1 files changed, 383 insertions, 248 deletions

diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c
index 1184210e6d0c..d7f0993b8056 100644
--- a/arch/x86/kernel/io_apic.c
+++ b/arch/x86/kernel/io_apic.c
@@ -108,8 +108,33 @@ static int __init parse_noapic(char *str)
 early_param("noapic", parse_noapic);
 
 struct irq_pin_list;
+
+/*
+ * This is performance-critical, we want to do it O(1)
+ *
+ * the indexing order of this array favors 1:1 mappings
+ * between pins and IRQs.
+ */
+
+struct irq_pin_list {
+        int apic, pin;
+        struct irq_pin_list *next;
+};
+
+static struct irq_pin_list *get_one_free_irq_2_pin(int cpu)
+{
+        struct irq_pin_list *pin;
+        int node;
+
+        node = cpu_to_node(cpu);
+
+        pin = kzalloc_node(sizeof(*pin), GFP_ATOMIC, node);
+        printk(KERN_DEBUG "  alloc irq_2_pin on cpu %d node %d\n", cpu, node);
+
+        return pin;
+}
+
 struct irq_cfg {
-        unsigned int irq;
         struct irq_pin_list *irq_2_pin;
         cpumask_t domain;
         cpumask_t old_domain;
@@ -119,81 +144,95 @@ struct irq_cfg {
 };
 
 /* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */
+#ifdef CONFIG_SPARSE_IRQ
+static struct irq_cfg irq_cfgx[] = {
+#else
 static struct irq_cfg irq_cfgx[NR_IRQS] = {
-        [0]  = { .irq =  0, .domain = CPU_MASK_ALL, .vector = IRQ0_VECTOR,  },
-        [1]  = { .irq =  1, .domain = CPU_MASK_ALL, .vector = IRQ1_VECTOR,  },
-        [2]  = { .irq =  2, .domain = CPU_MASK_ALL, .vector = IRQ2_VECTOR,  },
-        [3]  = { .irq =  3, .domain = CPU_MASK_ALL, .vector = IRQ3_VECTOR,  },
-        [4]  = { .irq =  4, .domain = CPU_MASK_ALL, .vector = IRQ4_VECTOR,  },
-        [5]  = { .irq =  5, .domain = CPU_MASK_ALL, .vector = IRQ5_VECTOR,  },
-        [6]  = { .irq =  6, .domain = CPU_MASK_ALL, .vector = IRQ6_VECTOR,  },
-        [7]  = { .irq =  7, .domain = CPU_MASK_ALL, .vector = IRQ7_VECTOR,  },
-        [8]  = { .irq =  8, .domain = CPU_MASK_ALL, .vector = IRQ8_VECTOR,  },
-        [9]  = { .irq =  9, .domain = CPU_MASK_ALL, .vector = IRQ9_VECTOR,  },
-        [10] = { .irq = 10, .domain = CPU_MASK_ALL, .vector = IRQ10_VECTOR, },
-        [11] = { .irq = 11, .domain = CPU_MASK_ALL, .vector = IRQ11_VECTOR, },
-        [12] = { .irq = 12, .domain = CPU_MASK_ALL, .vector = IRQ12_VECTOR, },
-        [13] = { .irq = 13, .domain = CPU_MASK_ALL, .vector = IRQ13_VECTOR, },
-        [14] = { .irq = 14, .domain = CPU_MASK_ALL, .vector = IRQ14_VECTOR, },
-        [15] = { .irq = 15, .domain = CPU_MASK_ALL, .vector = IRQ15_VECTOR, },
+#endif
+        [0]  = { .domain = CPU_MASK_ALL, .vector = IRQ0_VECTOR,  },
+        [1]  = { .domain = CPU_MASK_ALL, .vector = IRQ1_VECTOR,  },
+        [2]  = { .domain = CPU_MASK_ALL, .vector = IRQ2_VECTOR,  },
+        [3]  = { .domain = CPU_MASK_ALL, .vector = IRQ3_VECTOR,  },
+        [4]  = { .domain = CPU_MASK_ALL, .vector = IRQ4_VECTOR,  },
+        [5]  = { .domain = CPU_MASK_ALL, .vector = IRQ5_VECTOR,  },
+        [6]  = { .domain = CPU_MASK_ALL, .vector = IRQ6_VECTOR,  },
+        [7]  = { .domain = CPU_MASK_ALL, .vector = IRQ7_VECTOR,  },
+        [8]  = { .domain = CPU_MASK_ALL, .vector = IRQ8_VECTOR,  },
+        [9]  = { .domain = CPU_MASK_ALL, .vector = IRQ9_VECTOR,  },
+        [10] = { .domain = CPU_MASK_ALL, .vector = IRQ10_VECTOR, },
+        [11] = { .domain = CPU_MASK_ALL, .vector = IRQ11_VECTOR, },
+        [12] = { .domain = CPU_MASK_ALL, .vector = IRQ12_VECTOR, },
+        [13] = { .domain = CPU_MASK_ALL, .vector = IRQ13_VECTOR, },
+        [14] = { .domain = CPU_MASK_ALL, .vector = IRQ14_VECTOR, },
+        [15] = { .domain = CPU_MASK_ALL, .vector = IRQ15_VECTOR, },
 };
 
-#define for_each_irq_cfg(irq, cfg)              \
-        for (irq = 0, cfg = irq_cfgx; irq < nr_irqs; irq++, cfg++)
-
-static struct irq_cfg *irq_cfg(unsigned int irq)
+void __init arch_early_irq_init(void)
 {
-        return irq < nr_irqs ? irq_cfgx + irq : NULL;
+        struct irq_cfg *cfg;
+        struct irq_desc *desc;
+        int count;
+        int i;
+
+        cfg = irq_cfgx;
+        count = ARRAY_SIZE(irq_cfgx);
+
+        for (i = 0; i < count; i++) {
+                desc = irq_to_desc(i);
+                desc->chip_data = &cfg[i];
+        }
 }
 
