MAPPING_BEYOND_END. */ - leal (INIT_MAP_BEYOND_END+PTE_IDENT_ATTR)(%edi),%ebp + movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp cmpl %ebp,%eax jb 10b - movl %edi,pa(init_pg_tables_end) + addl $__PAGE_OFFSET, %edi + movl %edi, pa(_brk_end) shrl $12, %eax movl %eax, pa(max_pfn_mapped) @@ -636,6 +631,7 @@ swapper_pg_fixmap: .fill 1024,4,0 ENTRY(empty_zero_page) .fill 4096,1,0 + /* * This starts the data section. */ diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index e948b28a5a9..55b94614e34 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -193,6 +193,9 @@ static int __kprobes can_boost(kprobe_opcode_t *opcodes) kprobe_opcode_t opcode; kprobe_opcode_t *orig_opcodes = opcodes; + if (search_exception_tables((unsigned long)opcodes)) + return 0; /* Page fault may occur on this address. */ + retry: if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1) return 0; diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c index 478bca986ec..33019ddb56b 100644 --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -138,12 +138,6 @@ static void kvm_set_pte_atomic(pte_t *ptep, pte_t pte) kvm_mmu_write(ptep, pte_val(pte)); } -static void kvm_set_pte_present(struct mm_struct *mm, unsigned long addr, - pte_t *ptep, pte_t pte) -{ - kvm_mmu_write(ptep, pte_val(pte)); -} - static void kvm_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { @@ -220,7 +214,6 @@ static void paravirt_ops_setup(void) #if PAGETABLE_LEVELS >= 3 #ifdef CONFIG_X86_PAE pv_mmu_ops.set_pte_atomic = kvm_set_pte_atomic; - pv_mmu_ops.set_pte_present = kvm_set_pte_present; pv_mmu_ops.pte_clear = kvm_pte_clear; pv_mmu_ops.pmd_clear = kvm_pmd_clear; #endif diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index 47673e02ae5..290cb57f469 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -109,9 +109,6 @@ static void __init MP_bus_info(struct mpc_bus *m) } else printk(KERN_WARNING "Unknown bustype %s - ignoring\n", str); } -#endif - -#ifdef CONFIG_X86_IO_APIC static int bad_ioapic(unsigned long address) { @@ -224,8 +221,12 @@ static void __init MP_intsrc_info(struct mpc_intsrc *m) if (++mp_irq_entries == MAX_IRQ_SOURCES) panic("Max # of irq sources exceeded!!\n"); } +#else /* CONFIG_X86_IO_APIC */ +static inline void __init MP_bus_info(struct mpc_bus *m) {} +static inline void __init MP_ioapic_info(struct mpc_ioapic *m) {} +static inline void __init MP_intsrc_info(struct mpc_intsrc *m) {} +#endif /* CONFIG_X86_IO_APIC */ -#endif static void __init MP_lintsrc_info(struct mpc_lintsrc *m) { @@ -275,6 +276,12 @@ static int __init smp_check_mpc(struct mpc_table *mpc, char *oem, char *str) return 1; } +static void skip_entry(unsigned char **ptr, int *count, int size) +{ + *ptr += size; + *count += size; +} + static int __init smp_read_mpc(struct mpc_table *mpc, unsigned early) { char str[16]; @@ -310,55 +317,27 @@ static int __init smp_read_mpc(struct mpc_table *mpc, unsigned early) while (count < mpc->length) { switch (*mpt) { case MP_PROCESSOR: - { - struct mpc_cpu *m = (struct mpc_cpu *)mpt; - /* ACPI may have already provided this data */ - if (!acpi_lapic) - MP_processor_info(m); - mpt += sizeof(*m); - count += sizeof(*m); - break; - } + /* ACPI may have already provided this data */ + if (!acpi_lapic) + MP_processor_info((struct mpc_cpu *)mpt); + skip_entry(&mpt, &count, sizeof(struct mpc_cpu)); + break; case MP_BUS: - { - struct mpc_bus *m = (struct mpc_bus *)mpt; -#ifdef CONFIG_X86_IO_APIC - MP_bus_info(m); -#endif - mpt += sizeof(*m); - count += sizeof(*m); - break; - } + MP_bus_info((struct mpc_bus *)mpt); + skip_entry(&mpt, &count, sizeof(struct mpc_bus)); + break; case MP_IOAPIC: - { -#ifdef CONFIG_X86_IO_APIC - struct mpc_ioapic *m = (struct mpc_ioapic *)mpt; - MP_ioapic_info(m); -#endif - mpt += sizeof(struct mpc_ioapic); - count += sizeof(struct mpc_ioapic); - break; - } + MP_ioapic_info((struct mpc_ioapic *)mpt); + skip_entry(&mpt, &count, sizeof(struct mpc_ioapic)); + break; case MP_INTSRC: - { -#ifdef CONFIG_X86_IO_APIC - struct mpc_intsrc *m = (struct mpc_intsrc *)mpt; - - MP_intsrc_info(m); -#endif - mpt += sizeof(struct mpc_intsrc); - count += sizeof(struct mpc_intsrc); - break; - } + MP_intsrc_info((struct mpc_intsrc *)mpt); + skip_entry(&mpt, &count, sizeof(struct mpc_intsrc)); + break; case MP_LINTSRC: - { - struct mpc_lintsrc *m = - (struct mpc_lintsrc *)mpt; - MP_lintsrc_info(m); - mpt += sizeof(*m); - count += sizeof(*m); - break; - } + MP_lintsrc_info((struct mpc_lintsrc *)mpt); + skip_entry(&mpt, &count, sizeof(struct mpc_lintsrc)); + break; default: /* wrong mptable */ printk(KERN_ERR "Your mptable is wrong, contact your HW vendor!\n"); @@ -689,6 +668,31 @@ void __init get_smp_config(void) __get_smp_config(0); } +static void smp_reserve_bootmem(struct mpf_intel *mpf) +{ + unsigned long size = get_mpc_size(mpf->physptr); +#ifdef CONFIG_X86_32 + /* + * We cannot access to MPC table to compute table size yet, + * as only few megabytes from the bottom is mapped now. + * PC-9800's MPC table places on the very last of physical + * memory; so that simply reserving PAGE_SIZE from mpf->physptr + * yields BUG() in reserve_bootmem. + * also need to make sure physptr is below than max_low_pfn + * we don't need reserve the area above max_low_pfn + */ + unsigned long end = max_low_pfn * PAGE_SIZE; + + if (mpf->physptr < end) { + if (mpf->physptr + size > end) + size = end - mpf->physptr; + reserve_bootmem_generic(mpf->physptr, size, BOOTMEM_DEFAULT); + } +#else + reserve_bootmem_generic(mpf->physptr, size, BOOTMEM_DEFAULT); +#endif +} + static int __init smp_scan_config(unsigned long base, unsigned long length, unsigned reserve) { @@ -717,35 +721,9 @@ static int __init smp_scan_config(unsigned long base, unsigned long length, if (!reserve) return 1; reserve_bootmem_generic(virt_to_phys(mpf), sizeof(*mpf), - BOOTMEM_DEFAULT); - if (mpf->physptr) { - unsigned long size = get_mpc_size(mpf->physptr); -#ifdef CONFIG_X86_32 - /* - * We cannot access to MPC table to compute - * table size yet, as only few megabytes from - * the bottom is mapped now. - * PC-9800's MPC table places on the very last - * of physical memory; so that simply reserving - * PAGE_SIZE from mpf->physptr yields BUG() - * in reserve_bootmem. - * also need to make sure physptr is below than - * max_low_pfn - * we don't need reserve the area