From c1726d6f1ad2f1d83e5db1e0142756e9255a82b3 Mon Sep 17 00:00:00 2001 From: Yasuaki Ishimatsu Date: Tue, 17 Jul 2007 21:21:26 +0900 Subject: [IA64] Use per iosapic lock for indirect iosapic register access Use per-iosapic lock for indirect iosapic register access. It reduces lock contention. Signed-off-by: Kenji Kaneshige Signed-off-by: Yasuaki Ishimatsu Signed-off-by: Tony Luck --- include/asm-ia64/iosapic.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-ia64/iosapic.h b/include/asm-ia64/iosapic.h index 421cb6b62a7c..09bdc3898df8 100644 --- a/include/asm-ia64/iosapic.h +++ b/include/asm-ia64/iosapic.h @@ -53,13 +53,13 @@ #define NR_IOSAPICS 256 -static inline unsigned int iosapic_read(char __iomem *iosapic, unsigned int reg) +static inline unsigned int __iosapic_read(char __iomem *iosapic, unsigned int reg) { writel(reg, iosapic + IOSAPIC_REG_SELECT); return readl(iosapic + IOSAPIC_WINDOW); } -static inline void iosapic_write(char __iomem *iosapic, unsigned int reg, u32 val) +static inline void __iosapic_write(char __iomem *iosapic, unsigned int reg, u32 val) { writel(reg, iosapic + IOSAPIC_REG_SELECT); writel(val, iosapic + IOSAPIC_WINDOW); -- cgit v1.2.3 From e1b30a392835e92581db09a4e8b4b2ad53a0c370 Mon Sep 17 00:00:00 2001 From: Yasuaki Ishimatsu Date: Tue, 17 Jul 2007 21:22:23 +0900 Subject: [IA64] Add mapping table between irq and vector Add mapping tables between irqs and vectors, and its management code. This is necessary for supporting multiple vector domain because 1:1 mapping between irq and vector will be changed to n:1. The irq == vector relationship between irqs and vectors is explicitly remained for percpu interrupts, platform interrupts, isa IRQs and vectors assigned using assign_irq_vector() because some programs might depend on it. And I should consider the following problem. When pci drivers enabled/disabled devices dynamically, its irq number is changed to the different one. Therefore, suspend/resume code may happen problem. To fix this problem, I bound gsi to irq. Signed-off-by: Kenji Kaneshige Signed-off-by: Yasuaki Ishimatsu Signed-off-by: Tony Luck --- arch/ia64/kernel/iosapic.c | 61 ++++++++------ arch/ia64/kernel/irq.c | 2 +- arch/ia64/kernel/irq_ia64.c | 201 ++++++++++++++++++++++++++++++++++++-------- arch/ia64/kernel/smpboot.c | 4 + include/asm-ia64/hw_irq.h | 15 +++- 5 files changed, 219 insertions(+), 64 deletions(-) (limited to 'include') diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index cffb443a557c..cf27cfb4d165 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -117,6 +117,9 @@ static DEFINE_SPINLOCK(iosapic_lock); * These tables map IA-64 vectors to the IOSAPIC pin that generates this * vector. */ + +#define NO_REF_RTE 0 + static struct iosapic { char __iomem *addr; /* base address of IOSAPIC */ unsigned int gsi_base; /* GSI base */ @@ -204,7 +207,7 @@ inline int gsi_to_vector (unsigned int gsi) { int irq = __gsi_to_irq(gsi); - if (irq < 0) + if (check_irq_used(irq) < 0) return -1; return irq_to_vector(irq); } @@ -619,14 +622,18 @@ register_intr (unsigned int gsi, int irq, unsigned char delivery, iosapic_intr_info[irq].count++; iosapic_lists[index].rtes_inuse++; } - else if (irq_is_shared(irq)) { + else if (rte->refcnt == NO_REF_RTE) { struct iosapic_intr_info *info = &iosapic_intr_info[irq]; - if (info->trigger != trigger || info->polarity != polarity) { + if (info->count > 0 && + (info->trigger != trigger || info->polarity != polarity)){ printk (KERN_WARNING "%s: cannot override the interrupt\n", __FUNCTION__); return -EINVAL; } + rte->refcnt++; + iosapic_intr_info[irq].count++; + iosapic_lists[index].rtes_inuse++; } iosapic_intr_info[irq].polarity = polarity; @@ -756,12 +763,17 @@ iosapic_register_intr (unsigned int gsi, irq = __gsi_to_irq(gsi); if (irq > 0) { rte = find_rte(irq, gsi); - rte->refcnt++; - goto unlock_iosapic_lock; - } + if(iosapic_intr_info[irq].count == 0) { + assign_irq_vector(irq); + dynamic_irq_init(irq); + } else if (rte->refcnt != NO_REF_RTE) { + rte->refcnt++; + goto unlock_iosapic_lock; + } + } else + irq = create_irq(); /* If vector is running out, we try to find a sharable vector */ - irq = create_irq(); if (irq < 0) { irq = iosapic_find_sharable_irq(trigger, polarity); if (irq < 0) @@ -832,18 +844,14 @@ iosapic_unregister_intr (unsigned int gsi) if (--rte->refcnt > 0) goto out; - /* Remove the rte entry from the list */ idesc = irq_desc + irq; - spin_lock(&idesc->lock); - list_del(&rte->rte_list); - spin_unlock(&idesc->lock); + rte->refcnt = NO_REF_RTE; /* Mask the interrupt */ low32 = iosapic_intr_info[irq].low32 | IOSAPIC_MASK; iosapic_write(rte->iosapic, IOSAPIC_RTE_LOW(rte->rte_index), low32); iosapic_intr_info[irq].count--; - iosapic_free_rte(rte); index = find_iosapic(gsi); iosapic_lists[index].rtes_inuse--; WARN_ON(iosapic_lists[index].rtes_inuse < 0); @@ -857,21 +865,20 @@ iosapic_unregister_intr (unsigned int gsi) (polarity == IOSAPIC_POL_HIGH ? "high" : "low"), cpu_logical_id(dest), dest, irq_to_vector(irq)); - if (list_empty(&iosapic_intr_info[irq].rtes)) { - /* Sanity check */ - BUG_ON(iosapic_intr_info[irq].count); + if (iosapic_intr_info[irq].count == 0) { #ifdef CONFIG_SMP /* Clear affinity */ cpus_setall(idesc->affinity); #endif /* Clear the interrupt information */ - memset(&iosapic_intr_info[irq], 0, - sizeof(struct iosapic_intr_info)); + iosapic_intr_info[irq].dest = 0; + iosapic_intr_info[irq].dmode = 0; + iosapic_intr_info[irq].polarity = 0; + iosapic_intr_info[irq].trigger = 0; iosapic_intr_info[irq].low32 |= IOSAPIC_MASK; - INIT_LIST_HEAD(&iosapic_intr_info[irq].rtes); - /* Destroy IRQ */ - destroy_irq(irq); + /* Destroy and reserve IRQ */ + destroy_and_reserve_irq(irq); } out: spin_unlock_irqrestore(&iosapic_lock, flags); @@ -892,8 +899,8 @@ iosapic_register_platform_intr (u32 int_type, unsigned int gsi, switch (int_type) { case ACPI_INTERRUPT_PMI: - vector = iosapic_vector; - irq = vector; /* FIXME */ + irq = vector = iosapic_vector; + bind_irq_vector(irq, vector); /* * since PMI vector is alloc'd by FW(ACPI) not by kernel, * we need to make sure the vector is available @@ -909,8 +916,8 @@ iosapic_register_platform_intr (u32 int_type, unsigned int gsi, delivery = IOSAPIC_INIT; break; case ACPI_INTERRUPT_CPEI: - vector = IA64_CPE_VECTOR; - irq = vector; /* FIXME */ + irq = vector = IA64_CPE_VECTOR; + BUG_ON(bind_irq_vector(irq, vector)); delivery = IOSAPIC_LOWEST_PRIORITY; mask = 1; break; @@ -945,8 +952,8 @@ iosapic_override_isa_irq (unsigned int isa_irq, unsigned int gsi, int vector, irq; unsigned int dest = cpu_physical_id(smp_processor_id()); - vector = isa_irq_to_vector(isa_irq); - irq = vector; /* FIXME */ + irq = vector = isa_irq_to_vector(isa_irq); + BUG_ON(bind_irq_vector(irq, vector)); register_intr(gsi, irq, IOSAPIC_LOWEST_PRIORITY, polarity, trigger); DBG("ISA: IRQ %u -> GSI %u (%s,%s) -> CPU %d (0x%04x) vector %d\n", @@ -966,6 +973,8 @@ iosapic_system_init (int system_pcat_compat) iosapic_intr_info[irq].low32 = IOSAPIC_MASK; /* mark as unused */ INIT_LIST_HEAD(&iosapic_intr_info[irq].rtes); + + iosapic_intr_info[irq].count = 0; } pcat_compat = system_pcat_compat; diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c index 407b45870489..cc3ee4ef37af 100644 --- a/arch/ia64/kernel/irq.c +++ b/arch/ia64/kernel/irq.c @@ -35,7 +35,7 @@ void ack_bad_irq(unsigned int irq) #ifdef CONFIG_IA64_GENERIC unsigned int __ia64_local_vector_to_irq (ia64_vector vec) { - return (unsigned int) vec; + return __get_cpu_var(vector_irq)[vec]; } #endif diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c index bc47049f060f..072427c2c3f6 100644 --- a/arch/ia64/kernel/irq_ia64.c +++ b/arch/ia64/kernel/irq_ia64.c @@ -46,6 +46,12 @@ #define IRQ_DEBUG 0 +#define IRQ_VECTOR_UNASSIGNED (0) + +#define IRQ_UNUSED (0) +#define IRQ_USED (1) +#define IRQ_RSVD (2) + /* These can be overridden in platform_irq_init */ int ia64_first_device_vector = IA64_DEF_FIRST_DEVICE_VECTOR; int ia64_last_device_vector = IA64_DEF_LAST_DEVICE_VECTOR; @@ -64,46 +70,161 @@ __u8 isa_irq_to_vector_map[16] = { }; EXPORT_SYMBOL(isa_irq_to_vector_map); -static unsigned long ia64_vector_mask[BITS_TO_LONGS(IA64_MAX_DEVICE_VECTORS)]; +DEFINE_SPINLOCK(vector_lock); + +struct irq_cfg irq_cfg[NR_IRQS] __read_mostly = { + [0 ... NR_IRQS - 1] = { .vector = IRQ_VECTOR_UNASSIGNED } +}; + +DEFINE_PER_CPU(int[IA64_NUM_VECTORS], vector_irq) = { + [0 ... IA64_NUM_VECTORS - 1] = IA64_SPURIOUS_INT_VECTOR +}; + +static int irq_status[NR_IRQS] = { + [0 ... NR_IRQS -1] = IRQ_UNUSED +}; + +int check_irq_used(int irq) +{ + if (irq_status[irq] == IRQ_USED) + return 1; + + return -1; +} + +static void reserve_irq(unsigned int irq) +{ + unsigned long flags; + + spin_lock_irqsave(&vector_lock, flags); + irq_status[irq] = IRQ_RSVD; + spin_unlock_irqrestore(&vector_lock, flags); +} + +static inline int find_unassigned_irq(void) +{ + int irq; + + for (irq = IA64_FIRST_DEVICE_VECTOR; irq < NR_IRQS; irq++) + if (irq_status[irq] == IRQ_UNUSED) + return irq; + return -ENOSPC; +} + +static inline int find_unassigned_vector(void) +{ + int vector; + + for (vector = IA64_FIRST_DEVICE_VECTOR; + vector <= IA64_LAST_DEVICE_VECTOR; vector++) + if (__get_cpu_var(vector_irq[vector]) == IA64_SPURIOUS_INT_VECTOR) + return vector; + return -ENOSPC; +} + +static int __bind_irq_vector(int irq, int vector) +{ + int cpu; + + if (irq_to_vector(irq) == vector) + return 0; + if (irq_to_vector(irq) != IRQ_VECTOR_UNASSIGNED) + return -EBUSY; + for_each_online_cpu(cpu) + per_cpu(vector_irq, cpu)[vector] = irq; + irq_cfg[irq].vector = vector; + irq_status[irq] = IRQ_USED; + return 0; +} + +int bind_irq_vector(int irq, int vector) +{ + unsigned long flags; + int ret; + + spin_lock_irqsave(&vector_lock, flags); + ret = __bind_irq_vector(irq, vector); + spin_unlock_irqrestore(&vector_lock, flags); + return ret; +} + +static void clear_irq_vector(int irq) +{ + unsigned long flags; + int vector, cpu; + + spin_lock_irqsave(&vector_lock, flags); + BUG_ON((unsigned)irq >= NR_IRQS); + BUG_ON(irq_cfg[irq].vector == IRQ_VECTOR_UNASSIGNED); + vector = irq_cfg[irq].vector; + for_each_online_cpu(cpu) + per_cpu(vector_irq, cpu)[vector] = IA64_SPURIOUS_INT_VECTOR; + irq_cfg[irq].vector = IRQ_VECTOR_UNASSIGNED; + irq_status[irq] = IRQ_UNUSED; + spin_unlock_irqrestore(&vector_lock, flags); +} int assign_irq_vector (int irq) { - int pos, vector; - again: - pos = find_first_zero_bit(ia64_vector_mask, IA64_NUM_DEVICE_VECTORS); - vector = IA64_FIRST_DEVICE_VECTOR + pos; - if (vector > IA64_LAST_DEVICE_VECTOR) - return -ENOSPC; - if (test_and_set_bit(pos, ia64_vector_mask)) - goto again; + unsigned long flags; + int vector = -ENOSPC; + + if (irq < 0) { + goto out; + } + spin_lock_irqsave(&vector_lock, flags); + vector = find_unassigned_vector(); + if (vector < 0) + goto out; + BUG_ON(__bind_irq_vector(irq, vector)); + spin_unlock_irqrestore(&vector_lock, flags); + out: return vector; } void free_irq_vector (int vector) { - int pos; - - if (vector < IA64_FIRST_DEVICE_VECTOR || vector > IA64_LAST_DEVICE_VECTOR) + if (vector < IA64_FIRST_DEVICE_VECTOR || + vector > IA64_LAST_DEVICE_VECTOR) return; - - pos = vector - IA64_FIRST_DEVICE_VECTOR; - if (!test_and_clear_bit(pos, ia64_vector_mask)) - printk(KERN_WARNING "%s: double free!