-static struct irq_cfg *irq_cfg_alloc(unsigned int irq)
+#ifdef CONFIG_SPARSE_IRQ
+static struct irq_cfg *irq_cfg(unsigned int irq)
 {
-        return irq_cfg(irq);
+        struct irq_cfg *cfg = NULL;
+        struct irq_desc *desc;
+
+        desc = irq_to_desc(irq);
+        if (desc)
+                cfg = desc->chip_data;
+
+        return cfg;
 }
 
-/*
- * Rough estimation of how many shared IRQs there are, can be changed
- * anytime.
- */
-#define MAX_PLUS_SHARED_IRQS NR_IRQS
-#define PIN_MAP_SIZE (MAX_PLUS_SHARED_IRQS + NR_IRQS)
+static struct irq_cfg *get_one_free_irq_cfg(int cpu)
+{
+        struct irq_cfg *cfg;
+        int node;
 
-/*
- * This is performance-critical, we want to do it O(1)
- *
- * the indexing order of this array favors 1:1 mappings
- * between pins and IRQs.
- */
+        node = cpu_to_node(cpu);
 
-struct irq_pin_list {
-        int apic, pin;
-        struct irq_pin_list *next;
-};
+        cfg = kzalloc_node(sizeof(*cfg), GFP_ATOMIC, node);
+        printk(KERN_DEBUG "  alloc irq_cfg on cpu %d node %d\n", cpu, node);
 
-static struct irq_pin_list irq_2_pin_head[PIN_MAP_SIZE];
-static struct irq_pin_list *irq_2_pin_ptr;
+        return cfg;
+}
 
-static void __init irq_2_pin_init(void)
+void arch_init_chip_data(struct irq_desc *desc, int cpu)
 {
-        struct irq_pin_list *pin = irq_2_pin_head;
-        int i;
-
-        for (i = 1; i < PIN_MAP_SIZE; i++)
-                pin[i-1].next = &pin[i];
+        struct irq_cfg *cfg;
 
-        irq_2_pin_ptr = &pin[0];
+        cfg = desc->chip_data;
+        if (!cfg) {
+                desc->chip_data = get_one_free_irq_cfg(cpu);
+                if (!desc->chip_data) {
+                        printk(KERN_ERR "can not alloc irq_cfg\n");
+                        BUG_ON(1);
+                }
+        }
 }
 
-static struct irq_pin_list *get_one_free_irq_2_pin(void)
+#else
+static struct irq_cfg *irq_cfg(unsigned int irq)
 {
-        struct irq_pin_list *pin = irq_2_pin_ptr;
+        return irq < nr_irqs ? irq_cfgx + irq : NULL;
+}
 
-        if (!pin)
-                panic("can not get more irq_2_pin\n");
+#endif
 
-        irq_2_pin_ptr = pin->next;
-        pin->next = NULL;
-        return pin;
+static inline void set_extra_move_desc(struct irq_desc *desc, cpumask_t mask)
+{
 }
 
 struct io_apic {
@@ -237,11 +276,10 @@ static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned
         writel(value, &io_apic->data);
 }
 
-static bool io_apic_level_ack_pending(unsigned int irq)
+static bool io_apic_level_ack_pending(struct irq_cfg *cfg)
 {
         struct irq_pin_list *entry;
         unsigned long flags;
-        struct irq_cfg *cfg = irq_cfg(irq);
 
         spin_lock_irqsave(&ioapic_lock, flags);
         entry = cfg->irq_2_pin;
@@ -323,13 +361,12 @@ static void ioapic_mask_entry(int apic, int pin)
 }
 
 #ifdef CONFIG_SMP
-static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, u8 vector)
+static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq_cfg *cfg)
 {
         int apic, pin;
-        struct irq_cfg *cfg;
         struct irq_pin_list *entry;
+        u8 vector = cfg->vector;
 
-        cfg = irq_cfg(irq);
         entry = cfg->irq_2_pin;
         for (;;) {
                 unsigned int reg;
@@ -359,24 +396,27 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, u8 vector)
         }
 }
 
-static int assign_irq_vector(int irq, cpumask_t mask);
+static int assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask);
 
-static void set_ioapic_affinity_irq(unsigned int irq,
-                                    const struct cpumask *mask)
+static void set_ioapic_affinity_irq_desc(struct irq_desc *desc,
+                                         const struct cpumask *mask)
 {
         struct irq_cfg *cfg;
         unsigned long flags;
         unsigned int dest;
         cpumask_t tmp;
-        struct irq_desc *desc;
+        unsigned int irq;
 
         if (!cpumask_intersects(mask, cpu_online_mask))
                 return;
 
-        cfg = irq_cfg(irq);
-        if (assign_irq_vector(irq, *mask))
+        irq = desc->irq;
+        cfg = desc->chip_data;
+        if (assign_irq_vector(irq, cfg, *mask))
                 return;
 
+        set_extra_move_desc(desc, *mask);
+
         cpumask_and(&tmp, &cfg->domain, mask);
         dest = cpu_mask_to_apicid(tmp);
         /*
@@ -384,12 +424,21 @@ static void set_ioapic_affinity_irq(unsigned int irq,
          */
         dest = SET_APIC_LOGICAL_ID(dest);
 