above max_low_pfn - */ - unsigned long end = max_low_pfn * PAGE_SIZE; - - if (mpf->physptr < end) { - if (mpf->physptr + size > end) - size = end - mpf->physptr; - reserve_bootmem_generic(mpf->physptr, size, - BOOTMEM_DEFAULT); - } -#else - reserve_bootmem_generic(mpf->physptr, size, BOOTMEM_DEFAULT); -#endif - } + if (mpf->physptr) + smp_reserve_bootmem(mpf); return 1; } @@ -848,7 +826,57 @@ static int __init get_MP_intsrc_index(struct mpc_intsrc *m) #define SPARE_SLOT_NUM 20 static struct mpc_intsrc __initdata *m_spare[SPARE_SLOT_NUM]; -#endif + +static void check_irq_src(struct mpc_intsrc *m, int *nr_m_spare) +{ + int i; + + apic_printk(APIC_VERBOSE, "OLD "); + print_MP_intsrc_info(m); + + i = get_MP_intsrc_index(m); + if (i > 0) { + assign_to_mpc_intsrc(&mp_irqs[i], m); + apic_printk(APIC_VERBOSE, "NEW "); + print_mp_irq_info(&mp_irqs[i]); + return; + } + if (!i) { + /* legacy, do nothing */ + return; + } + if (*nr_m_spare < SPARE_SLOT_NUM) { + /* + * not found (-1), or duplicated (-2) are invalid entries, + * we need to use the slot later + */ + m_spare[*nr_m_spare] = m; + *nr_m_spare += 1; + } +} +#else /* CONFIG_X86_IO_APIC */ +static inline void check_irq_src(struct mpc_intsrc *m, int *nr_m_spare) {} +#endif /* CONFIG_X86_IO_APIC */ + +static int check_slot(unsigned long mpc_new_phys, unsigned long mpc_new_length, + int count) +{ + if (!mpc_new_phys) { + pr_info("No spare slots, try to append...take your risk, " + "new mpc_length %x\n", count); + } else { + if (count <= mpc_new_length) + pr_info("No spare slots, try to append..., " + "new mpc_length %x\n", count); + else { + pr_err("mpc_new_length %lx is too small\n", + mpc_new_length); + return -1; + } + } + + return 0; +} static int __init replace_intsrc_all(struct mpc_table *mpc, unsigned long mpc_new_phys, @@ -856,71 +884,30 @@ static int __init replace_intsrc_all(struct mpc_table *mpc, { #ifdef CONFIG_X86_IO_APIC int i; - int nr_m_spare = 0; #endif - int count = sizeof(*mpc); + int nr_m_spare = 0; unsigned char *mpt = ((unsigned char *)mpc) + count; printk(KERN_INFO "mpc_length %x\n", mpc->length); while (count < mpc->length) { switch (*mpt) { case MP_PROCESSOR: - { - struct mpc_cpu *m = (struct mpc_cpu *)mpt; - mpt += sizeof(*m); - count += sizeof(*m); - break; - } + skip_entry(&mpt, &count, sizeof(struct mpc_cpu)); + break; case MP_BUS: - { - struct mpc_bus *m = (struct mpc_bus *)mpt; - mpt += sizeof(*m); - count += sizeof(*m); - break; - } + skip_entry(&mpt, &count, sizeof(struct mpc_bus)); + break; case MP_IOAPIC: - { - mpt += sizeof(struct mpc_ioapic); - count += sizeof(struct mpc_ioapic); - break; - } + skip_entry(&mpt, &count, sizeof(struct mpc_ioapic)); + break; case MP_INTSRC: - { -#ifdef CONFIG_X86_IO_APIC - struct mpc_intsrc *m = (struct mpc_intsrc *)mpt; - - apic_printk(APIC_VERBOSE, "OLD "); - print_MP_intsrc_info(m); - i = get_MP_intsrc_index(m); - if (i > 0) { - assign_to_mpc_intsrc(&mp_irqs[i], m); - apic_printk(APIC_VERBOSE, "NEW "); - print_mp_irq_info(&mp_irqs[i]); - } else if (!i) { - /* legacy, do nothing */ - } else if (nr_m_spare < SPARE_SLOT_NUM) { - /* - * not found (-1), or duplicated (-2) - * are invalid