\n", __FUNCTION__); + clear_irq_vector(vector); } int reserve_irq_vector (int vector) { - int pos; - if (vector < IA64_FIRST_DEVICE_VECTOR || vector > IA64_LAST_DEVICE_VECTOR) return -EINVAL; + return !!bind_irq_vector(vector, vector); +} - pos = vector - IA64_FIRST_DEVICE_VECTOR; - return test_and_set_bit(pos, ia64_vector_mask); +/* + * Initialize vector_irq on a new cpu. This function must be called + * with vector_lock held. + */ +void __setup_vector_irq(int cpu) +{ + int irq, vector; + + /* Clear vector_irq */ + for (vector = 0; vector < IA64_NUM_VECTORS; ++vector) + per_cpu(vector_irq, cpu)[vector] = IA64_SPURIOUS_INT_VECTOR; + /* Mark the inuse vectors */ + for (irq = 0; irq < NR_IRQS; ++irq) { + if ((vector = irq_to_vector(irq)) != IRQ_VECTOR_UNASSIGNED) + per_cpu(vector_irq, cpu)[vector] = irq; + } +} + +void destroy_and_reserve_irq(unsigned int irq) +{ + dynamic_irq_cleanup(irq); + + clear_irq_vector(irq); + reserve_irq(irq); } /* @@ -111,18 +232,29 @@ reserve_irq_vector (int vector) */ int create_irq(void) { - int vector = assign_irq_vector(AUTO_ASSIGN); - - if (vector >= 0) - dynamic_irq_init(vector); - - return vector; + unsigned long flags; + int irq, vector; + + irq = -ENOSPC; + spin_lock_irqsave(&vector_lock, flags); + vector = find_unassigned_vector(); + if (vector < 0) + goto out; + irq = find_unassigned_irq(); + if (irq < 0) + goto out; + BUG_ON(__bind_irq_vector(irq, vector)); + out: + spin_unlock_irqrestore(&vector_lock, flags); + if (irq >= 0) + dynamic_irq_init(irq); + return irq; } void destroy_irq(unsigned int irq) { dynamic_irq_cleanup(irq); - free_irq_vector(irq); + clear_irq_vector(irq); } #ifdef CONFIG_SMP @@ -301,14 +433,13 @@ register_percpu_irq (ia64_vector vec, struct irqaction *action) irq_desc_t *desc; unsigned int irq; - for (irq = 0; irq < NR_IRQS; ++irq) - if (irq_to_vector(irq) == vec) { - desc = irq_desc + irq; - desc->status |= IRQ_PER_CPU; - desc->chip = &irq_type_ia64_lsapic; - if (action) - setup_irq(irq, action); - } + irq = vec; + BUG_ON(bind_irq_vector(irq, vec)); + desc = irq_desc + irq; + desc->status |= IRQ_PER_CPU; + desc->chip = &irq_type_ia64_lsapic; + if (action) + setup_irq(irq, action); } void __init diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c index 3c9d8e6089cf..9f5c90b594b9 100644 --- a/arch/ia64/kernel/smpboot.c +++ b/arch/ia64/kernel/smpboot.c @@ -395,9 +395,13 @@ smp_callin (void) fix_b0_for_bsp(); lock_ipi_calllock(); + spin_lock(&vector_lock); + /* Setup the per cpu irq handling data structures */ + __setup_vector_irq(cpuid); cpu_set(cpuid, cpu_online_map); unlock_ipi_calllock(); per_cpu(cpu_state, cpuid) = CPU_ONLINE; + spin_unlock(&vector_lock); smp_setup_percpu_timer(); diff --git a/include/asm-ia64/hw_irq.h b/include/asm-ia64/hw_irq.h index c054d7a9aaa7..4eff7ff2da82 100644 --- a/include/asm-ia64/hw_irq.h +++ b/include/asm-ia64/hw_irq.h @@ -90,13 +90,24 @@ enum { extern __u8 isa_irq_to_vector_map[16]; #define isa_irq_to_vector(x) isa_irq_to_vector_map[(x)] +struct irq_cfg { + ia64_vector vector; +}; +extern spinlock_t vector_lock; +extern struct irq_cfg irq_cfg[NR_IRQS]; +DECLARE_PER_CPU(int[IA64_NUM_VECTORS], vector_irq); + extern struct hw_interrupt_type irq_type_ia64_lsapic; /* CPU-internal interrupt controller */ +extern int bind_irq_vector(int irq, int vector); extern int assign_irq_vector (int irq); /* allocate a free vector */ extern void free_irq_vector (int vector); extern int reserve_irq_vector (int vector); +extern void __setup_vector_irq(int cpu); extern void ia64_send_ipi (int cpu, int vector, int delivery_mode, int redirect); extern void register_percpu_irq (ia64_vector vec, struct irqaction *action); +extern int check_irq_used (int irq); +extern void destroy_and_reserve_irq (unsigned int irq); static inline void ia64_resend_irq(unsigned int vector) { @@ -113,7 +124,7 @@ extern irq_desc_t irq_desc[NR_IRQS]; static inline unsigned int __ia64_local_vector_to_irq (ia64_vector vec) { - return (unsigned int) vec; + return __get_cpu_var(vector_irq)[vec]; } #endif @@ -131,7 +142,7 @@ __ia64_local_vector_to_irq (ia64_vector vec) static inline ia64_vector irq_to_vector (int irq) { - return (ia64_vector) irq; + return irq_cfg[irq].vector; } /* -- cgit v1.2.3 From 4994be1b3fe9120c88022ff5c0c33f6312b17adb Mon Sep 17 00:00:00 2001 From: Yasuaki Ishimatsu Date: Tue, 17 Jul 2007 21:22:33 +0900 Subject: [IA64] Add support for vector domain Add fundamental support for multiple vector domain. There still exists only one vector domain even with this patch. IRQ migration across domain is not supported yet by this patch. Signed-off-by: Kenji Kaneshige Signed-off-by: Yasuaki Ishimatsu Signed-off-by: Tony Luck --- arch/ia64/kernel/iosapic.c | 13 +++-- arch/ia64/kernel/irq_ia64.c | 120 ++++++++++++++++++++++++++++++++------------ arch/ia64/kernel/msi_ia64.c | 9 +++- include/asm-ia64/hw_irq.h | 4 +- include/asm-ia64/irq.h | 9 +++- 5 files changed, 113 insertions(+), 42 deletions(-) (limited to 'include') diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index cf27cfb4d165..e647254c2707 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -354,6 +354,8 @@ iosapic_set_affinity (unsigned int irq, cpumask_t mask) irq &= (~IA64_IRQ_REDIRECTED); + /* IRQ migration across domain is not supported yet */ + cpus_and(mask, mask, irq_to_domain(irq)); if (cpus_empty(mask)) return; @@ -663,6 +665,7 @@ get_target_cpu (unsigned int gsi, int irq) #ifdef CONFIG_SMP static int cpu = -1; extern int cpe_vector; + cpumask_t domain = irq_to_domain(irq); /* * In case of vector shared by multiple RTEs, all RTEs that @@ -701,7 +704,7 @@ get_target_cpu (unsigned int gsi, int irq) goto skip_numa_setup; cpu_mask = node_to_cpumask(iosapic_lists[iosapic_index].node); - + cpus_and(cpu_mask, cpu_mask, domain); for_each_cpu_mask(numa_cpu, cpu_mask) { if (!cpu_online(numa_cpu)) cpu_clear(numa_cpu, cpu_mask); @@ -731,7 +734,7 @@ skip_numa_setup: do { if (++cpu >= NR_CPUS) cpu = 0; - } while (!cpu_online(cpu)); + } while (!cpu_online(cpu) || !cpu_isset(cpu, domain)); return cpu_physical_id(cpu); #else /* CONFIG_SMP */ @@ -900,7 +903,7 @@ iosapic_register_platform_intr (u32 int_type, unsigned int gsi, switch (int_type) { case ACPI_INTERRUPT_PMI: irq = vector = iosapic_vector; - bind_irq_vector(irq, vector); + bind_irq_vector(irq, vector, CPU_MASK_ALL); /* * since PMI vector is alloc'd by FW(ACPI) not by kernel, * we need to make sure the vector is available @@ -917,7 +920,7 @@ iosapic_register_platform_intr (u32 int_type, unsigned int gsi, break; case ACPI_INTERRUPT_CPEI: irq = vector = IA64_CPE_VECTOR; - BUG_ON(bind_irq_vector(irq, vector)); + BUG_ON(bind_irq_vector(irq, vector, CPU_MASK_ALL)); delivery = IOSAPIC_LOWEST_PRIORITY; mask = 1; break; @@ -953,7 +956,7 @@ iosapic_override_isa_irq (unsigned int isa_irq, unsigned int gsi, unsigned int dest = cpu_physical_id(smp_processor_id()); irq = vector = isa_irq_to_vector(isa_irq); - BUG_ON(bind_irq_vector(irq, vector)); + BUG_ON(bind_irq_vector(irq, vector, CPU_MASK_ALL)); register_intr(gsi, irq, IOSAPIC_LOWEST_PRIORITY, polarity, trigger); DBG("ISA: IRQ %u -> GSI %u (%s,%s) -> CPU %d (0x%04x) vector %d\n", diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c index 072427c2c3f6..a3667631ed80 100644 --- a/arch/ia64/kernel/irq_ia64.c +++ b/arch/ia64/kernel/irq_ia64.c @@ -60,6 +60,8 @@ int ia64_last_device_vector = IA64_DEF_LAST_DEVICE_VECTOR; void __iomem *ipi_base_addr = ((void __iomem *) (__IA64_UNCACHED_OFFSET | IA64_IPI_DEFAULT_BASE_ADDR)); +static cpumask_t vector_allocation_domain(int cpu); + /* * Legacy IRQ to IA-64 vector translation table. */ @@ -73,13 +75,20 @@ EXPORT_SYMBOL(isa_irq_to_vector_map); DEFINE_SPINLOCK(vector_lock); struct irq_cfg irq_cfg[NR_IRQS] __read_mostly = { - [0 ... NR_IRQS - 1] = { .vector = IRQ_VECTOR_UNASSIGNED } + [0 ... NR_IRQS - 1] = { + .vector = IRQ_VECTOR_UNASSIGNED, + .domain = CPU_MASK_NONE + } }; DEFINE_PER_CPU(int[IA64_NUM_VECTORS], vector_irq) = { [0 ... IA64_NUM_VECTORS - 1] = IA64_SPURIOUS_INT_VECTOR }; +static cpumask_t vector_table[IA64_MAX_DEVICE_VECTORS] = { + [0 ... IA64_MAX_DEVICE_VECTORS - 1] = CPU_MASK_NONE +}; + static int irq_status[NR_IRQS] = { [0 ... NR_IRQS -1] = IRQ_UNUSED }; @@ -111,39 +120,54 @@ static inline int find_unassigned_irq(void) return -ENOSPC; } -static inline int find_unassigned_vector(void) +static inline int find_unassigned_vector(cpumask_t domain) { - int vector; + cpumask_t mask; + int pos; + + cpus_and(mask, domain, cpu_online_map); + if (cpus_empty(mask)) + return -EINVAL; - for (vector = IA64_FIRST_DEVICE_VECTOR; - vector <= IA64_LAST_DEVICE_VECTOR; vector++) - if (__get_cpu_var(vector_irq[vector]) == IA64_SPURIOUS_INT_VECTOR) - return vector; + for (pos = 0; pos < IA64_NUM_DEVICE_VECTORS; pos++) { + cpus_and(mask, domain, vector_table[pos]); + if (!cpus_empty(mask)) + continue; + return IA64_FIRST_DEVICE_VECTOR + pos; + } return -ENOSPC; } -static int __bind_irq_vector(int irq, int vector) +static int __bind_irq_vector(int irq, int vector, cpumask_t domain) { - int cpu; + cpumask_t mask; + int cpu, pos; + struct irq_cfg *cfg = &irq_cfg[irq]; - if (irq_to_vector(irq) == vector) + cpus_and(mask, domain, cpu_online_map); + if (cpus_empty(mask)) + return -EINVAL; + if ((cfg->vector == vector) && cpus_equal(cfg->domain, domain)) return 