-        desc = irq_to_desc(irq);
         spin_lock_irqsave(&ioapic_lock, flags);
-        __target_IO_APIC_irq(irq, dest, cfg->vector);
+        __target_IO_APIC_irq(irq, dest, cfg);
         cpumask_copy(&desc->affinity, mask);
         spin_unlock_irqrestore(&ioapic_lock, flags);
 }
+
+static void set_ioapic_affinity_irq(unsigned int irq,
+                                    const struct cpumask *mask)
+{
+        struct irq_desc *desc;
+
+        desc = irq_to_desc(irq);
+
+        set_ioapic_affinity_irq_desc(desc, mask);
+}
 #endif /* CONFIG_SMP */
 
 /*
@@ -397,16 +446,18 @@ static void set_ioapic_affinity_irq(unsigned int irq,
  * shared ISA-space IRQs, so we have to support them. We are super
  * fast in the common case, and fast for shared ISA-space IRQs.
  */
-static void add_pin_to_irq(unsigned int irq, int apic, int pin)
+static void add_pin_to_irq_cpu(struct irq_cfg *cfg, int cpu, int apic, int pin)
 {
-        struct irq_cfg *cfg;
         struct irq_pin_list *entry;
 
-        /* first time to refer irq_cfg, so with new */
-        cfg = irq_cfg_alloc(irq);
         entry = cfg->irq_2_pin;
         if (!entry) {
-                entry = get_one_free_irq_2_pin();
+                entry = get_one_free_irq_2_pin(cpu);
+                if (!entry) {
+                        printk(KERN_ERR "can not alloc irq_2_pin to add %d - %d\n",
+                                        apic, pin);
+                        return;
+                }
                 cfg->irq_2_pin = entry;
                 entry->apic = apic;
                 entry->pin = pin;
@@ -421,7 +472,7 @@ static void add_pin_to_irq(unsigned int irq, int apic, int pin)
                 entry = entry->next;
         }
 
-        entry->next = get_one_free_irq_2_pin();
+        entry->next = get_one_free_irq_2_pin(cpu);
         entry = entry->next;
         entry->apic = apic;
         entry->pin = pin;
@@ -430,11 +481,10 @@ static void add_pin_to_irq(unsigned int irq, int apic, int pin)
 /*
  * Reroute an IRQ to a different pin.
  */
-static void __init replace_pin_at_irq(unsigned int irq,
+static void __init replace_pin_at_irq_cpu(struct irq_cfg *cfg, int cpu,
                                       int oldapic, int oldpin,
                                       int newapic, int newpin)
 {
-        struct irq_cfg *cfg = irq_cfg(irq);
         struct irq_pin_list *entry = cfg->irq_2_pin;
         int replaced = 0;
 
@@ -451,18 +501,16 @@ static void __init replace_pin_at_irq(unsigned int irq,
 
         /* why? call replace before add? */
         if (!replaced)
-                add_pin_to_irq(irq, newapic, newpin);
+                add_pin_to_irq_cpu(cfg, cpu, newapic, newpin);
 }
 
-static inline void io_apic_modify_irq(unsigned int irq,
+static inline void io_apic_modify_irq(struct irq_cfg *cfg,
                                 int mask_and, int mask_or,
                                 void (*final)(struct irq_pin_list *entry))
 {
         int pin;
-        struct irq_cfg *cfg;
         struct irq_pin_list *entry;
 
-        cfg = irq_cfg(irq);
         for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) {
                 unsigned int reg;
                 pin = entry->pin;
@@ -475,9 +523,9 @@ static inline void io_apic_modify_irq(unsigned int irq,
         }
 }
 
-static void __unmask_IO_APIC_irq(unsigned int irq)
+static void __unmask_IO_APIC_irq(struct irq_cfg *cfg)
 {
-        io_apic_modify_irq(irq, ~IO_APIC_REDIR_MASKED, 0, NULL);
+        io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL);
 }
 
 #ifdef CONFIG_X86_64
@@ -492,47 +540,64 @@ void io_apic_sync(struct irq_pin_list *entry)
         readl(&io_apic->data);
 }
 
-static void __mask_IO_APIC_irq(unsigned int irq)
+static void __mask_IO_APIC_irq(struct irq_cfg *cfg)
 {
-        io_apic_modify_irq(irq, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync);
+        io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync);
 }
 #else /* CONFIG_X86_32 */
-static void __mask_IO_APIC_irq(unsigned int irq)
+static void __mask_IO_APIC_irq(struct irq_cfg *cfg)
 {
-        io_apic_modify_irq(irq, ~0, IO_APIC_REDIR_MASKED, NULL);
+        io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, NULL);
 }
 
-static void __mask_and_edge_IO_APIC_irq(unsigned int irq)
+static void __mask_and_edge_IO_APIC_irq(struct irq_cfg *cfg)
 {
-        io_apic_modify_irq(irq, ~IO_APIC_REDIR_LEVEL_TRIGGER,
+        io_apic_modify_irq(cfg, ~IO_APIC_REDIR_LEVEL_TRIGGER,
                         IO_APIC_REDIR_MASKED, NULL);
 }
 
-static void __unmask_and_level_IO_APIC_irq(unsigned int irq)
+static void __unmask_and_level_IO_APIC_irq(struct irq_cfg *cfg)
 {
-        io_apic_modify_irq(irq, ~IO_APIC_REDIR_MASKED,
+        io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED,
                         IO_APIC_REDIR_LEVEL_TRIGGER, NULL);
 }
 #endif /* CONFIG_X86_32 */
 
-static void mask_IO_APIC_irq (unsigned int irq)
+static void mask_IO_APIC_irq_desc(struct irq_desc *desc)
 {
+        struct irq_cfg *cfg = desc->chip_data;
         unsigned long flags;
 