entries, - * we need to use the slot later - */ - m_spare[nr_m_spare] = m; - nr_m_spare++; - } -#endif - mpt += sizeof(struct mpc_intsrc); - count += sizeof(struct mpc_intsrc); - break; - } + check_irq_src((struct mpc_intsrc *)mpt, &nr_m_spare); + skip_entry(&mpt, &count, sizeof(struct mpc_intsrc)); + break; case MP_LINTSRC: - { - struct mpc_lintsrc *m = - (struct mpc_lintsrc *)mpt; - mpt += sizeof(*m); - count += sizeof(*m); - break; - } + skip_entry(&mpt, &count, sizeof(struct mpc_lintsrc)); + break; default: /* wrong mptable */ printk(KERN_ERR "Your mptable is wrong, contact your HW vendor!\n"); @@ -950,16 +937,8 @@ static int __init replace_intsrc_all(struct mpc_table *mpc, } else { struct mpc_intsrc *m = (struct mpc_intsrc *)mpt; count += sizeof(struct mpc_intsrc); - if (!mpc_new_phys) { - printk(KERN_INFO "No spare slots, try to append...take your risk, new mpc_length %x\n", count); - } else { - if (count <= mpc_new_length) - printk(KERN_INFO "No spare slots, try to append..., new mpc_length %x\n", count); - else { - printk(KERN_ERR "mpc_new_length %lx is too small\n", mpc_new_length); - goto out; - } - } + if (!check_slot(mpc_new_phys, mpc_new_length, count)) + goto out; assign_to_mpc_intsrc(&mp_irqs[i], m); mpc->length = count; mpt += sizeof(struct mpc_intsrc); diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index 63dd358d8ee..8e45f446488 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -470,7 +470,6 @@ struct pv_mmu_ops pv_mmu_ops = { #if PAGETABLE_LEVELS >= 3 #ifdef CONFIG_X86_PAE .set_pte_atomic = native_set_pte_atomic, - .set_pte_present = native_set_pte_present, .pte_clear = native_pte_clear, .pmd_clear = native_pmd_clear, #endif diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 6afa5232dbb..156f87582c6 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -65,11 +65,11 @@ void exit_thread(void) { struct task_struct *me = current; struct thread_struct *t = &me->thread; + unsigned long *bp = t->io_bitmap_ptr; - if (me->thread.io_bitmap_ptr) { + if (bp) { struct tss_struct *tss = &per_cpu(init_tss, get_cpu()); - kfree(t->io_bitmap_ptr); t->io_bitmap_ptr = NULL; clear_thread_flag(TIF_IO_BITMAP); /* @@ -78,6 +78,7 @@ void exit_thread(void) memset(tss->io_bitmap, 0xff, t->io_bitmap_max); t->io_bitmap_max = 0; put_cpu(); + kfree(bp); } ds_exit_thread(current); diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index f28c56e6bf9..a0d26237d7c 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -112,8 +112,13 @@ #define ARCH_SETUP #endif +RESERVE_BRK(dmi_alloc, 65536); + unsigned int boot_cpu_id __read_mostly; +static __initdata unsigned long _brk_start = (unsigned long)__brk_base; +unsigned long _brk_end = (unsigned long)__brk_base; + #ifdef CONFIG_X86_64 int default_cpu_present_to_apicid(int mps_cpu) { @@ -158,12 +163,6 @@ static struct resource bss_resource = { #ifdef CONFIG_X86_32 -/* This value is set up by the early boot code to point to the value - immediately after the boot time page tables. It contains a *physical* - address, and must not be in the .bss segment! */ -unsigned long init_pg_tables_start __initdata = ~0UL; -unsigned long init_pg_tables_end __initdata = ~0UL; - static struct resource video_ram_resource = { .name = "Video RAM area", .start = 0xa0000, @@ -219,12 +218,6 @@ unsigned long mmu_cr4_features = X86_CR4_PAE; int bootloader_type; /* - * Early DMI memory - */ -int dmi_alloc_index; -char dmi_alloc_data[DMI_MAX_DATA]; - -/* * Setup options */ struct screen_info screen_info; @@ -269,6 +262,35 @@ static inline void copy_edd(void) } #endif +void * __init extend_brk(size_t size, size_t align) +{ + size_t mask = align - 1; + void *ret; + + BUG_ON(_brk_start == 0); + BUG_ON(align & mask); + + _brk_end = (_brk_end + mask) & ~mask; + BUG_ON((char *)(_brk_end + size) > __brk_limit); + + ret = (void *)_brk_end; + _brk_end += size; + + memset(ret, 0, size); + + return ret; +} + +static void __init reserve_brk(void) +{ + if (_brk_end > _brk_start) + reserve_early(__pa(_brk_start), __pa(_brk_end), "BRK"); + + /* Mark brk area as locked down and no longer taking any + new allocations */ + _brk_start = 0; +} + #ifdef CONFIG_BLK_DEV_INITRD #ifdef CONFIG_X86_32 @@ -717,11 +739,7 @@ void __init setup_arch(char **cmdline_p) init_mm.start_code = (unsigned long) _text; init_mm.end_code = (unsigned long) _etext; init_mm.end_data = (unsigned long) _edata; -#ifdef CONFIG_X86_32 - init_mm.brk = init_pg_tables_end + PAGE_OFFSET; -#else - init_mm.brk = (unsigned long) &_end; -#endif + init_mm.brk = _brk_end; code_resource.start = virt_to_phys(_text); code_resource.end = virt_to_phys(_etext)-1; @@ -842,6 +860,8 @@ void __init setup_arch(char **cmdline_p) setup_bios_corruption_check(); #endif + reserve_brk(); + /* max_pfn_mapped is updated here */ max_low_pfn_mapped = init_memory_mapping(0, max_low_pfn<<PAGE_SHIFT); max_pfn_mapped = max_low_pfn_mapped; diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index d038b9c45cf..79c07324728 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -750,7 +750,7 @@ static int __init uv_bau_init(void) int node; int nblades; int last_blade; - int cur_cpu = 0; + int cur_cpu; if (!is_uv_system()) return 0; @@ -760,6 +760,7 @@ static int __init uv_bau_init(void) uv_mmask = (1UL << uv_hub_info->n_val) - 1; nblades = 0; last_blade = -1; + cur_cpu = 0; for_each_online_node(node) { blade = uv_node_to_blade_id(node); if (blade == last_blade) diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 83d53ce5d4c..462b9ba67e9 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -273,30 +273,43 @@ static unsigned long pit_calibrate_tsc(u32 latch, unsigned long ms, int loopmin) * use the TSC value at the transitions to calculate a pretty * good value for the TSC frequencty. */ -static inline int pit_expect_msb(unsigned char val) +static inline int pit_expect_msb(unsigned char val, u64 *tscp, unsigned long *deltap) { - int count = 0; + int count; + u64 tsc = 0; for (count = 0; count < 50000; count++) { /* Ignore LSB */ inb(0x42); if (inb(0x42) != val) break; + tsc = get_cycles(); } - return count > 50; + *deltap = get_cycles() - tsc; + *tscp = tsc; + + /* + * We require _some_ success, but the quality control + * will be based on the error terms on the TSC values. + */ + return count > 5; } /* - * How many MSB values do we want to see? We aim for a - * 