0; - if (irq_to_vector(irq) != IRQ_VECTOR_UNASSIGNED) + if (cfg->vector != IRQ_VECTOR_UNASSIGNED) return -EBUSY; - for_each_online_cpu(cpu) + for_each_cpu_mask(cpu, mask) per_cpu(vector_irq, cpu)[vector] = irq; - irq_cfg[irq].vector = vector; + cfg->vector = vector; + cfg->domain = domain; irq_status[irq] = IRQ_USED; + pos = vector - IA64_FIRST_DEVICE_VECTOR; + cpus_or(vector_table[pos], vector_table[pos], domain); return 0; } -int bind_irq_vector(int irq, int vector) +int bind_irq_vector(int irq, int vector, cpumask_t domain) { unsigned long flags; int ret; spin_lock_irqsave(&vector_lock, flags); - ret = __bind_irq_vector(irq, vector); + ret = __bind_irq_vector(irq, vector, domain); spin_unlock_irqrestore(&vector_lock, flags); return ret; } @@ -151,16 +175,24 @@ int bind_irq_vector(int irq, int vector) static void clear_irq_vector(int irq) { unsigned long flags; - int vector, cpu; + int vector, cpu, pos; + cpumask_t mask; + cpumask_t domain; + struct irq_cfg *cfg = &irq_cfg[irq]; spin_lock_irqsave(&vector_lock, flags); BUG_ON((unsigned)irq >= NR_IRQS); - BUG_ON(irq_cfg[irq].vector == IRQ_VECTOR_UNASSIGNED); - vector = irq_cfg[irq].vector; - for_each_online_cpu(cpu) + BUG_ON(cfg->vector == IRQ_VECTOR_UNASSIGNED); + vector = cfg->vector; + domain = cfg->domain; + cpus_and(mask, cfg->domain, cpu_online_map); + for_each_cpu_mask(cpu, mask) per_cpu(vector_irq, cpu)[vector] = IA64_SPURIOUS_INT_VECTOR; - irq_cfg[irq].vector = IRQ_VECTOR_UNASSIGNED; + cfg->vector = IRQ_VECTOR_UNASSIGNED; + cfg->domain = CPU_MASK_NONE; irq_status[irq] = IRQ_UNUSED; + pos = vector - IA64_FIRST_DEVICE_VECTOR; + cpus_andnot(vector_table[pos], vector_table[pos], domain); spin_unlock_irqrestore(&vector_lock, flags); } @@ -168,18 +200,26 @@ int assign_irq_vector (int irq) { unsigned long flags; - int vector = -ENOSPC; + int vector, cpu; + cpumask_t domain; + + vector = -ENOSPC; + spin_lock_irqsave(&vector_lock, flags); if (irq < 0) { goto out; } - spin_lock_irqsave(&vector_lock, flags); - vector = find_unassigned_vector(); + for_each_online_cpu(cpu) { + domain = vector_allocation_domain(cpu); + vector = find_unassigned_vector(domain); + if (vector >= 0) + break; + } if (vector < 0) goto out; - BUG_ON(__bind_irq_vector(irq, vector)); - spin_unlock_irqrestore(&vector_lock, flags); + BUG_ON(__bind_irq_vector(irq, vector, domain)); out: + spin_unlock_irqrestore(&vector_lock, flags); return vector; } @@ -198,7 +238,7 @@ reserve_irq_vector (int vector) if (vector < IA64_FIRST_DEVICE_VECTOR || vector > IA64_LAST_DEVICE_VECTOR) return -EINVAL; - return !!bind_irq_vector(vector, vector); + return !!bind_irq_vector(vector, vector, CPU_MASK_ALL); } /* @@ -214,11 +254,19 @@ void __setup_vector_irq(int cpu) per_cpu(vector_irq, cpu)[vector] = IA64_SPURIOUS_INT_VECTOR; /* Mark the inuse vectors */ for (irq = 0; irq < NR_IRQS; ++irq) { - if ((vector = irq_to_vector(irq)) != IRQ_VECTOR_UNASSIGNED) - per_cpu(vector_irq, cpu)[vector] = irq; + if (!cpu_isset(cpu, irq_cfg[irq].domain)) + continue; + vector = irq_to_vector(irq); + per_cpu(vector_irq, cpu)[vector] = irq; } } +static cpumask_t vector_allocation_domain(int cpu) +{ + return CPU_MASK_ALL; +} + + void destroy_and_reserve_irq(unsigned int irq) { dynamic_irq_cleanup(irq); @@ -233,17 +281,23 @@ void destroy_and_reserve_irq(unsigned int irq) int create_irq(void) { unsigned long flags; - int irq, vector; + int irq, vector, cpu; + cpumask_t domain; - irq = -ENOSPC; + irq = vector = -ENOSPC; spin_lock_irqsave(&vector_lock, flags); - vector = find_unassigned_vector(); + for_each_online_cpu(cpu) { + domain = vector_allocation_domain(cpu); + vector = find_unassigned_vector(domain); + if (vector >= 0) + break; + } if (vector < 0) goto out; irq = find_unassigned_irq(); if (irq < 0) goto out; - BUG_ON(__bind_irq_vector(irq, vector)); + BUG_ON(__bind_irq_vector(irq, vector, domain)); out: spin_unlock_irqrestore(&vector_lock, flags); if (irq >= 0) @@ -434,7 +488,7 @@ register_percpu_irq (ia64_vector vec, struct irqaction *action) unsigned int irq; irq = vec; - BUG_ON(bind_irq_vector(irq, vec)); + BUG_ON(bind_irq_vector(irq, vec, CPU_MASK_ALL)); desc = irq_desc + irq; desc->status |= IRQ_PER_CPU; desc->chip = &irq_type_ia64_lsapic; diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c index c81080df70df..1d22670cc88b 100644 --- a/arch/ia64/kernel/msi_ia64.c +++ b/arch/ia64/kernel/msi_ia64.c @@ -52,6 +52,11 @@ static void ia64_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask) struct msi_msg msg; u32 addr; + /* IRQ migration across domain is not supported yet */ + cpus_and(cpu_mask, cpu_mask, irq_to_domain(irq)); + if (cpus_empty(cpu_mask)) + return; + read_msi_msg(irq, &msg); addr = msg.address_lo; @@ -69,13 +74,15 @@ int ia64_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *desc) struct msi_msg msg; unsigned long dest_phys_id; int irq, vector; + cpumask_t mask; irq = create_irq(); if (irq < 0) return irq; set_irq_msi(irq, desc); - dest_phys_id = cpu_physical_id(first_cpu(cpu_online_map)); + cpus_and(mask, irq_to_domain(irq), cpu_online_map); + dest_phys_id = cpu_physical_id(first_cpu(mask)); vector = irq_to_vector(irq); msg.address_hi = 0; diff --git a/include/asm-ia64/hw_irq.h b/include/asm-ia64/hw_irq.h index 4eff7ff2da82..97ea3900d74a 100644 --- a/include/asm-ia64/hw_irq.h +++ b/include/asm-ia64/hw_irq.h @@ -92,14 +92,16 @@ extern __u8 isa_irq_to_vector_map[16]; struct irq_cfg { ia64_vector vector; + cpumask_t domain; }; extern spinlock_t vector_lock; extern struct irq_cfg irq_cfg[NR_IRQS]; +#define irq_to_domain(x) irq_cfg[(x)].domain DECLARE_PER_CPU(int[IA64_NUM_VECTORS], vector_irq); extern struct hw_interrupt_type irq_type_ia64_lsapic; /* CPU-internal interrupt controller */ -extern int bind_irq_vector(int irq, int vector); +extern int bind_irq_vector(int irq, int vector, cpumask_t domain); extern int assign_irq_vector (int irq); /* allocate a free vector */ extern void free_irq_vector (int vector); extern int reserve_irq_vector (int vector); diff --git a/include/asm-ia64/irq.h b/include/asm-ia64/irq.h index 67221615e317..35b360b82e43 100644 --- a/include/asm-ia64/irq.h +++ b/include/asm-ia64/irq.h @@ -14,8 +14,13 @@ #include #include -#define NR_IRQS 256 -#define NR_IRQ_VECTORS NR_IRQS +#define NR_VECTORS 256 + +#if (NR_VECTORS + 32 * NR_CPUS) < 1024 +#define NR_IRQS (NR_VECTORS + 32 * NR_CPUS) +#else +#define NR_IRQS 1024 +#endif static __inline__ int irq_canonicalize (int irq) -- cgit v1.2.3 From cd378f18cf73d92bf0b6e1e6b5759b5dd729a9f2 Mon Sep 17 00:00:00 2001 From: Yasuaki Ishimatsu Date: Tue, 17 Jul 2007 21:22:48 +0900 Subject: [IA64] Support irq migration across domain Add support for IRQ migration across vector domain. Signed-off-by: Kenji Kaneshige Signed-off-by: Yasuaki Ishimatsu Signed-off-by: Tony Luck --- arch/ia64/kernel/iosapic.c | 20 +++++++++++++++++--- arch/ia64/kernel/irq_ia64.c | 42 +++++++++++++++++++++++++++++++++++++++--- arch/ia64/kernel/msi_ia64.c | 20 ++++++++++++++------ include/asm-ia64/hw_irq.h | 1 + include/asm-ia64/iosapic.h | 2 ++ 5 files changed, 73 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index e647254c2707..c101c8bff27b 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -354,11 +354,13 @@ iosapic_set_affinity (unsigned int irq, cpumask_t mask) irq &= (~IA64_IRQ_REDIRECTED); - /* IRQ migration across domain is not supported yet */ - cpus_and(mask, mask, irq_to_domain(irq)); + cpus_and(mask, mask, cpu_online_map); if (cpus_empty(mask)) return; + if (reassign_irq_vector(irq, first_cpu(mask))) + return; + dest = cpu_physical_id(first_cpu(mask)); if (list_empty(&iosapic_intr_info[irq].rtes)) @@ -376,6 +378,8 @@ iosapic_set_affinity (unsigned int irq, cpumask_t mask) else /* change delivery mode to fixed */ low32 |= (IOSAPIC_FIXED << IOSAPIC_DELIVERY_SHIFT); + low32 &= IOSAPIC_VECTOR_MASK; + low32 |= irq_to_vector(irq); iosapic_intr_info[irq].low32 = low32; iosapic_intr_info[irq].dest = dest; @@ -404,10 +408,20 @@ iosapic_end_level_irq (unsigned int irq) { ia64_vector vec = irq_to_vector(irq); struct iosapic_rte_info *rte; + int do_unmask_irq = 0; + + if (unlikely(irq_desc[irq].status & IRQ_MOVE_PENDING)) { + do_unmask_irq = 1; + mask_irq(irq); + } - move_native_irq(irq); list_for_each_entry(rte, &iosapic_intr_info[irq].rtes, rte_list) iosapic_eoi(rte->iosapic->addr, vec); + + if (unlikely(do_unmask_irq)) { + move_masked_irq(irq); + unmask_irq(irq); + } } #define iosapic_shutdown_level_irq mask_irq diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c index a3667631ed80..22806b94025a 100644 --- a/arch/ia64/kernel/irq_ia64.c +++ b/arch/ia64/kernel/irq_ia64.c @@ -172,15 +172,13 @@ int bind_irq_vector(int irq, int vector, cpumask_t domain) return ret; } -static void clear_irq_vector(int irq) +static void __clear_irq_vector(int irq) { - unsigned long flags; int vector, cpu, pos; cpumask_t mask; cpumask_t domain; struct irq_cfg *cfg = &irq_cfg[irq]; - spin_lock_irqsave(&vector_lock, flags); BUG_ON((unsigned)irq >= NR_IRQS); BUG_ON(cfg->vector == IRQ_VECTOR_UNASSIGNED); vector = cfg->vector; @@ -193,6 +191,14 @@ static void clear_irq_vector(int irq) irq_status[irq] = IRQ_UNUSED; pos = vector - IA64_FIRST_DEVICE_VECTOR; cpus_andnot(vector_table[pos], vector_table[pos], domain); +} + +static void clear_irq_vector(int irq) +{ + unsigned long flags; + + spin_lock_irqsave(&vector_lock, flags); + __clear_irq_vector(irq); spin_unlock_irqrestore(&vector_lock, flags); } @@ -275,6 +281,36 @@ void destroy_and_reserve_irq(unsigned int irq) reserve_irq(irq); } +static int __reassign_irq_vector(int irq, int cpu) +{ + struct irq_cfg *cfg = &irq_cfg[irq]; + int vector; + cpumask_t domain; + + if (cfg->vector == IRQ_VECTOR_UNASSIGNED || !cpu_online(cpu)) + return -EINVAL; + if (cpu_isset(cpu, cfg->domain)) + return 0; + domain = vector_allocation_domain(cpu); + vector = find_unassigned_vector(domain); + if (vector < 0) + return -ENOSPC; + __clear_irq_vector(irq); + BUG_ON(__bind_irq_vector(irq, vector, domain)); + return 0; +} + +int reassign_irq_vector(int irq, int cpu) +{ + unsigned long flags; + int ret; + + spin_lock_irqsave(&vector_lock, flags); + ret = __reassign_irq_vector(irq, cpu); + spin_unlock_irqrestore(&vector_lock, flags); + return ret; +} + /* * Dynamic irq allocate