+        BUG_ON(!cfg);
+
         spin_lock_irqsave(&ioapic_lock, flags);
-        __mask_IO_APIC_irq(irq);
+        __mask_IO_APIC_irq(cfg);
         spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
-static void unmask_IO_APIC_irq (unsigned int irq)
+static void unmask_IO_APIC_irq_desc(struct irq_desc *desc)
 {
+        struct irq_cfg *cfg = desc->chip_data;
         unsigned long flags;
 
         spin_lock_irqsave(&ioapic_lock, flags);
-        __unmask_IO_APIC_irq(irq);
+        __unmask_IO_APIC_irq(cfg);
         spin_unlock_irqrestore(&ioapic_lock, flags);
 }
 
+static void mask_IO_APIC_irq(unsigned int irq)
+{
+        struct irq_desc *desc = irq_to_desc(irq);
+
+        mask_IO_APIC_irq_desc(desc);
+}
+static void unmask_IO_APIC_irq(unsigned int irq)
+{
+        struct irq_desc *desc = irq_to_desc(irq);
+
+        unmask_IO_APIC_irq_desc(desc);
+}
+
 static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
 {
         struct IO_APIC_route_entry entry;
@@ -809,7 +874,7 @@ EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector);
  */
 static int EISA_ELCR(unsigned int irq)
 {
-        if (irq < 16) {
+        if (irq < NR_IRQS_LEGACY) {
                 unsigned int port = 0x4d0 + (irq >> 3);
                 return (inb(port) >> (irq & 7)) & 1;
         }
@@ -1034,7 +1099,7 @@ void unlock_vector_lock(void)
         spin_unlock(&vector_lock);
 }
 
-static int __assign_irq_vector(int irq, cpumask_t mask)
+static int __assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask)
 {
         /*
          * NOTE! The local APIC isn't very good at handling
@@ -1050,16 +1115,13 @@ static int __assign_irq_vector(int irq, cpumask_t mask)
         static int current_vector = FIRST_DEVICE_VECTOR, current_offset = 0;
         unsigned int old_vector;
         int cpu;
-        struct irq_cfg *cfg;
 
-        cfg = irq_cfg(irq);
+        if ((cfg->move_in_progress) || cfg->move_cleanup_count)
+                return -EBUSY;
 
         /* Only try and allocate irqs on cpus that are present */
         cpus_and(mask, mask, cpu_online_map);
 
-        if ((cfg->move_in_progress) || cfg->move_cleanup_count)
-                return -EBUSY;
-
         old_vector = cfg->vector;
         if (old_vector) {
                 cpumask_t tmp;
@@ -1113,24 +1175,22 @@ next:
         return -ENOSPC;
 }
 
-static int assign_irq_vector(int irq, cpumask_t mask)
+static int assign_irq_vector(int irq, struct irq_cfg *cfg, cpumask_t mask)
 {
         int err;
         unsigned long flags;
 
         spin_lock_irqsave(&vector_lock, flags);
-        err = __assign_irq_vector(irq, mask);
+        err = __assign_irq_vector(irq, cfg, mask);
         spin_unlock_irqrestore(&vector_lock, flags);
         return err;
 }
 
-static void __clear_irq_vector(int irq)
+static void __clear_irq_vector(int irq, struct irq_cfg *cfg)
 {
-        struct irq_cfg *cfg;
         cpumask_t mask;
         int cpu, vector;
 
-        cfg = irq_cfg(irq);
         BUG_ON(!cfg->vector);
 
         vector = cfg->vector;
@@ -1162,9 +1222,13 @@ void __setup_vector_irq(int cpu)
         /* This function must be called with vector_lock held */
         int irq, vector;
         struct irq_cfg *cfg;
+        struct irq_desc *desc;
 
         /* Mark the inuse vectors */
-        for_each_irq_cfg(irq, cfg) {
+        for_each_irq_desc(irq, desc) {
+                if (!desc)
+                        continue;
+                cfg = desc->chip_data;
                 if (!cpu_isset(cpu, cfg->domain))
                         continue;
                 vector = cfg->vector;
@@ -1215,11 +1279,8 @@ static inline int IO_APIC_irq_trigger(int irq)
 }
 #endif
 
-static void ioapic_register_intr(int irq, unsigned long trigger)
+static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long trigger)
 {
-        struct irq_desc *desc;
-
-        desc = irq_to_desc(irq);
 
         if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
             trigger == IOAPIC_LEVEL)
@@ -1311,7 +1372,7 @@ static int setup_ioapic_entry(int apic, int irq,
         return 0;
 }
 
-static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq,
+static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq, struct irq_desc *desc,
                               int trigger, int polarity)
 {
         struct irq_cfg *cfg;
@@ -1321,10 +1382,10 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq,
         if (!IO_APIC_IRQ(irq))
                 return;
 
-        cfg = irq_cfg(irq);
+        cfg = desc->chip_data;
 
         mask = TARGET_CPUS;
-        if (assign_irq_vector(irq, mask))
+        if (assign_irq_vector(irq, cfg, mask))
                 return;
 
         cpus_and(mask, cfg->domain, mask);
@@ -1341,12 +1402,12 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq,
                                cfg->vector)) {
                 printk("Failed to setup ioapic entry for ioapic  %d, pin %d\n",
                        mp_ioapics[apic].mp_apicid, pin);
-                __clear_irq_vector(irq);
+                __clear_irq_vector(irq, cfg);
                 return;
         }
 