15ms calibration, which assuming a 2us counter read - * error should give us roughly 150 ppm precision for - * the calibration. + * How many MSB values do we want to see? We aim for + * a maximum error rate of 500ppm (in practice the + * real error is much smaller), but refuse to spend + * more than 25ms on it. */ -#define QUICK_PIT_MS 15 -#define QUICK_PIT_ITERATIONS (QUICK_PIT_MS * PIT_TICK_RATE / 1000 / 256) +#define MAX_QUICK_PIT_MS 25 +#define MAX_QUICK_PIT_ITERATIONS (MAX_QUICK_PIT_MS * PIT_TICK_RATE / 1000 / 256) static unsigned long quick_pit_calibrate(void) { + int i; + u64 tsc, delta; + unsigned long d1, d2; + /* Set the Gate high, disable speaker */ outb((inb(0x61) & ~0x02) | 0x01, 0x61); @@ -315,45 +328,52 @@ static unsigned long quick_pit_calibrate(void) outb(0xff, 0x42); outb(0xff, 0x42); - if (pit_expect_msb(0xff)) { - int i; - u64 t1, t2, delta; - unsigned char expect = 0xfe; - - t1 = get_cycles(); - for (i = 0; i < QUICK_PIT_ITERATIONS; i++, expect--) { - if (!pit_expect_msb(expect)) - goto failed; + /* + * The PIT starts counting at the next edge, so we + * need to delay for a microsecond. The easiest way + * to do that is to just read back the 16-bit counter + * once from the PIT. + */ + inb(0x42); + inb(0x42); + + if (pit_expect_msb(0xff, &tsc, &d1)) { + for (i = 1; i <= MAX_QUICK_PIT_ITERATIONS; i++) { + if (!pit_expect_msb(0xff-i, &delta, &d2)) + break; + + /* + * Iterate until the error is less than 500 ppm + */ + delta -= tsc; + if (d1+d2 < delta >> 11) + goto success; } - t2 = get_cycles(); - - /* - * Make sure we can rely on the second TSC timestamp: - */ - if (!pit_expect_msb(expect)) - goto failed; - - /* - * Ok, if we get here, then we've seen the - * MSB of the PIT decrement QUICK_PIT_ITERATIONS - * times, and each MSB had many hits, so we never - * had any sudden jumps. - * - * As a result, we can depend on there not being - * any odd delays anywhere, and the TSC reads are - * reliable. - * - * kHz = ticks / time-in-seconds / 1000; - * kHz = (t2 - t1) / (QPI * 256 / PIT_TICK_RATE) / 1000 - * kHz = ((t2 - t1) * PIT_TICK_RATE) / (QPI * 256 * 1000) - */ - delta = (t2 - t1)*PIT_TICK_RATE; - do_div(delta, QUICK_PIT_ITERATIONS*256*1000); - printk("Fast TSC calibration using PIT\n"); - return delta; } -failed: + printk("Fast TSC calibration failed\n"); return 0; + +success: + /* + * Ok, if we get here, then we've seen the + * MSB of the PIT decrement 'i' times, and the + * error has shrunk to less than 500 ppm. + * + * As a result, we can depend on there not being + * any odd delays anywhere, and the TSC reads are + * reliable (within the error). We also adjust the + * delta to the middle of the error bars, just + * because it looks nicer. + * + * kHz = ticks / time-in-seconds / 1000; + * kHz = (t2 - t1) / (I * 256 / PIT_TICK_RATE) / 1000 + * kHz = ((t2 - t1) * PIT_TICK_RATE) / (I * 256 * 1000) + */ + delta += (long)(d2 - d1)/2; + delta *= PIT_TICK_RATE; + do_div(delta, i*256*1000); + printk("Fast TSC calibration using PIT\n"); + return delta; } /** diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index 2cc4a90e2cb..95deb9f2211 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -395,11 +395,6 @@ static void vmi_set_pte_atomic(pte_t *ptep, pte_t pteval) vmi_ops.update_pte(ptep, VMI_PAGE_PT); } -static void vmi_set_pte_present(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte) -{ - vmi_ops.set_pte(pte, ptep, vmi_flags_addr_defer(mm, addr, VMI_PAGE_PT, 1)); -} - static void vmi_set_pud(pud_t *pudp, pud_t pudval) { /* Um, eww */ @@ -750,7 +745,6 @@ static inline int __init activate_vmi(void) pv_mmu_ops.set_pmd = vmi_set_pmd; #ifdef CONFIG_X86_PAE pv_mmu_ops.set_pte_atomic = vmi_set_pte_atomic; - pv_mmu_ops.set_pte_present = vmi_set_pte_present; pv_mmu_ops.set_pud = vmi_set_pud; pv_mmu_ops.pte_clear = vmi_pte_clear; pv_mmu_ops.pmd_clear = vmi_pmd_clear; diff --git a/arch/x86/kernel/vmlinux_32.lds.S b/arch/x86/kernel/vmlinux_32.lds.S index 0d860963f26..62ad500d55f 100644 --- a/arch/x86/kernel/vmlinux_32.lds.S +++ b/arch/x86/kernel/vmlinux_32.lds.S @@ -189,15 +189,24 @@ SECTIONS *(.bss) . = ALIGN(4); __bss_stop = .; - _end = . ; - /* This is where the kernel creates the early boot page tables */ + } + + .brk : AT(ADDR(.brk) - LOAD_OFFSET) { . = ALIGN(PAGE_SIZE); - pg0 = . ; + __brk_base = . ; + . += 64 * 1024 ; /* 64k alignment slop space */ + *(.brk_reservation) /* areas brk users have reserved */ + __brk_limit = . ; + } + + .end : AT(ADDR(.end) - LOAD_OFFSET) { + _end = . ; } /* Sections to be discarded */ /DISCARD/ : { *(.exitcall.exit) + *(.discard) } STABS_DEBUG @@ -205,6 +214,12 @@ SECTIONS DWARF_DEBUG } +/* + * Build-time check on the image size: + */ +ASSERT((_end - LOAD_OFFSET <= KERNEL_IMAGE_SIZE), + "kernel image bigger than KERNEL_IMAGE_SIZE") + #ifdef CONFIG_KEXEC /* Link time checks */ #include <asm/kexec.h> diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index 5bf54e40c6e..c8742507b03 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -29,8 +29,8 @@ SECTIONS { . = __START_KERNEL; phys_startup_64 = startup_64 - LOAD_OFFSET; - _text = .; /* Text and read-only data */ .text : AT(ADDR(.text) - LOAD_OFFSET) { + _text = .; /* Text and read-only data */ /* First the code that has to be first for bootstrapping */ *(.text.head) _stext = .; @@ -61,13 +61,13 @@ SECTIONS .data : AT(ADDR(.data) - LOAD_OFFSET) { DATA_DATA CONSTRUCTORS + _edata = .; /* End of data section */ } :data - _edata = .; /* End of data section */ - . = ALIGN(PAGE_SIZE); - . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); .data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) { + . = ALIGN(PAGE_SIZE); + . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); *(.data.cacheline_aligned) } . = ALIGN(CONFIG_X86_INTERNODE_CACHE_BYTES); @@ -125,29 +125,29 @@ SECTIONS #undef VVIRT_OFFSET #undef VVIRT - . = ALIGN(THREAD_SIZE); /* init_task */ .