and deallocation for MSI */ diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c index 1d22670cc88b..2fdbd5c3f213 100644 --- a/arch/ia64/kernel/msi_ia64.c +++ b/arch/ia64/kernel/msi_ia64.c @@ -13,6 +13,7 @@ #define MSI_DATA_VECTOR_SHIFT 0 #define MSI_DATA_VECTOR(v) (((u8)v) << MSI_DATA_VECTOR_SHIFT) +#define MSI_DATA_VECTOR_MASK 0xffffff00 #define MSI_DATA_DELIVERY_SHIFT 8 #define MSI_DATA_DELIVERY_FIXED (0 << MSI_DATA_DELIVERY_SHIFT) @@ -50,22 +51,29 @@ static struct irq_chip ia64_msi_chip; static void ia64_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask) { struct msi_msg msg; - u32 addr; + u32 addr, data; + int cpu = first_cpu(cpu_mask); - /* IRQ migration across domain is not supported yet */ - cpus_and(cpu_mask, cpu_mask, irq_to_domain(irq)); - if (cpus_empty(cpu_mask)) + if (!cpu_online(cpu)) + return; + + if (reassign_irq_vector(irq, cpu)) return; read_msi_msg(irq, &msg); addr = msg.address_lo; addr &= MSI_ADDR_DESTID_MASK; - addr |= MSI_ADDR_DESTID_CPU(cpu_physical_id(first_cpu(cpu_mask))); + addr |= MSI_ADDR_DESTID_CPU(cpu_physical_id(cpu)); msg.address_lo = addr; + data = msg.data; + data &= MSI_DATA_VECTOR_MASK; + data |= MSI_DATA_VECTOR(irq_to_vector(irq)); + msg.data = data; + write_msi_msg(irq, &msg); - irq_desc[irq].affinity = cpu_mask; + irq_desc[irq].affinity = cpumask_of_cpu(cpu); } #endif /* CONFIG_SMP */ diff --git a/include/asm-ia64/hw_irq.h b/include/asm-ia64/hw_irq.h index 97ea3900d74a..efa1b8f7251d 100644 --- a/include/asm-ia64/hw_irq.h +++ b/include/asm-ia64/hw_irq.h @@ -106,6 +106,7 @@ extern int assign_irq_vector (int irq); /* allocate a free vector */ extern void free_irq_vector (int vector); extern int reserve_irq_vector (int vector); extern void __setup_vector_irq(int cpu); +extern int reassign_irq_vector(int irq, int cpu); extern void ia64_send_ipi (int cpu, int vector, int delivery_mode, int redirect); extern void register_percpu_irq (ia64_vector vec, struct irqaction *action); extern int check_irq_used (int irq); diff --git a/include/asm-ia64/iosapic.h b/include/asm-ia64/iosapic.h index 09bdc3898df8..b8f712859140 100644 --- a/include/asm-ia64/iosapic.h +++ b/include/asm-ia64/iosapic.h @@ -47,6 +47,8 @@ #define IOSAPIC_MASK_SHIFT 16 #define IOSAPIC_MASK (1< Date: Mon, 16 Jul 2007 15:33:40 +1000 Subject: [IA64] fallocate system call sys_fallocate for ia64. This uses an empty slot #1303 erroneously marked as reserved for move_pages (which had already been allocated as syscall #1276) Signed-Off-By: Dave Chinner Signed-off-by: Tony Luck --- arch/ia64/kernel/entry.S | 2 +- include/asm-ia64/unistd.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index 95f517515235..c36f43c94600 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -1581,7 +1581,7 @@ sys_call_table: data8 sys_sync_file_range // 1300 data8 sys_tee data8 sys_vmsplice - data8 sys_ni_syscall // reserved for move_pages + data8 sys_fallocate data8 sys_getcpu data8 sys_epoll_pwait // 1305 data8 sys_utimensat diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h index 441c9e001776..315f8de950a2 100644 --- a/include/asm-ia64/unistd.h +++ b/include/asm-ia64/unistd.h @@ -292,7 +292,7 @@ #define __NR_sync_file_range 1300 #define __NR_tee 1301 #define __NR_vmsplice 1302 -/* 1303 reserved for move_pages */ +#define __NR_fallocate 1303 #define __NR_getcpu 1304 #define __NR_epoll_pwait 1305 #define __NR_utimensat 1306 -- cgit v1.2.3 From 0aa366f351d044703e25c8425e508170e80d83b1 Mon Sep 17 00:00:00 2001 From: Tony Luck Date: Fri, 20 Jul 2007 11:22:30 -0700 Subject: [IA64] Convert to generic timekeeping/clocksource This is a merge of Peter Keilty's initial patch (which was revived by Bob Picco) for this with Hidetoshi Seto's fixes and scaling improvements. Acked-by: Bob Picco Signed-off-by: Tony Luck --- Documentation/kernel-parameters.txt | 2 + arch/ia64/Kconfig | 6 +- arch/ia64/configs/bigsur_defconfig | 2 +- arch/ia64/configs/gensparse_defconfig | 2 +- arch/ia64/configs/sim_defconfig | 2 +- arch/ia64/configs/sn2_defconfig | 2 +- arch/ia64/configs/tiger_defconfig | 2 +- arch/ia64/configs/zx1_defconfig | 2 +- arch/ia64/defconfig | 2 +- arch/ia64/kernel/asm-offsets.c | 35 ++++--- arch/ia64/kernel/cyclone.c | 46 ++++++--- arch/ia64/kernel/fsys.S | 179 +++++++++++++++++----------------- arch/ia64/kernel/fsyscall_gtod_data.h | 23 +++++ arch/ia64/kernel/time.c | 96 ++++++++++++++++-- arch/ia64/sn/kernel/sn2/timer.c | 29 ++++-- drivers/acpi/processor_idle.c | 4 +- drivers/char/hpet.c | 70 +++++++------ include/linux/clocksource.h | 6 ++ 18 files changed, 335 insertions(+), 175 deletions(-) create mode 100644 arch/ia64/kernel/fsyscall_gtod_data.h (limited to 'include') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 9a541486fb7e..68115d7b0a7a 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1154,6 +1154,8 @@ and is between 256 and 4096 characters. It is defined in the file nointroute [IA-64] + nojitter [IA64] Disables jitter checking for ITC timers. + nolapic [IA-32,APIC] Do not enable or use the local APIC. nolapic_timer [IA-32,APIC] Do not use the local APIC timer. diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 616c96e73483..36c7b9682aa6 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -62,7 +62,11 @@ config GENERIC_CALIBRATE_DELAY bool default y -config TIME_INTERPOLATION +config GENERIC_TIME + bool + default y + +config GENERIC_TIME_VSYSCALL bool default y diff --git a/arch/ia64/configs/bigsur_defconfig b/arch/ia64/configs/bigsur_defconfig index 90e9c2e61bf4..9eb48c0927b0 100644 --- a/arch/ia64/configs/bigsur_defconfig +++ b/arch/ia64/configs/bigsur_defconfig @@ -85,7 +85,7 @@ CONFIG_MMU=y CONFIG_SWIOTLB=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y CONFIG_GENERIC_CALIBRATE_DELAY=y -CONFIG_TIME_INTERPOLATION=y +CONFIG_GENERIC_TIME=y CONFIG_EFI=y CONFIG_GENERIC_IOMAP=y CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y diff --git a/arch/ia64/configs/gensparse_defconfig b/arch/ia64/configs/gensparse_defconfig index 0d29aa2066b3..3a9ed951db08 100644 --- a/arch/ia64/configs/gensparse_defconfig +++ b/arch/ia64/configs/gensparse_defconfig @@ -86,7 +86,7 @@ CONFIG_MMU=y CONFIG_SWIOTLB=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y CONFIG_GENERIC_CALIBRATE_DELAY=y -CONFIG_TIME_INTERPOLATION=y +CONFIG_GENERIC_TIME=y CONFIG_EFI=y CONFIG_GENERIC_IOMAP=y CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y diff --git a/arch/ia64/configs/sim_defconfig b/arch/ia64/configs/sim_defconfig index d9146c31ea13..c420d9f3df98 100644 --- a/arch/ia64/configs/sim_defconfig +++ b/arch/ia64/configs/sim_defconfig @@ -86,7 +86,7 @@ CONFIG_MMU=y CONFIG_SWIOTLB=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y CONFIG_GENERIC_CALIBRATE_DELAY=y -CONFIG_TIME_INTERPOLATION=y +CONFIG_GENERIC_TIME=y CONFIG_EFI=y CONFIG_GENERIC_IOMAP=y CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig index 64e951de4e57..4c9ffc47bc7a 100644 --- a/arch/ia64/configs/sn2_defconfig +++ b/arch/ia64/configs/sn2_defconfig @@ -93,7 +93,7 @@ CONFIG_SWIOTLB=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y CONFIG_GENERIC_FIND_NEXT_BIT=y CONFIG_GENERIC_CALIBRATE_DELAY=y -CONFIG_TIME_INTERPOLATION=y +CONFIG_GENERIC_TIME=y CONFIG_DMI=y CONFIG_EFI=y CONFIG_GENERIC_IOMAP=y diff --git a/arch/ia64/configs/tiger_defconfig b/arch/ia64/configs/tiger_defconfig index a1446931b401..e208747d5de4 100644 --- a/arch/ia64/configs/tiger_defconfig +++ b/arch/ia64/configs/tiger_defconfig @@ -98,7 +98,7 @@ CONFIG_RWSEM_XCHGADD_ALGORITHM=y # CONFIG_ARCH_HAS_ILOG2_U64 is not set CONFIG_GENERIC_FIND_NEXT_BIT=y CONFIG_GENERIC_CALIBRATE_DELAY=y -CONFIG_TIME_INTERPOLATION=y +CONFIG_GENERIC_TIME=y CONFIG_DMI=y CONFIG_EFI=y CONFIG_GENERIC_IOMAP=y diff --git a/arch/ia64/configs/zx1_defconfig b/arch/ia64/configs/zx1_defconfig index 1c7955c16358..4a060fc39934 100644 --- a/arch/ia64/configs/zx1_defconfig +++ b/arch/ia64/configs/zx1_defconfig @@ -96,7 +96,7 @@ CONFIG_RWSEM_XCHGADD_ALGORITHM=y # CONFIG_ARCH_HAS_ILOG2_U64 is not set CONFIG_GENERIC_FIND_NEXT_BIT=y CONFIG_GENERIC_CALIBRATE_DELAY=y -CONFIG_TIME_INTERPOLATION=y +CONFIG_GENERIC_TIME=y CONFIG_DMI=y CONFIG_EFI=y CONFIG_GENERIC_IOMAP=y diff --git a/arch/ia64/defconfig b/arch/ia64/defconfig index 90bd9601cdde..461f8ee738fb 100644 --- a/arch/ia64/defconfig +++ b/arch/ia64/defconfig @@ -98,7 +98,7 @@ CONFIG_RWSEM_XCHGADD_ALGORITHM=y # CONFIG_ARCH_HAS_ILOG2_U64 is not set CONFIG_GENERIC_FIND_NEXT_BIT=y CONFIG_GENERIC_CALIBRATE_DELAY=y -CONFIG_TIME_INTERPOLATION=y +CONFIG_GENERIC_TIME=y CONFIG_DMI=y CONFIG_EFI=y CONFIG_GENERIC_IOMAP=y diff --git a/arch/ia64/kernel/asm-offsets.c b/arch/ia64/kernel/asm-offsets.c index 2236fabbb3c6..0aebc6f79e95 100644 --- a/arch/ia64/kernel/asm-offsets.c +++ b/arch/ia64/kernel/asm-offsets.c @@ -7,6 +7,7 @@ #define ASM_OFFSETS_C 1 #include +#include #include #include @@ -15,6 +16,7 @@ #include #include "../kernel/sigframe.h" +#include "../kernel/fsyscall_gtod_data.h" #define DEFINE(sym, val) \ asm volatile("\n->" #sym " %0 " #val : : "i" (val)) @@ -256,17 +258,24 @@ void foo(void) BLANK(); /* used by fsys_gettimeofday in arch/ia64/kernel/fsys.S */ - DEFINE(IA64_TIME_INTERPOLATOR_ADDRESS_OFFSET, offsetof (struct time_interpolator, addr)); - DEFINE(IA64_TIME_INTERPOLATOR_SOURCE_OFFSET, offsetof (struct time_interpolator, source)); - DEFINE(IA64_TIME_INTERPOLATOR_SHIFT_OFFSET, offsetof (struct time_interpolator, shift)); - DEFINE(IA64_TIME_INTERPOLATOR_NSEC_OFFSET, offsetof (struct time_interpolator, nsec_per_cyc)); - DEFINE(IA64_TIME_INTERPOLATOR_OFFSET_OFFSET, offsetof (struct time_interpolator, offset)); - DEFINE(IA64_TIME_INTERPOLATOR_LAST_CYCLE_OFFSET, offsetof (struct time_interpolator, last_cycle)); - DEFINE(IA64_TIME_INTERPOLATOR_LAST_COUNTER_OFFSET, offsetof (struct time_interpolator, last_counter)); - DEFINE(IA64_TIME_INTERPOLATOR_JITTER_OFFSET, offsetof (struct time_interpolator, jitter)); - DEFINE(IA64_TIME_INTERPOLATOR_MASK_OFFSET, offsetof (struct time_interpolator, mask)); - DEFINE(IA64_TIME_SOURCE_CPU, TIME_SOURCE_CPU); - DEFINE(IA64_TIME_SOURCE_MMIO64, TIME_SOURCE_MMIO64); - DEFINE(IA64_TIME_SOURCE_MMIO32, TIME_SOURCE_MMIO32); - DEFINE(IA64_TIMESPEC_TV_NSEC_OFFSET, offsetof (struct timespec, tv_nsec)); + DEFINE(IA64_GTOD_LOCK_OFFSET, + offsetof (struct