-        ioapic_register_intr(irq, trigger);
-        if (irq < 16)
+        ioapic_register_intr(irq, desc, trigger);
+        if (irq < NR_IRQS_LEGACY)
                 disable_8259A_irq(irq);
 
         ioapic_write_entry(apic, pin, entry);
@@ -1356,6 +1417,9 @@ static void __init setup_IO_APIC_irqs(void)
 {
         int apic, pin, idx, irq;
         int notcon = 0;
+        struct irq_desc *desc;
+        struct irq_cfg *cfg;
+        int cpu = boot_cpu_id;
 
         apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n");
 
@@ -1387,9 +1451,15 @@ static void __init setup_IO_APIC_irqs(void)
                         if (multi_timer_check(apic, irq))
                                 continue;
 #endif
-                        add_pin_to_irq(irq, apic, pin);
+                        desc = irq_to_desc_alloc_cpu(irq, cpu);
+                        if (!desc) {
+                                printk(KERN_INFO "can not get irq_desc for %d\n", irq);
+                                continue;
+                        }
+                        cfg = desc->chip_data;
+                        add_pin_to_irq_cpu(cfg, cpu, apic, pin);
 
-                        setup_IO_APIC_irq(apic, pin, irq,
+                        setup_IO_APIC_irq(apic, pin, irq, desc,
                                         irq_trigger(idx), irq_polarity(idx));
                 }
         }
@@ -1448,6 +1518,7 @@ __apicdebuginit(void) print_IO_APIC(void)
         union IO_APIC_reg_03 reg_03;
         unsigned long flags;
         struct irq_cfg *cfg;
+        struct irq_desc *desc;
         unsigned int irq;
 
         if (apic_verbosity == APIC_QUIET)
@@ -1537,8 +1608,13 @@ __apicdebuginit(void) print_IO_APIC(void)
         }
         }
         printk(KERN_DEBUG "IRQ to pin mappings:\n");
-        for_each_irq_cfg(irq, cfg) {
-                struct irq_pin_list *entry = cfg->irq_2_pin;
+        for_each_irq_desc(irq, desc) {
+                struct irq_pin_list *entry;
+
+                if (!desc)
+                        continue;
+                cfg = desc->chip_data;
+                entry = cfg->irq_2_pin;
                 if (!entry)
                         continue;
                 printk(KERN_DEBUG "IRQ%d ", irq);
@@ -2022,14 +2098,16 @@ static unsigned int startup_ioapic_irq(unsigned int irq)
 {
         int was_pending = 0;
         unsigned long flags;
+        struct irq_cfg *cfg;
 
         spin_lock_irqsave(&ioapic_lock, flags);
-        if (irq < 16) {
+        if (irq < NR_IRQS_LEGACY) {
                 disable_8259A_irq(irq);
                 if (i8259A_irq_pending(irq))
                         was_pending = 1;
         }
-        __unmask_IO_APIC_irq(irq);
+        cfg = irq_cfg(irq);
+        __unmask_IO_APIC_irq(cfg);
         spin_unlock_irqrestore(&ioapic_lock, flags);
 
         return was_pending;
@@ -2092,35 +2170,37 @@ static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration);
  * as simple as edge triggered migration and we can do the irq migration
  * with a simple atomic update to IO-APIC RTE.
  */
-static void migrate_ioapic_irq(int irq, cpumask_t mask)
+static void migrate_ioapic_irq_desc(struct irq_desc *desc, cpumask_t mask)
 {
         struct irq_cfg *cfg;
-        struct irq_desc *desc;
         cpumask_t tmp, cleanup_mask;
         struct irte irte;
         int modify_ioapic_rte;
         unsigned int dest;
         unsigned long flags;
+        unsigned int irq;
 
         cpus_and(tmp, mask, cpu_online_map);
         if (cpus_empty(tmp))
                 return;
 
+        irq = desc->irq;
         if (get_irte(irq, &irte))
                 return;
 
-        if (assign_irq_vector(irq, mask))
+        cfg = desc->chip_data;
+        if (assign_irq_vector(irq, cfg, mask))
                 return;
 
-        cfg = irq_cfg(irq);
+        set_extra_move_desc(desc, mask);
+
         cpus_and(tmp, cfg->domain, mask);
         dest = cpu_mask_to_apicid(tmp);
 
-        desc = irq_to_desc(irq);
         modify_ioapic_rte = desc->status & IRQ_LEVEL;
         if (modify_ioapic_rte) {
                 spin_lock_irqsave(&ioapic_lock, flags);
-                __target_IO_APIC_irq(irq, dest, cfg->vector);
+                __target_IO_APIC_irq(irq, dest, cfg);
                 spin_unlock_irqrestore(&ioapic_lock, flags);
         }
 
@@ -2142,14 +2222,14 @@ static void migrate_ioapic_irq(int irq, cpumask_t mask)
         desc->affinity = mask;
 }
 
-static int migrate_irq_remapped_level(int irq)
+static int migrate_irq_remapped_level_desc(struct irq_desc *desc)
 {
         int ret = -1;
-        struct irq_desc *desc = irq_to_desc(irq);
+        struct irq_cfg *cfg = desc->chip_data;
 
-        mask_IO_APIC_irq(irq);
+        mask_IO_APIC_irq_desc(desc);
 
-        if (io_apic_level_ack_pending(irq)) {
+        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
@@ -2161,14 +2241,15 @@ static int migrate_irq_remapped_level(int irq)
         }
 
         /* everthing is clear. we have right of way */
-        migrate_ioapic_irq(irq, desc->pending_mask);
+        migrate_ioapic_irq_desc(desc, desc->pending_mask);
 
         ret = 0;
         desc->status &= ~IRQ_MOVE_PENDING;
         cpus_clear(desc->pending_mask);
 
 unmask:
-        unmask_IO_APIC_irq(irq);
+        unmask_IO_APIC_irq_desc(desc);
+
         return ret;
 }
 