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) { + . = ALIGN(THREAD_SIZE); /* init_task */ *(.data.init_task) }:data.init - . = ALIGN(PAGE_SIZE); .data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) { + . = ALIGN(PAGE_SIZE); *(.data.page_aligned) } - /* might get freed after init */ - . = ALIGN(PAGE_SIZE); - __smp_alt_begin = .; - __smp_locks = .; .smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) { + /* might get freed after init */ + . = ALIGN(PAGE_SIZE); + __smp_alt_begin = .; + __smp_locks = .; *(.smp_locks) + __smp_locks_end = .; + . = ALIGN(PAGE_SIZE); + __smp_alt_end = .; } - __smp_locks_end = .; - . = ALIGN(PAGE_SIZE); - __smp_alt_end = .; . = ALIGN(PAGE_SIZE); /* Init code and data */ - __init_begin = .; + __init_begin = .; /* paired with __init_end */ .init.text : AT(ADDR(.init.text) - LOAD_OFFSET) { _sinittext = .; INIT_TEXT @@ -159,40 +159,42 @@ SECTIONS __initdata_end = .; } - . = ALIGN(16); - __setup_start = .; - .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) { *(.init.setup) } - __setup_end = .; - __initcall_start = .; + .init.setup : AT(ADDR(.init.setup) - LOAD_OFFSET) { + . = ALIGN(16); + __setup_start = .; + *(.init.setup) + __setup_end = .; + } .initcall.init : AT(ADDR(.initcall.init) - LOAD_OFFSET) { + __initcall_start = .; INITCALLS + __initcall_end = .; } - __initcall_end = .; - __con_initcall_start = .; .con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET) { + __con_initcall_start = .; *(.con_initcall.init) + __con_initcall_end = .; } - __con_initcall_end = .; - __x86_cpu_dev_start = .; .x86_cpu_dev.init : AT(ADDR(.x86_cpu_dev.init) - LOAD_OFFSET) { + __x86_cpu_dev_start = .; *(.x86_cpu_dev.init) + __x86_cpu_dev_end = .; } - __x86_cpu_dev_end = .; SECURITY_INIT . = ALIGN(8); .parainstructions : AT(ADDR(.parainstructions) - LOAD_OFFSET) { - __parainstructions = .; + __parainstructions = .; *(.parainstructions) - __parainstructions_end = .; + __parainstructions_end = .; } - . = ALIGN(8); - __alt_instructions = .; .altinstructions : AT(ADDR(.altinstructions) - LOAD_OFFSET) { + . = ALIGN(8); + __alt_instructions = .; *(.altinstructions) + __alt_instructions_end = .; } - __alt_instructions_end = .; .altinstr_replacement : AT(ADDR(.altinstr_replacement) - LOAD_OFFSET) { *(.altinstr_replacement) } @@ -207,9 +209,11 @@ SECTIONS #ifdef CONFIG_BLK_DEV_INITRD . = ALIGN(PAGE_SIZE); - __initramfs_start = .; - .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { *(.init.ramfs) } - __initramfs_end = .; + .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { + __initramfs_start = .; + *(.init.ramfs) + __initramfs_end = .; + } #endif #ifdef CONFIG_SMP @@ -229,20 +233,29 @@ SECTIONS . = ALIGN(PAGE_SIZE); __init_end = .; - . = ALIGN(PAGE_SIZE); - __nosave_begin = .; .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { - *(.data.nosave) + . = ALIGN(PAGE_SIZE); + __nosave_begin = .; + *(.data.nosave) + . = ALIGN(PAGE_SIZE); + __nosave_end = .; } :data.init2 /* use another section data.init2, see PERCPU_VADDR() above */ - . = ALIGN(PAGE_SIZE); - __nosave_end = .; - __bss_start = .; /* BSS */ .bss : AT(ADDR(.bss) - LOAD_OFFSET) { + . = ALIGN(PAGE_SIZE); + __bss_start = .; /* BSS */ *(.bss.page_aligned) *(.bss) - } - __bss_stop = .; + __bss_stop = .; + } + + .brk : AT(ADDR(.brk) - LOAD_OFFSET) { + . = ALIGN(PAGE_SIZE); + __brk_base = . ; + . += 64 * 1024 ; /* 64k alignment slop space */ + *(.brk_reservation) /* areas brk users have reserved */ + __brk_limit = . ; + } _end = . ; @@ -250,6 +263,7 @@ SECTIONS /DISCARD/ : { *(.exitcall.exit) *(.eh_frame) + *(.discard) } STABS_DEBUG |