fsyscall_gtod_data_t, lock)); + DEFINE(IA64_GTOD_WALL_TIME_OFFSET, + offsetof (struct fsyscall_gtod_data_t, wall_time)); + DEFINE(IA64_GTOD_MONO_TIME_OFFSET, + offsetof (struct fsyscall_gtod_data_t, monotonic_time)); + DEFINE(IA64_CLKSRC_MASK_OFFSET, + offsetof (struct fsyscall_gtod_data_t, clk_mask)); + DEFINE(IA64_CLKSRC_MULT_OFFSET, + offsetof (struct fsyscall_gtod_data_t, clk_mult)); + DEFINE(IA64_CLKSRC_SHIFT_OFFSET, + offsetof (struct fsyscall_gtod_data_t, clk_shift)); + DEFINE(IA64_CLKSRC_MMIO_OFFSET, + offsetof (struct fsyscall_gtod_data_t, clk_fsys_mmio)); + DEFINE(IA64_CLKSRC_CYCLE_LAST_OFFSET, + offsetof (struct fsyscall_gtod_data_t, clk_cycle_last)); + DEFINE(IA64_ITC_JITTER_OFFSET, + offsetof (struct itc_jitter_data_t, itc_jitter)); + DEFINE(IA64_ITC_LASTCYCLE_OFFSET, + offsetof (struct itc_jitter_data_t, itc_lastcycle)); } diff --git a/arch/ia64/kernel/cyclone.c b/arch/ia64/kernel/cyclone.c index e00b21514f7c..2fd96d9062a1 100644 --- a/arch/ia64/kernel/cyclone.c +++ b/arch/ia64/kernel/cyclone.c @@ -3,6 +3,7 @@ #include #include #include +#include #include /* IBM Summit (EXA) Cyclone counter code*/ @@ -18,13 +19,21 @@ void __init cyclone_setup(void) use_cyclone = 1; } +static void __iomem *cyclone_mc; -struct time_interpolator cyclone_interpolator = { - .source = TIME_SOURCE_MMIO64, - .shift = 16, - .frequency = CYCLONE_TIMER_FREQ, - .drift = -100, - .mask = (1LL << 40) - 1 +static cycle_t read_cyclone(void) +{ + return (cycle_t)readq((void __iomem *)cyclone_mc); +} + +static struct clocksource clocksource_cyclone = { + .name = "cyclone", + .rating = 300, + .read = read_cyclone, + .mask = (1LL << 40) - 1, + .mult = 0, /*to be caluclated*/ + .shift = 16, + .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; int __init init_cyclone_clock(void) @@ -44,13 +53,15 @@ int __init init_cyclone_clock(void) offset = (CYCLONE_CBAR_ADDR); reg = (u64*)ioremap_nocache(offset, sizeof(u64)); if(!reg){ - printk(KERN_ERR "Summit chipset: Could not find valid CBAR register.\n"); + printk(KERN_ERR "Summit chipset: Could not find valid CBAR" + " register.\n"); use_cyclone = 0; return -ENODEV; } base = readq(reg); if(!base){ - printk(KERN_ERR "Summit chipset: Could not find valid CBAR value.\n"); + printk(KERN_ERR "Summit chipset: Could not find valid CBAR" + " value.\n"); use_cyclone = 0; return -ENODEV; } @@ -60,7 +71,8 @@ int __init init_cyclone_clock(void) offset = (base + CYCLONE_PMCC_OFFSET); reg = (u64*)ioremap_nocache(offset, sizeof(u64)); if(!reg){ - printk(KERN_ERR "Summit chipset: Could not find valid PMCC register.\n"); + printk(KERN_ERR "Summit chipset: Could not find valid PMCC" + " register.\n"); use_cyclone = 0; return -ENODEV; } @@ -71,7 +83,8 @@ int __init init_cyclone_clock(void) offset = (base + CYCLONE_MPCS_OFFSET); reg = (u64*)ioremap_nocache(offset, sizeof(u64)); if(!reg){ - printk(KERN_ERR "Summit chipset: Could not find valid MPCS register.\n"); + printk(KERN_ERR "Summit chipset: Could not find valid MPCS" + " register.\n"); use_cyclone = 0; return -ENODEV; } @@ -82,7 +95,8 @@ int __init init_cyclone_clock(void) offset = (base + CYCLONE_MPMC_OFFSET); cyclone_timer = (u32*)ioremap_nocache(offset, sizeof(u32)); if(!cyclone_timer){ - printk(KERN_ERR "Summit chipset: Could not find valid MPMC register.\n"); + printk(KERN_ERR "Summit chipset: Could not find valid MPMC" + " register.\n"); use_cyclone = 0; return -ENODEV; } @@ -93,7 +107,8 @@ int __init init_cyclone_clock(void) int stall = 100; while(stall--) barrier(); if(readl(cyclone_timer) == old){ - printk(KERN_ERR "Summit chipset: Counter not counting! DISABLED\n"); + printk(KERN_ERR "Summit chipset: Counter not counting!" + " DISABLED\n"); iounmap(cyclone_timer); cyclone_timer = 0; use_cyclone = 0; @@ -101,8 +116,11 @@ int __init init_cyclone_clock(void) } } /* initialize last tick */ - cyclone_interpolator.addr = cyclone_timer; - register_time_interpolator(&cyclone_interpolator); + cyclone_mc = cyclone_timer; + clocksource_cyclone.fsys_mmio = cyclone_timer; + clocksource_cyclone.mult = clocksource_hz2mult(CYCLONE_TIMER_FREQ, + clocksource_cyclone.shift); + clocksource_register(&clocksource_cyclone); return 0; } diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S index 3f926c2dc708..44841971f077 100644 --- a/arch/ia64/kernel/fsys.S +++ b/arch/ia64/kernel/fsys.S @@ -147,12 +147,11 @@ ENTRY(fsys_set_tid_address) FSYS_RETURN END(fsys_set_tid_address) -/* - * Ensure that the time interpolator structure is compatible with the asm code - */ -#if IA64_TIME_INTERPOLATOR_SOURCE_OFFSET !=0 || IA64_TIME_INTERPOLATOR_SHIFT_OFFSET != 2 \ - || IA64_TIME_INTERPOLATOR_JITTER_OFFSET != 3 || IA64_TIME_INTERPOLATOR_NSEC_OFFSET != 4 -#error fsys_gettimeofday incompatible with changes to struct time_interpolator +#if IA64_GTOD_LOCK_OFFSET !=0 +#error fsys_gettimeofday incompatible with changes to struct fsyscall_gtod_data_t +#endif +#if IA64_ITC_JITTER_OFFSET !=0 +#error fsys_gettimeofday incompatible with changes to struct itc_jitter_data_t #endif #define CLOCK_REALTIME 0 #define CLOCK_MONOTONIC 1 @@ -179,126 +178,124 @@ ENTRY(fsys_gettimeofday) // r11 = preserved: saved ar.pfs // r12 = preserved: memory stack // r13 = preserved: thread pointer - // r14 = address of mask / mask + // r14 = address of mask / mask value // r15 = preserved: system call number // r16 = preserved: current task pointer - // r17 = wall to monotonic use - // r18 = time_interpolator->offset - // r19 = address of wall_to_monotonic - // r20 = pointer to struct time_interpolator / pointer to time_interpolator->address - // r21 = shift factor - // r22 = address of time interpolator->last_counter - // r23 = address of time_interpolator->last_cycle - // r24 = adress of time_interpolator->offset - // r25 = last_cycle value - // r26 = last_counter value - // r27 = pointer to xtime + // r17 = (not used) + // r18 = (not used) + // r19 = address of itc_lastcycle + // r20 = struct fsyscall_gtod_data (= address of gtod_lock.sequence) + // r21 = address of mmio_ptr + // r22 = address of wall_time or monotonic_time + // r23 = address of shift / value + // r24 = address mult factor / cycle_last value + // r25 = itc_lastcycle value + // r26 = address clocksource cycle_last + // r27 = (not used) // r28 = sequence number at the beginning of critcal section - // r29 = address of seqlock + // r29 = address of itc_jitter // r30 = time processing flags / memory address // r31 = pointer to result // Predicates // p6,p7 short term use // p8 = timesource ar.itc // p9 = timesource mmio64 - // p10 = timesource mmio32 + // p10 = timesource mmio32 - not used // p11 = timesource not to be handled by asm code - // p12 = memory time source ( = p9 | p10) - // p13 = do cmpxchg with time_interpolator_last_cycle + // p12 = memory time source ( = p9 | p10) - not used + // p13 = do cmpxchg with itc_lastcycle // p14 = Divide by 1000 // p15 = Add monotonic // - // Note that instructions are optimized for McKinley. McKinley can process two - // bundles simultaneously and therefore we continuously try to feed the CPU - // two bundles and then a stop. - tnat.nz p6,p0 = r31 // branch deferred since it does not fit into bundle structure + // Note that instructions are optimized for McKinley. McKinley can + // process two bundles simultaneously and therefore we continuously + // try to feed the CPU two bundles and then a stop. + // + // Additional note that code has changed a lot. Optimization is TBD. + // Comments begin with "?" are maybe outdated. + tnat.nz p6,p0 = r31 // ? branch deferred to fit later bundle mov pr = r30,0xc000 // Set predicates according to function add r2 = TI_FLAGS+IA64_TASK_SIZE,r16 - movl r20 = time_interpolator + movl r20 = fsyscall_gtod_data // load fsyscall gettimeofday data address ;; - ld8 r20 = [r20] // get pointer to time_interpolator structure - movl r29 = xtime_lock + movl r29 = itc_jitter_data // itc_jitter + add r22 = IA64_GTOD_WALL_TIME_OFFSET,r20 // wall_time ld4 r2 = [r2] // process work pending flags - movl r27 = xtime - ;; // only one bundle here - ld8 r21 = [r20] // first quad with control information + ;; +(p15) add r22 = IA64_GTOD_MONO_TIME_OFFSET,r20 // monotonic_time + add r21 = IA64_CLKSRC_MMIO_OFFSET,r20 + add r19 = IA64_ITC_LASTCYCLE_OFFSET,r29 and r2 = TIF_ALLWORK_MASK,r2 -(p6) br.cond.spnt.few .fail_einval // deferred branch +(p6) br.cond.spnt.few .fail_einval // ? deferred branch ;; - add r10 = IA64_TIME_INTERPOLATOR_ADDRESS_OFFSET,r20 - extr r3 = r21,32,32 // time_interpolator->nsec_per_cyc - extr r8 = r21,0,16 // time_interpolator->source + add r26 = IA64_CLKSRC_CYCLE_LAST_OFFSET,r20 // clksrc_cycle_last cmp.ne p6, p0 = 0, r2 // Fallback if work is scheduled (p6) br.cond.spnt.many fsys_fallback_syscall ;; - cmp.eq p8,p12 = 0,r8 // Check for cpu timer - cmp.eq p9,p0 = 1,r8 // MMIO64 ? - extr r2 = r21,24,8 // time_interpolator->jitter - cmp.eq p10,p0 = 2,r8 // MMIO32 ? - cmp.ltu p11,p0 = 2,r8 // function or other clock -(p11) br.cond.spnt.many fsys_fallback_syscall + // Begin critical section +.time_redo: + ld4.acq r28 = [r20] // gtod_lock.sequence, Must take first + ;; + and r28 = ~1,r28 // And make sequence even to force retry if odd ;; - setf.sig f7 = r3 // Setup for scaling of counter -(p15) movl r19 = wall_to_monotonic -(p12) ld8 r30 = [r10] - cmp.ne p13,p0 = r2,r0 // need jitter compensation? - extr r21 = r21,16,8 // shift factor + ld8 r30 = [r21] // clocksource->mmio_ptr + add r24 = IA64_CLKSRC_MULT_OFFSET,r20 + ld4 r2 = [r29] // itc_jitter value + add r23 = IA64_CLKSRC_SHIFT_OFFSET,r20 + add r14 = IA64_CLKSRC_MASK_OFFSET,r20 ;; -.time_redo: - .pred.rel.mutex p8,p9,p10 - ld4.acq r28 = [r29] // xtime_lock.sequence. Must come first for locking purposes + ld4 r3 = [r24] // clocksource mult value + ld8 r14 = [r14] // clocksource mask value + cmp.eq p8,p9 = 0,r30 // use cpu timer if no mmio_ptr ;; - and r28 = ~1,r28 // Make sequence even to force retry if odd + setf.sig f7 = r3 // Setup for mult scaling of counter +(p8) cmp.ne p13,p0 = r2,r0 // need itc_jitter compensation, set p13 + ld4 r23 = [r23] // clocksource shift value + ld8 r24 = [r26] // get clksrc_cycle_last value +(p9) cmp.eq p13,p0 = 0,r30 // if mmio_ptr, clear p13 jitter control ;; + .pred.rel.mutex p8,p9 (p8) mov r2 = ar.itc // CPU_TIMER. 