@@ -2178,6 +2259,9 @@ static void ir_irq_migration(struct work_struct *work)
         struct irq_desc *desc;
 
         for_each_irq_desc(irq, desc) {
+                if (!desc)
+                        continue;
+
                 if (desc->status & IRQ_MOVE_PENDING) {
                         unsigned long flags;
 
@@ -2198,19 +2282,24 @@ static void ir_irq_migration(struct work_struct *work)
 /*
  * Migrates the IRQ destination in the process context.
  */
-static void set_ir_ioapic_affinity_irq(unsigned int irq,
-                                       const struct cpumask *mask)
+static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
+                                            const struct cpumask *mask)
 {
-        struct irq_desc *desc = irq_to_desc(irq);
-
         if (desc->status & IRQ_LEVEL) {
                 desc->status |= IRQ_MOVE_PENDING;
                 cpumask_copy(&desc->pending_mask, mask);
-                migrate_irq_remapped_level(irq);
+                migrate_irq_remapped_level_desc(desc);
                 return;
         }
 
-        migrate_ioapic_irq(irq, *mask);
+        migrate_ioapic_irq_desc(desc, mask);
+}
+static void set_ir_ioapic_affinity_irq(unsigned int irq,
+                                       const struct cpumask *mask)
+{
+        struct irq_desc *desc = irq_to_desc(irq);
+
+        set_ir_ioapic_affinity_irq_desc(desc, mask);
 }
 #endif
 
@@ -2230,6 +2319,9 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void)
                 struct irq_cfg *cfg;
                 irq = __get_cpu_var(vector_irq)[vector];
 
+                if (irq == -1)
+                        continue;
+
                 desc = irq_to_desc(irq);
                 if (!desc)
                         continue;
@@ -2251,9 +2343,10 @@ unlock:
         irq_exit();
 }
 
-static void irq_complete_move(unsigned int irq)
+static void irq_complete_move(struct irq_desc **descp)
 {
-        struct irq_cfg *cfg = irq_cfg(irq);
+        struct irq_desc *desc = *descp;
+        struct irq_cfg *cfg = desc->chip_data;
         unsigned vector, me;
 
         if (likely(!cfg->move_in_progress))
@@ -2271,8 +2364,9 @@ static void irq_complete_move(unsigned int irq)
         }
 }
 #else
-static inline void irq_complete_move(unsigned int irq) {}
+static inline void irq_complete_move(struct irq_desc **descp) {}
 #endif
+
 #ifdef CONFIG_INTR_REMAP
 static void ack_x2apic_level(unsigned int irq)
 {
@@ -2283,11 +2377,14 @@ static void ack_x2apic_edge(unsigned int irq)
 {
         ack_x2APIC_irq();
 }
+
 #endif
 
 static void ack_apic_edge(unsigned int irq)
 {
-        irq_complete_move(irq);
+        struct irq_desc *desc = irq_to_desc(irq);
+
+        irq_complete_move(&desc);
         move_native_irq(irq);
         ack_APIC_irq();
 }
@@ -2296,18 +2393,21 @@ atomic_t irq_mis_count;
 
 static void ack_apic_level(unsigned int irq)
 {
+        struct irq_desc *desc = irq_to_desc(irq);
+
 #ifdef CONFIG_X86_32
         unsigned long v;
         int i;
 #endif
+        struct irq_cfg *cfg;
         int do_unmask_irq = 0;
 
-        irq_complete_move(irq);
+        irq_complete_move(&desc);
 #ifdef CONFIG_GENERIC_PENDING_IRQ
         /* If we are moving the irq we need to mask it */
-        if (unlikely(irq_to_desc(irq)->status & IRQ_MOVE_PENDING)) {
+        if (unlikely(desc->status & IRQ_MOVE_PENDING)) {
                 do_unmask_irq = 1;
-                mask_IO_APIC_irq(irq);
+                mask_IO_APIC_irq_desc(desc);
         }
 #endif
 
@@ -2331,7 +2431,8 @@ static void ack_apic_level(unsigned int irq)
         * operation to prevent an edge-triggered interrupt escaping meanwhile.
         * The idea is from Manfred Spraul.  --macro
         */
-        i = irq_cfg(irq)->vector;
+        cfg = desc->chip_data;
+        i = cfg->vector;
 
         v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1));
 #endif
@@ -2370,17 +2471,18 @@ static void ack_apic_level(unsigned int irq)
                  * accurate and is causing problems then it is a hardware bug
                  * and you can go talk to the chipset vendor about it.
                  */
-                if (!io_apic_level_ack_pending(irq))
+                cfg = desc->chip_data;
+                if (!io_apic_level_ack_pending(cfg))
                         move_masked_irq(irq);
-                unmask_IO_APIC_irq(irq);
+                unmask_IO_APIC_irq_desc(desc);
         }
 
 #ifdef CONFIG_X86_32
         if (!(v & (1 << (i & 0x1f)))) {
                 atomic_inc(&irq_mis_count);
                 spin_lock(&ioapic_lock);
-                __mask_and_edge_IO_APIC_irq(irq);
-                __unmask_and_level_IO_APIC_irq(irq);
+                __mask_and_edge_IO_APIC_irq(cfg);
+                __unmask_and_level_IO_APIC_irq(cfg);
                 spin_unlock(&ioapic_lock);
         }
 #endif
@@ -2431,20 +2533,22 @@ static inline void init_IO_APIC_traps(void)
          * Also, we've got to be careful not to trash gate
          * 0x80, because int 0x80 is hm, kind of importantish. ;)
          */
-        for_each_irq_cfg(irq, cfg) {
-                if (IO_APIC_IRQ(irq) && !cfg->vector) {
+        for_each_irq_desc(irq, desc) {
+                if (!desc)
+                        continue;
+
+                cfg = desc->chip_data;
+                if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) {
                         /*
                          * Hmm.. We don't have an entry for this,
                          * so default to an old-fashioned 8259
                          * interrupt if we can..
                          */
-                        if (irq < 16)
+                        if (irq < NR_IRQS_LEGACY)
                                 make_8259A_irq(irq);
-                        else {
-                                desc = irq_to_desc(irq);
+                        else
                                 /* Strange. Oh, well.. */
                                 desc->chip = &no_irq_chip;
-                        }
                 }
         }
 }
@@ -2469,7 +2573,7 @@ static void unmask_lapic_irq(unsigned int irq)
         apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED);
 }
 
-static void ack_lapic_irq (unsigned int irq)
+static void ack_lapic_irq(unsigned int irq)
 {
         ack_APIC_irq();
 }
@@ -2481,11 +2585,8 @@ static struct irq_chip lapic_chip __read_mostly = {
         .ack            = ack_lapic_irq,
 };
 
-static void lapic_register_intr(int irq)
+static void lapic_register_intr(int irq, struct irq_desc *desc)
 {
-        struct irq_desc *desc;
-
-        desc = irq_to_desc(irq);
         desc->status &= ~IRQ_LEVEL;
         set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq,
                                       "edge");
@@ -2589,7 +2690,9 @@ int timer_through_8259 __initdata;
  */
 static inline void __init check_timer(void)
 {
-        struct irq_cfg *cfg = irq_cfg(0);
+        struct irq_desc *desc = irq_to_desc(0);
+        struct irq_cfg *cfg = desc->chip_data;
+        int cpu = boot_cpu_id;
         int apic1, pin1, apic2, pin2;
         unsigned long flags;
         unsigned int ver;
@@ -2604,7 +2707,7 @@ static inline void __init check_timer(void)
          * get/set the timer IRQ vector:
          */
         disable_8259A_irq(0);
-        assign_irq_vector(0, TARGET_CPUS);
+        assign_irq_vector(0, cfg, TARGET_CPUS);
 
         /*
          * As IRQ0 is to be enabled in the 8259A, the virtual
@@ -2655,10 +2758,10 @@ static inline void __init check_timer(void)
                  * Ok, does IRQ0 through the IOAPIC work?
                  */
                 if (no_pin1) {
-                        add_pin_to_irq(0, apic1, pin1);
+                        add_pin_to_irq_cpu(cfg, cpu, apic1, pin1);
                         setup_timer_IRQ0_pin(apic1, pin1, cfg->vector);
                 }
-                unmask_IO_APIC_irq(0);
+                unmask_IO_APIC_irq_desc(desc);
                 if (timer_irq_works()) {
                         if (nmi_watchdog == NMI_IO_APIC) {
                                 setup_nmi();
@@ -2684,9 +2787,9 @@ static inline void __init check_timer(void)
                 /*
                  * legacy devices should be connected to IO APIC #0
                  */
-                replace_pin_at_irq(0, apic1, pin1, apic2, pin2);
+                replace_pin_at_irq_cpu(cfg, cpu, apic1, pin1, apic2, pin2);
                 setup_timer_IRQ0_pin(apic2, pin2, cfg->vector);
-                unmask_IO_APIC_irq(0);
+                unmask_IO_APIC_irq_desc(desc);
                 enable_8259A_irq(0);
                 if (timer_irq_works()) {
                         apic_printk(APIC_QUIET, KERN_INFO "....... works.\n");
@@ -2718,7 +2821,7 @@ static inline void __init check_timer(void)
         apic_printk(APIC_QUIET, KERN_INFO
                     "...trying to set up timer as Virtual Wire IRQ...\n");
 
-        lapic_register_intr(0);
+        lapic_register_intr(0, desc);
         apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector);     /* Fixed mode */
         enable_8259A_irq(0);
 
@@ -2903,22 +3006,26 @@ unsigned int create_irq_nr(unsigned int irq_want)
         unsigned int irq;
         unsigned int new;
         unsigned long flags;
-        struct irq_cfg *cfg_new;
-
-        irq_want = nr_irqs - 1;
+        struct irq_cfg *cfg_new = NULL;
+        int cpu = boot_cpu_id;
+        struct irq_desc *desc_new = NULL;
 
         irq = 0;
         spin_lock_irqsave(&vector_lock, flags);
-        for (new = irq_want; new > 0; new--) {
+        for (new = irq_want; new < NR_IRQS; new++) {
                 if (platform_legacy_irq(new))
                         continue;
-                cfg_new = irq_cfg(new);
-                if (cfg_new && cfg_new->vector != 0)
+
+                desc_new = irq_to_desc_alloc_cpu(new, cpu);
+                if (!desc_new) {
+                        printk(KERN_INFO "can not get irq_desc for %d\n", new);
+                        continue;
+                }
+                cfg_new = desc_new->chip_data;
+
+                if (cfg_new->vector != 0)
                         continue;
-                /* check if need to create one */
-                if (!cfg_new)
-                        cfg_new = irq_cfg_alloc(new);
-                if (__assign_irq_vector(new, TARGET_CPUS) == 0)
+                if (__assign_irq_vector(new, cfg_new, TARGET_CPUS) == 0)
                         irq = new;
                 break;
         }
@@ -2926,15 +3033,21 @@ unsigned int create_irq_nr(unsigned int irq_want)
 
         if (irq > 0) {
                 dynamic_irq_init(irq);
+                /* restore it, in case dynamic_irq_init clear it */
+                if (desc_new)
+                        desc_new->chip_data = cfg_new;
         }
         return irq;
 }
 