36 clocks latency!!! - add r22 = IA64_TIME_INTERPOLATOR_LAST_COUNTER_OFFSET,r20 -(p9) ld8 r2 = [r30] // readq(ti->address). Could also have latency issues.. -(p10) ld4 r2 = [r30] // readw(ti->address) -(p13) add r23 = IA64_TIME_INTERPOLATOR_LAST_CYCLE_OFFSET,r20 - ;; // could be removed by moving the last add upward - ld8 r26 = [r22] // time_interpolator->last_counter -(p13) ld8 r25 = [r23] // time interpolator->last_cycle - add r24 = IA64_TIME_INTERPOLATOR_OFFSET_OFFSET,r20 -(p15) ld8 r17 = [r19],IA64_TIMESPEC_TV_NSEC_OFFSET - ld8 r9 = [r27],IA64_TIMESPEC_TV_NSEC_OFFSET - add r14 = IA64_TIME_INTERPOLATOR_MASK_OFFSET, r20 - ;; - ld8 r18 = [r24] // time_interpolator->offset - ld8 r8 = [r27],-IA64_TIMESPEC_TV_NSEC_OFFSET // xtime.tv_nsec -(p13) sub r3 = r25,r2 // Diff needed before comparison (thanks davidm) - ;; - ld8 r14 = [r14] // time_interpolator->mask -(p13) cmp.gt.unc p6,p7 = r3,r0 // check if it is less than last. p6,p7 cleared - sub r10 = r2,r26 // current_counter - last_counter - ;; -(p6) sub r10 = r25,r26 // time we got was less than last_cycle +(p9) ld8 r2 = [r30] // MMIO_TIMER. Could also have latency issues.. +(p13) ld8 r25 = [r19] // get itc_lastcycle value + ;; // ? could be removed by moving the last add upward + ld8 r9 = [r22],IA64_TIMESPEC_TV_NSEC_OFFSET // tv_sec + ;; + ld8 r8 = [r22],-IA64_TIMESPEC_TV_NSEC_OFFSET // tv_nsec +(p13) sub r3 = r25,r2 // Diff needed before comparison (thanks davidm) + ;; +(p13) cmp.gt.unc p6,p7 = r3,r0 // check if it is less than last. p6,p7 cleared + sub r10 = r2,r24 // current_cycle - last_cycle + ;; +(p6) sub r10 = r25,r24 // time we got was less than last_cycle (p7) mov ar.ccv = r25 // more than last_cycle. Prep for cmpxchg ;; +(p7) cmpxchg8.rel r3 = [r19],r2,ar.ccv + ;; +(p7) cmp.ne p7,p0 = r25,r3 // if cmpxchg not successful + ;; +(p7) sub r10 = r3,r24 // then use new last_cycle instead + ;; and r10 = r10,r14 // Apply mask ;; setf.sig f8 = r10 nop.i 123 ;; -(p7) cmpxchg8.rel r3 = [r23],r2,ar.ccv -EX(.fail_efault, probe.w.fault r31, 3) // This takes 5 cycles and we have spare time + // fault check takes 5 cycles and we have spare time +EX(.fail_efault, probe.w.fault r31, 3) xmpy.l f8 = f8,f7 // nsec_per_cyc*(counter-last_counter) -(p15) add r9 = r9,r17 // Add wall to monotonic.secs to result secs ;; -(p15) ld8 r17 = [r19],-IA64_TIMESPEC_TV_NSEC_OFFSET -(p7) cmp.ne p7,p0 = r25,r3 // if cmpxchg not successful redo - // simulate tbit.nz.or p7,p0 = r28,0 + // ? simulate tbit.nz.or p7,p0 = r28,0 getf.sig r2 = f8 mf - add r8 = r8,r18 // Add time interpolator offset ;; - ld4 r10 = [r29] // xtime_lock.sequence -(p15) add r8 = r8, r17 // Add monotonic.nsecs to nsecs - shr.u r2 = r2,r21 - ;; // overloaded 3 bundles! - // End critical section. + ld4 r10 = [r20] // gtod_lock.sequence + shr.u r2 = r2,r23 // shift by factor + ;; // ? overloaded 3 bundles! add r8 = r8,r2 // Add xtime.nsecs - cmp4.ne.or p7,p0 = r28,r10 -(p7) br.cond.dpnt.few .time_redo // sequence number changed ? + cmp4.ne p7,p0 = r28,r10 +(p7) br.cond.dpnt.few .time_redo // sequence number changed, redo + // End critical section. // Now r8=tv->tv_nsec and r9=tv->tv_sec mov r10 = r0 movl r2 = 1000000000 @@ -308,19 +305,19 @@ EX(.fail_efault, probe.w.fault r31, 3) // This takes 5 cycles and we have spare .time_normalize: mov r21 = r8 cmp.ge p6,p0 = r8,r2 -(p14) shr.u r20 = r8, 3 // We can repeat this if necessary just wasting some time +(p14) shr.u r20 = r8, 3 // We can repeat this if necessary just wasting time ;; (p14) setf.sig f8 = r20 (p6) sub r8 = r8,r2 -(p6) add r9 = 1,r9 // two nops before the branch. -(p14) setf.sig f7 = r3 // Chances for repeats are 1 in 10000 for gettod +(p6) add r9 = 1,r9 // two nops before the branch. +(p14) setf.sig f7 = r3 // Chances for repeats are 1 in 10000 for gettod (p6) br.cond.dpnt.few .time_normalize ;; // Divided by 8 though shift. Now divide by 125 // The compiler was able to do that with a multiply // and a shift and we do the same -EX(.fail_efault, probe.w.fault r23, 3) // This also costs 5 cycles -(p14) xmpy.hu f8 = f8, f7 // xmpy has 5 cycles latency so use it... +EX(.fail_efault, probe.w.fault r23, 3) // This also costs 5 cycles +(p14) xmpy.hu f8 = f8, f7 // xmpy has 5 cycles latency so use it ;; mov r8 = r0 (p14) getf.sig r2 = f8 diff --git a/arch/ia64/kernel/fsyscall_gtod_data.h b/arch/ia64/kernel/fsyscall_gtod_data.h new file mode 100644 index 000000000000..490dab55fba3 --- /dev/null +++ b/arch/ia64/kernel/fsyscall_gtod_data.h @@ -0,0 +1,23 @@ +/* + * (c) Copyright 2007 Hewlett-Packard Development Company, L.P. + * Contributed by Peter Keilty + * + * fsyscall gettimeofday data + */ + +struct fsyscall_gtod_data_t { + seqlock_t lock; + struct timespec wall_time; + struct timespec monotonic_time; + cycle_t clk_mask; + u32 clk_mult; + u32 clk_shift; + void *clk_fsys_mmio; + cycle_t clk_cycle_last; +} __attribute__ ((aligned (L1_CACHE_BYTES))); + +struct itc_jitter_data_t { + int itc_jitter; + cycle_t itc_lastcycle; +} __attribute__ ((aligned (L1_CACHE_BYTES))); + diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c index 3486fe7d6e65..627785c48ea9 100644 --- a/arch/ia64/kernel/time.c +++ b/arch/ia64/kernel/time.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -28,6 +29,16 @@ #include #include +#include "fsyscall_gtod_data.h" + +static cycle_t itc_get_cycles(void); + +struct fsyscall_gtod_data_t fsyscall_gtod_data = { + .lock = SEQLOCK_UNLOCKED, +}; + +struct itc_jitter_data_t itc_jitter_data; + volatile int time_keeper_id = 0; /* smp_processor_id() of time-keeper */ #ifdef CONFIG_IA64_DEBUG_IRQ @@ -37,11 +48,16 @@ EXPORT_SYMBOL(last_cli_ip); #endif -static struct time_interpolator itc_interpolator = { - .shift = 16, - .mask = 0xffffffffffffffffLL, - .source = TIME_SOURCE_CPU +static struct clocksource clocksource_itc = { + .name = "itc", + .rating = 350, + .read = itc_get_cycles, + .mask = 0xffffffffffffffff, + .mult = 0, /*to be caluclated*/ + .shift = 16, + .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; +static struct clocksource *itc_clocksource; static irqreturn_t timer_interrupt (int irq, void *dev_id) @@ -210,8 +226,6 @@ ia64_init_itm (void) + itc_freq/2)/itc_freq; if (!(sal_platform_features & IA64_SAL_PLATFORM_FEATURE_ITC_DRIFT)) { - itc_interpolator.frequency = local_cpu_data->itc_freq; - itc_interpolator.drift = itc_drift; #ifdef CONFIG_SMP /* On IA64 in an SMP configuration ITCs are never accurately synchronized. * Jitter compensation requires a cmpxchg which may limit @@ -223,15 +237,50 @@ ia64_init_itm (void) * even going backward) if the ITC offsets between the individual CPUs * are too large. */ - if (!nojitter) itc_interpolator.jitter = 1; + if (!nojitter) + itc_jitter_data.itc_jitter = 1; #endif - register_time_interpolator(&itc_interpolator); } /* Setup the CPU local timer tick */ ia64_cpu_local_tick(); + + if (!itc_clocksource) { + /* Sort out mult/shift values: */ + clocksource_itc.mult = + clocksource_hz2mult(local_cpu_data->itc_freq, + clocksource_itc.shift); + clocksource_register(&clocksource_itc); + itc_clocksource = &clocksource_itc; + } } +static cycle_t itc_get_cycles() +{ + u64 lcycle, now, ret; + + if (!itc_jitter_data.itc_jitter) + return get_cycles(); + + lcycle = itc_jitter_data.itc_lastcycle; + now = get_cycles(); + if (lcycle && time_after(lcycle, now)) + return lcycle; + + /* + * Keep track of the last timer value returned. + * In an SMP environment, you could lose out in contention of + * cmpxchg. If so, your cmpxchg returns new value which the + * winner of contention updated to. Use the new value instead. + */ + ret = cmpxchg(&itc_jitter_data.itc_lastcycle, lcycle, now); + if (unlikely(ret != lcycle)) + return ret; + + return now; +} + + static struct irqaction timer_irqaction = { .handler = timer_interrupt, .flags = IRQF_DISABLED | IRQF_IRQPOLL, @@ -307,3 +356,34 @@ ia64_setup_printk_clock(void) if (!(sal_platform_features & IA64_SAL_PLATFORM_FEATURE_ITC_DRIFT)) ia64_printk_clock = ia64_itc_printk_clock; } + +void update_vsyscall(struct timespec *wall, struct clocksource *c) +{ + unsigned long flags; + + write_seqlock_irqsave(&fsyscall_gtod_data.lock, flags); + + /* copy fsyscall clock data */ + fsyscall_gtod_data.clk_mask = c->mask; + fsyscall_gtod_data.clk_mult = c->mult; + fsyscall_gtod_data.clk_shift = c->shift; + fsyscall_gtod_data.clk_fsys_mmio = c->fsys_mmio; + fsyscall_gtod_data.clk_cycle_last = c->cycle_last; + + /* copy kernel time structures */ + fsyscall_gtod_data.wall_time.tv_sec = wall->tv_sec; + fsyscall_gtod_data.wall_time.tv_nsec = wall->tv_nsec; + fsyscall_gtod_data.monotonic_time.tv_sec = wall_to_monotonic.tv_sec + + wall->tv_sec; + fsyscall_gtod_data.monotonic_time.tv_nsec = wall_to_monotonic.tv_nsec + + wall->tv_nsec; + + /* normalize */ + while (fsyscall_gtod_data.monotonic_time.tv_nsec >= NSEC_PER_SEC) { + fsyscall_gtod_data.monotonic_time.tv_nsec -= NSEC_PER_SEC; + fsyscall_gtod_data.monotonic_time.tv_sec++; + } + + write_sequnlock_irqrestore(&fsyscall_gtod_data.lock, flags); +} + diff --git a/arch/ia64/sn/kernel/sn2/timer.c b/arch/ia64/sn/kernel/sn2/timer.c index 56a88b6df4b4..19e25d2b64fc 100644 --- a/arch/ia64/sn/kernel/sn2/timer.c +++ b/arch/ia64/sn/kernel/sn2/timer.c @@ -11,6 +11,7 @@ #include #include #include +#include #include #include @@ -22,11 +23,21 @@ extern unsigned long sn_rtc_cycles_per_second; -static struct time_interpolator sn2_interpolator = { - .drift = -1, - .shift = 10, - .mask = (1LL << 55) - 1, - .source = TIME_SOURCE_MMIO64 +static void __iomem *sn2_mc; + +static cycle_t read_sn2(void) +{ + return (cycle_t)readq(sn2_mc); +} + +static struct clocksource clocksource_sn2 = { + .name = "sn2_rtc", + .rating = 300, + .read = read_sn2, + .mask = (1LL << 55) - 1, + .mult = 0, + .shift = 10, + .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; /* @@ -47,9 +58,11 @@ ia64_sn_udelay (unsigned long usecs) void __init sn_timer_init(void) { - sn2_interpolator.frequency = sn_rtc_cycles_per_second; - sn2_interpolator.addr = RTC_COUNTER_ADDR; - register_time_interpolator(&sn2_interpolator); + sn2_mc = RTC_COUNTER_ADDR; + clocksource_sn2.fsys_mmio = RTC_COUNTER_ADDR; + clocksource_sn2.mult = clocksource_hz2mult(sn_rtc_cycles_per_second, + clocksource_sn2.shift); + clocksource_register(&clocksource_sn2); ia64_udelay = &ia64_sn_udelay; } diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index 80ffc7829916..bb5d23be4260 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -475,7 +475,7 @@ static void acpi_processor_idle(void) /* Get end time (ticks) */ t2 = inl(acpi_gbl_FADT.xpm_timer_block.address); -#ifdef CONFIG_GENERIC_TIME +#if defined (CONFIG_GENERIC_TIME) && defined (CONFIG_X86_TSC) /* TSC halts in C2, so notify users */ mark_tsc_unstable("possible TSC halt in C2"); #endif @@ -517,7 +517,7 @@ static void acpi_processor_idle(void) acpi_set_register(ACPI_BITREG_ARB_DISABLE, 0); } -#ifdef CONFIG_GENERIC_TIME +#if defined (CONFIG_GENERIC_TIME) && defined (CONFIG_X86_TSC) /* TSC halts in C3, so notify users */ mark_tsc_unstable("TSC halts in C3"); #endif diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c index 0be700f4e8fd..ba0e74ad74bb 100644 --- a/drivers/char/hpet.c +++ b/drivers/char/hpet.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include @@ -51,8 +52,34 @@ #define HPET_RANGE_SIZE 1024 /* from HPET spec */ +#if BITS_PER_LONG == 64 +#define write_counter(V, MC) writeq(V, MC) +#define read_counter(MC) readq(MC) +#else +#define write_counter(V, MC) writel(V, MC) +#define read_counter(MC) readl(MC) +#endif + static u32 hpet_nhpet, hpet_max_freq = HPET_USER_FREQ; +static void __iomem *hpet_mctr; + +static cycle_t read_hpet(void) +{ + return (cycle_t)read_counter((void __iomem *)hpet_mctr); +} + +static struct clocksource clocksource_hpet = { + .name = "hpet", + .rating = 250, + .read = read_hpet, + .mask = 0xffffffffffffffff, + .mult = 0, /*to be caluclated*/ + .shift = 10, + .flags = CLOCK_SOURCE_IS_CONTINUOUS, +}; +static struct clocksource *hpet_clocksource; + /* A lock for concurrent access by app and isr hpet activity. */ static DEFINE_SPINLOCK(hpet_lock); /* A lock for concurrent intermodule access to hpet and isr hpet activity. */ @@ -79,7 +106,7 @@ struct hpets { struct hpets *hp_next; struct hpet __iomem *hp_hpet; unsigned long hp_hpet_phys; - struct time_interpolator *hp_interpolator; + struct clocksource *hp_clocksource; unsigned long long hp_tick_freq; unsigned long hp_delta; unsigned int hp_ntimer; @@ -94,13 +121,6 @@ static struct hpets *hpets; #define HPET_PERIODIC 0x0004 #define HPET_SHARED_IRQ 0x0008 -#if BITS_PER_LONG == 64 -#define write_counter(V, MC) writeq(V, MC) -#define read_counter(MC) readq(MC) -#else -#define write_counter(V, MC) writel(V, MC) -#define read_counter(MC) readl(MC) -#endif #ifndef readq static inline unsigned long long readq(void __iomem *addr) @@ -737,27 +757,6 @@ static ctl_table dev_root[] = { static struct ctl_table_header *sysctl_header; -static void hpet_register_interpolator(struct hpets *hpetp) -{ -#ifdef CONFIG_TIME_INTERPOLATION - struct time_interpolator *ti; - - ti = kzalloc(sizeof(*ti), GFP_KERNEL); - if (!ti) - return; - - ti->source = TIME_SOURCE_MMIO64; - ti->shift = 10; - ti->addr = &hpetp->hp_hpet->hpet_mc; - ti->frequency = hpetp->hp_tick_freq; - ti->drift = HPET_DRIFT; - ti->mask = -1; - - hpetp->hp_interpolator = ti; - register_time_interpolator(ti); -#endif -} - /* * Adjustment for when arming the timer with * initial conditions. That is, main counter @@ -909,7 +908,16 @@ int hpet_alloc(struct hpet_data *hdp) } hpetp->hp_delta = hpet_calibrate(hpetp); - hpet_register_interpolator(hpetp); + + if (!hpet_clocksource) { + hpet_mctr = (void __iomem *)&hpetp->hp_hpet->hpet_mc; + CLKSRC_FSYS_MMIO_SET(clocksource_hpet.fsys_mmio, hpet_mctr); + clocksource_hpet.mult = clocksource_hz2mult(hpetp->hp_tick_freq, + clocksource_hpet.shift); + clocksource_register(&clocksource_hpet); + hpetp->hp_clocksource = &clocksource_hpet; + hpet_clocksource = &clocksource_hpet; + } return 0; } @@ -995,7 +1003,7 @@ static int hpet_acpi_add(struct acpi_device *device) static int hpet_acpi_remove(struct acpi_device *device, int type) { - /* XXX need to unregister interpolator, dealloc mem, etc */ + /* XXX need to unregister clocksource, dealloc mem, etc */ return -EINVAL; } diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h index bf297b03a4e4..16ea3374dddf 100644 --- a/include/linux/clocksource.h +++ b/include/linux/clocksource.h @@ -67,6 +67,12 @@ struct clocksource { unsigned long flags; cycle_t (*vread)(void); void (*resume)(void); +#ifdef CONFIG_IA64 + void *fsys_mmio; /* used by fsyscall asm code */ +#define CLKSRC_FSYS_MMIO_SET(mmio, addr) ((mmio) = (addr)) +#else +#define CLKSRC_FSYS_MMIO_SET(mmio, addr) do { } while (0) +#endif /* timekeeping specific data, ignore */ cycle_t cycle_interval; -- cgit v1.2.3 From 1f564ad6d4182859612cbae452122e5eb2d62a76 Mon Sep 17 00:00:00 2001 From: Bob Picco Date: Wed, 18 Jul 2007 15:51:28 -0700 Subject: [IA64] remove time interpolator Remove time_interpolator code (This is generic code, but only user was ia64. It has been superseded by the CONFIG_GENERIC_TIME code). Signed-off-by: Bob Picco Signed-off-by: John Stultz Signed-off-by: Peter Keilty Signed-off-by: Tony Luck --- Documentation/time_interpolators.txt | 41 -------- include/linux/timex.h | 60 ----------- kernel/time.c | 88 ---------------- kernel/time/ntp.c | 10 -- kernel/time/timekeeping.c | 4 - kernel/timer.c | 188 ----------------------------------- 6 files changed, 391 deletions(-) delete mode 100644 Documentation/time_interpolators.txt (limited to 'include') diff --git a/Documentation/time_interpolators.txt b/Documentation/time_interpolators.txt deleted file mode 100644 index e3b60854fbc2..000000000000 --- a/Documentation/time_interpolators.txt +++ /dev/null @@ -1,41 +0,0 @@ -Time Interpolators ------------------- - -Time interpolators are a base of time calculation between timer ticks and -allow an accurate determination of time down to the accuracy of the time -source in nanoseconds. - -The architecture specific code typically provides gettimeofday and -settimeofday under Linux. The time interpolator provides both if an arch -defines CONFIG_TIME_INTERPOLATION. The arch still must set up timer tick -operations and call the necessary functions to advance the clock. - -With the time interpolator a standardized interface exists for time -interpolation between ticks. The provided logic is highly scalable -and has been tested in SMP situations of up to 512 CPUs. - -If CONFIG_TIME_INTERPOLATION is defined then the architecture specific code -(or the device drivers - like HPET) may register time interpolators. -These are typically defined in the following way: - -static struct time_interpolator my_interpolator { - .frequency = MY_FREQUENCY, - .source = TIME_SOURCE_MMIO32, - .shift = 8, /* scaling for higher accuracy */ - .drift = -1, /* Unknown drift */ - .jitter = 0 /* time source is stable */ -}; - -void time_init(void) -{ - .... - /* Initialization of the timer *. - my_interpolator.address = &my_timer; - register_time_interpolator(&my_interpolator); - .... -} - -For more details see include/linux/timex.h and kernel/timer.c. - -Christoph Lameter , October 31, 2004 - diff --git a/include/linux/timex.h b/include/linux/timex.h index da929dbbea2a..37ac3ff90faf 100644 --- a/include/linux/timex.h +++ b/include/linux/timex.h @@ -224,66 +224,6 @@ static inline int ntp_synced(void) __x < 0 ? -(-__x >> __s) : __x >> __s; \ }) - -#ifdef CONFIG_TIME_INTERPOLATION - -#define TIME_SOURCE_CPU 0 -#define TIME_SOURCE_MMIO64 1 -#define TIME_SOURCE_MMIO32 2 -#define TIME_SOURCE_FUNCTION 3 - -/* For proper operations time_interpolator clocks must run slightly slower - * than the standard clock since the interpolator may only correct by having - * time jump forward during a tick. A slower clock is usually a side effect - * of the integer divide of the nanoseconds in a second by the frequency. - * The accuracy of the division can be increased by specifying a shift. - * However, this may cause the clock not to be slow enough. - * The interpolator will self-tune the clock by slowing down if no - * resets occur or speeding up if the time jumps per analysis cycle - * become too high. - * - * Setting jitter compensates for a fluctuating timesource by comparing - * to the last value read from the timesource to insure that an earlier value - * is not returned by a later call. The price to pay - * for the compensation is that the timer routines are not as scalable anymore. - */ - -struct time_interpolator { - u16 source; /* time source flags */ - u8 shift; /* increases accuracy of multiply by shifting. */ - /* Note that bits may be lost if shift is set too high */ - u8 jitter; /* if set compensate for fluctuations */ - u32 nsec_per_cyc; /* set by register_time_interpolator() */ - void *addr; /* address of counter or function */ - cycles_t mask; /* mask the valid bits of the counter */ - unsigned long offset; /* nsec offset at last update of interpolator */ - u64 last_counter; /* counter value in units of the counter at last update */ - cycles_t last_cycle; /* Last timer value if TIME_SOURCE_JITTER is set */ - u64 frequency; /* frequency in counts/second */ - long drift; /* drift in parts-per-million (or -1) */ - unsigned long skips; /* skips forward */ - unsigned long ns_skipped; /* nanoseconds skipped */ - struct time_interpolator *next; -}; - -extern void register_time_interpolator(struct time_interpolator *); -extern void unregister_time_interpolator(struct time_interpolator *); -extern void time_interpolator_reset(void); -extern unsigned long time_interpolator_get_offset(void); -extern void time_interpolator_update(long delta_nsec); - -#else /* !CONFIG_TIME_INTERPOLATION */ - -static inline void time_interpolator_reset(void) -{ -} - -static inline void time_interpolator_update(long delta_nsec) -{ -} - -#endif /* !CONFIG_TIME_INTERPOLATION */ - #define TICK_LENGTH_SHIFT 32 #ifdef CONFIG_NO_HZ diff --git a/kernel/time.c b/kernel/time.c index ffe19149d770..e325597f5bf5 100644 --- a/kernel/time.c +++ b/kernel/time.c @@ -136,7 +136,6 @@ static inline void warp_clock(void) write_seqlock_irq(&xtime_lock); wall_to_monotonic.tv_sec -= sys_tz.tz_minuteswest * 60; xtime.tv_sec += sys_tz.tz_minuteswest * 60; - time_interpolator_reset(); write_sequnlock_irq(&xtime_lock); clock_was_set(); } @@ -309,92 +308,6 @@ struct timespec timespec_trunc(struct timespec t, unsigned gran) } EXPORT_SYMBOL(timespec_trunc); -#ifdef CONFIG_TIME_INTERPOLATION -void getnstimeofday (struct timespec *tv) -{ - unsigned long seq,sec,nsec; - - do { - seq = read_seqbegin(&xtime_lock); - sec = xtime.tv_sec; - nsec = xtime.tv_nsec+time_interpolator_get_offset(); - } while (unlikely(read_seqretry(&xtime_lock, seq))); - - while (unlikely(nsec >= NSEC_PER_SEC)) { - nsec -= NSEC_PER_SEC; - ++sec; - } - tv->tv_sec = sec; - tv->tv_nsec = nsec; -} -EXPORT_SYMBOL_GPL(getnstimeofday); - -int do_settimeofday (struct timespec *tv) -{ - time_t wtm_sec, sec = tv->tv_sec; - long wtm_nsec, nsec = tv->tv_nsec; - - if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC) - return -EINVAL; - - write_seqlock_irq(&xtime_lock); - { - wtm_sec = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec); - wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec); - - set_normalized_timespec(&xtime, sec, nsec); - set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec); - - time_adjust = 0; /* stop active adjtime() */ - time_status |= STA_UNSYNC; - time_maxerror = NTP_PHASE_LIMIT; - time_esterror = NTP_PHASE_LIMIT; - time_interpolator_reset(); - } - write_sequnlock_irq(&xtime_lock); - clock_was_set(); - return 0; -} -EXPORT_SYMBOL(do_settimeofday); - -void do_gettimeofday (struct timeval *tv) -{ - unsigned long seq, nsec, usec, sec, offset; - do { - seq = read_seqbegin(&xtime_lock); - offset = time_interpolator_get_offset(); - sec = xtime.tv_sec; - nsec = xtime.tv_nsec; - } while (unlikely(read_seqretry(&xtime_lock, seq))); - - usec = (nsec + offset) / 1000; - - while (unlikely(usec >= USEC_PER_SEC)) { - usec -= USEC_PER_SEC; - ++sec; - } - - tv->tv_sec = sec; - tv->tv_usec = usec; - - /* - * Make sure xtime.tv_sec [returned by sys_time()] always - * follows the gettimeofday() result precisely. This - * condition is extremely unlikely, it can hit at most - * once per second: - */ - if (unlikely(xtime.tv_sec != tv->tv_sec)) { - unsigned long flags; - - write_seqlock_irqsave(&xtime_lock, flags); - update_wall_time(); - write_sequnlock_irqrestore(&xtime_lock, flags); - } -} -EXPORT_SYMBOL(do_gettimeofday); - -#else /* CONFIG_TIME_INTERPOLATION */ - #ifndef CONFIG_GENERIC_TIME /* * Simulate gettimeofday using do_gettimeofday which only allows a timeval @@ -410,7 +323,6 @@ void getnstimeofday(struct timespec *tv) } EXPORT_SYMBOL_GPL(getnstimeofday); #endif -#endif /* CONFIG_TIME_INTERPOLATION */ /* Converts Gregorian date to seconds since 1970-01-01 00:00:00. * Assumes input in normal date format, i.e. 1980-12-31 23:59:59 diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index 438c6b723ee2..b5e352597cbb 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -116,11 +116,6 @@ void second_overflow(void) if (xtime.tv_sec % 86400 == 0) { xtime.tv_sec--; wall_to_monotonic.tv_sec++; - /* - * The timer interpolator will make time change - * gradually instead of an immediate jump by one second - */ - time_interpolator_update(-NSEC_PER_SEC); time_state = TIME_OOP; printk(KERN_NOTICE "Clock: inserting leap second " "23:59:60 UTC\n"); @@ -130,11 +125,6 @@ void second_overflow(void) if ((xtime.tv_sec + 1) % 86400 == 0) { xtime.tv_sec++; wall_to_monotonic.tv_sec--; - /* - * Use of time interpolator for a gradual change of - * time - */ - time_interpolator_update(NSEC_PER_SEC); time_state = TIME_WAIT; printk(KERN_NOTICE "Clock: deleting leap second " "23:59:59 UTC\n"); diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 728cedfd3cbd..027d46c906e0 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -466,10 +466,6 @@ void update_wall_time(void) second_overflow(); } - /* interpolator bits */ - time_interpolator_update(clock->xtime_interval - >> clock->shift); - /* accumulate error between NTP and clock interval */ clock->error += current_tick_length(); clock->error -= clock->xtime_interval << (TICK_LENGTH_SHIFT - clock->shift); diff --git a/kernel/timer.c b/kernel/timer.c index b7792fb03387..dbc03ab14eed 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -1349,194 +1349,6 @@ void __init init_timers(void) open_softirq(TIMER_SOFTIRQ, run_timer_softirq, NULL); } -#ifdef CONFIG_TIME_INTERPOLATION - -struct time_interpolator *time_interpolator __read_mostly; -static struct time_interpolator *time_interpolator_list __read_mostly; -static DEFINE_SPINLOCK(time_interpolator_lock); - -static inline cycles_t time_interpolator_get_cycles(unsigned int src) -{ - unsigned long (*x)(void); - - switch (src) - { - case TIME_SOURCE_FUNCTION: - x = time_interpolator->addr; - return x(); - - case TIME_SOURCE_MMIO64 : - return readq_relaxed((void __iomem *)time_interpolator->addr); - - case TIME_SOURCE_MMIO32 : - return readl_relaxed((void __iomem *)time_interpolator->addr); - - default: return get_cycles(); - } -} - -static inline u64 time_interpolator_get_counter(int writelock) -{ - unsigned int src = time_interpolator->source; - - if (time_interpolator->jitter) - { - cycles_t lcycle; - cycles_t now; - - do { - lcycle = time_interpolator->last_cycle; - now = time_interpolator_get_cycles(src); - if (lcycle && time_after(lcycle, now)) - return lcycle; - - /* When holding the xtime write lock, there's no need - * to add the overhead of the cmpxchg. Readers are - * force to retry until the write lock is released. - */ - if (writelock) { - time_interpolator->last_cycle = now; - return now; - } - /* Keep track of the last timer value returned. The use of cmpxchg here - * will cause contention in an SMP environment. - */ - } while (unlikely(cmpxchg(&time_interpolator->last_cycle, lcycle, now) != lcycle)); - return now; - } - else - return time_interpolator_get_cycles(src); -} - -void time_interpolator_reset(void) -{ - time_interpolator->offset = 0; - time_interpolator->last_counter = time_interpolator_get_counter(1); -} - -#define GET_TI_NSECS(count,i) (((((count) - i->last_counter) & (i)->mask) * (i)->nsec_per_cyc) >> (i)->shift) - -unsigned long time_interpolator_get_offset(void) -{ - /* If we do not have a time interpolator set up then just return zero */ - if (!time_interpolator) - return 0; - - return time_interpolator->offset + - GET_TI_NSECS(time_interpolator_get_counter(0), time_interpolator); -} - -#define INTERPOLATOR_ADJUST 65536 -#define INTERPOLATOR_MAX_SKIP 10*INTERPOLATOR_ADJUST - -void time_interpolator_update(long delta_nsec) -{ - u64 counter; - unsigned long offset; - - /* If there is no time interpolator set up then do nothing */ - if (!time_interpolator) - return; - - /* - * The interpolator compensates for late ticks by accumulating the late - * time in time_interpolator->offset. A tick earlier than expected will - * lead to a reset of the offset and a corresponding jump of the clock - * forward. Again this only works if the interpolator clock is running - * slightly slower than the regular clock and the tuning logic insures - * that. - */ - - counter = time_interpolator_get_counter(1); - offset = time_interpolator->offset + - GET_TI_NSECS(counter, time_interpolator); - - if (delta_nsec < 0 || (unsigned long) delta_nsec < offset) - time_interpolator->offset = offset - delta_nsec; - else { - time_interpolator->skips++; - time_interpolator->ns_skipped += delta_nsec - offset; - time_interpolator->offset = 0; - } - time_interpolator->last_counter = counter; - - /* Tuning logic for time interpolator invoked every minute or so. - * Decrease interpolator clock speed if no skips occurred and an offset is carried. - * Increase interpolator clock speed if we skip too much time. - */ - if (jiffies % INTERPOLATOR_ADJUST == 0) - { - if (time_interpolator->skips == 0 && time_interpolator->offset > tick_nsec) - time_interpolator->nsec_per_cyc--; - if (time_interpolator->ns_skipped > INTERPOLATOR_MAX_SKIP && time_interpolator->offset == 0) - time_interpolator->nsec_per_cyc++; - time_interpolator->skips = 0; - time_interpolator->ns_skipped = 0; - } -} - -static inline int -is_better_time_interpolator(struct time_interpolator *new) -{ - if (!time_interpolator) - return 1; - return new->frequency > 2*time_interpolator->frequency || - (unsigned long)new->drift < (unsigned long)time_interpolator->drift; -} - -void -register_time_interpolator(struct time_interpolator *ti) -{ - unsigned long flags; - - /* Sanity check */ - BUG_ON(ti->frequency == 0 || ti->mask == 0); - - ti->nsec_per_cyc = ((u64)NSEC_PER_SEC << ti->shift) / ti->frequency; - spin_lock(&time_interpolator_lock); - write_seqlock_irqsave(&xtime_lock, flags); - if (is_better_time_interpolator(ti)) { - time_interpolator = ti; - time_interpolator_reset(); - } - write_sequnlock_irqrestore(&xtime_lock, flags); - - ti->next = time_interpolator_list; - time_interpolator_list = ti; - spin_unlock(&time_interpolator_lock); -} - -void -unregister_time_interpolator(struct time_interpolator *ti) -{ - struct time_interpolator *curr, **prev; - unsigned long flags; - - spin_lock(&time_interpolator_lock); - prev = &time_interpolator_list; - for (curr = *prev; curr; curr = curr->next) { - if (curr == ti) { - *prev = curr->next; - break; - } - prev = &curr->next; - } - - write_seqlock_irqsave(&xtime_lock, flags); - if (ti == time_interpolator) { - /* we lost the best time-interpolator: */ - time_interpolator = NULL; - /* find the next-best interpolator */ - for (curr = time_interpolator_list; curr; curr = curr->next) - if (is_better_time_interpolator(curr)) - time_interpolator = curr; - time_interpolator_reset(); - } - write_sequnlock_irqrestore(&xtime_lock, flags); - spin_unlock(&time_interpolator_lock); -} -#endif /* CONFIG_TIME_INTERPOLATION */ - /** * msleep - sleep safely even with waitqueue interruptions * @msecs: Time in milliseconds to sleep for -- cgit v1.2.3 From bd807f9c5b90c236c3d173a5815d620bc1b1316d Mon Sep 17 00:00:00 2001 From: "Robert P. J. Day" Date: Wed, 18 Jul 2007 07:33:08 -0400 Subject: [IA64] Prevent people from directly including . Signed-off-by: Robert P. J. Day Signed-off-by: Tony Luck --- include/asm-ia64/rwsem.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include') diff --git a/include/asm-ia64/rwsem.h b/include/asm-ia64/rwsem.h index 2d1640cc240a..8aba06a7b038 100644 --- a/include/asm-ia64/rwsem.h +++ b/include/asm-ia64/rwsem.h @@ -21,6 +21,10 @@ #ifndef _ASM_IA64_RWSEM_H #define _ASM_IA64_RWSEM_H +#ifndef _LINUX_RWSEM_H +#error "Please don't include directly, use instead." +#endif + #include #include -- cgit v1.2.3