+static int nr_irqs_gsi = NR_IRQS_LEGACY;
 int create_irq(void)
 {
+        unsigned int irq_want;
         int irq;
 
-        irq = create_irq_nr(nr_irqs - 1);
+        irq_want = nr_irqs_gsi;
+        irq = create_irq_nr(irq_want);
 
         if (irq == 0)
                 irq = -1;
@@ -2945,14 +3058,22 @@ int create_irq(void)
 void destroy_irq(unsigned int irq)
 {
         unsigned long flags;
+        struct irq_cfg *cfg;
+        struct irq_desc *desc;
 
+        /* store it, in case dynamic_irq_cleanup clear it */
+        desc = irq_to_desc(irq);
+        cfg = desc->chip_data;
         dynamic_irq_cleanup(irq);
+        /* connect back irq_cfg */
+        if (desc)
+                desc->chip_data = cfg;
 
 #ifdef CONFIG_INTR_REMAP
         free_irte(irq);
 #endif
         spin_lock_irqsave(&vector_lock, flags);
-        __clear_irq_vector(irq);
+        __clear_irq_vector(irq, cfg);
         spin_unlock_irqrestore(&vector_lock, flags);
 }
 
@@ -2967,12 +3088,12 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
         unsigned dest;
         cpumask_t tmp;
 
+        cfg = irq_cfg(irq);
         tmp = TARGET_CPUS;
-        err = assign_irq_vector(irq, tmp);
+        err = assign_irq_vector(irq, cfg, tmp);
         if (err)
                 return err;
 
-        cfg = irq_cfg(irq);
         cpus_and(tmp, cfg->domain, tmp);
         dest = cpu_mask_to_apicid(tmp);
 
@@ -3030,34 +3151,34 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 #ifdef CONFIG_SMP
 static void set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
+        struct irq_desc *desc = irq_to_desc(irq);
         struct irq_cfg *cfg;
         struct msi_msg msg;
         unsigned int dest;
         cpumask_t tmp;
-        struct irq_desc *desc;
 
         if (!cpumask_intersects(mask, cpu_online_mask))
                 return;
 
-        if (assign_irq_vector(irq, *mask))
+        cfg = desc->chip_data;
+        if (assign_irq_vector(irq, cfg, *mask))
                 return;
 
-        cfg = irq_cfg(irq);
+        set_extra_move_desc(desc, *mask);
+
         cpumask_and(&tmp, &cfg->domain, mask);
         dest = cpu_mask_to_apicid(tmp);
 
-        read_msi_msg(irq, &msg);
+        read_msi_msg_desc(desc, &msg);
 
         msg.data &= ~MSI_DATA_VECTOR_MASK;
         msg.data |= MSI_DATA_VECTOR(cfg->vector);
         msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
         msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
-        write_msi_msg(irq, &msg);
-        desc = irq_to_desc(irq);
+        write_msi_msg_desc(desc, &msg);
         cpumask_copy(&desc->affinity, mask);
 }
-
 #ifdef CONFIG_INTR_REMAP
 /*
  * Migrate the MSI irq to another cpumask. This migration is
@@ -3066,11 +3187,11 @@ static void set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
 static void ir_set_msi_irq_affinity(unsigned int irq,
                                     const struct cpumask *mask)
 {
+        struct irq_desc *desc = irq_to_desc(irq);
         struct irq_cfg *cfg;
         unsigned int dest;
         cpumask_t tmp, cleanup_mask;
         struct irte irte;
-        struct irq_desc *desc;
 
         if (!cpumask_intersects(mask, cpu_online_mask))
                 return;
@@ -3078,10 +3199,12 @@ static void ir_set_msi_irq_affinity(unsigned int irq,
         if (get_irte(irq, &irte))
                 return;
 
-        if (assign_irq_vector(irq, *mask))
+        cfg = desc->chip_data;
+        if (assign_irq_vector(irq, cfg, *mask))
                 return;
 
-        cfg = irq_cfg(irq);
+        set_extra_move_desc(desc, mask);
+
         cpumask_and(&tmp, &cfg->domain, mask);
         dest = cpu_mask_to_apicid(tmp);
 
@@ -3105,9 +3228,9 @@ static void ir_set_msi_irq_affinity(unsigned int irq,
                 cfg->move_in_progress = 0;
         }
 
-        desc = irq_to_desc(irq);
         cpumask_copy(&desc->affinity, mask);
 }
+
 #endif
 #endif /* CONFIG_SMP */
 
@@ -3166,7 +3289,7 @@ static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec)
 }
 #endif
 
-static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq)
+static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
 {
         int ret;
         struct msi_msg msg;
@@ -3175,7 +3298,7 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq)
         if (ret < 0)
                 return ret;
 
-        set_irq_msi(irq, desc);
+        set_irq_msi(irq, msidesc);
         write_msi_msg(irq, &msg);
 
 #ifdef CONFIG_INTR_REMAP
@@ -3195,26 +3318,13 @@ static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq)
         return 0;
 }
 
-static unsigned int build_irq_for_pci_dev(struct pci_dev *dev)
-{
-        unsigned int irq;
-
-        irq = dev->bus->number;
-        irq <<= 8;
-        irq |= dev->devfn;
-        irq <<= 12;
-
-        return irq;
-}
-
-int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc)
+int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc)
 {
         unsigned int irq;
         int ret;
         unsigned int irq_want;
 
-        irq_want = build_irq_for_pci_dev(dev) + 0x100;
-
+        irq_want = nr_irqs_gsi;
         irq = create_irq_nr(irq_want);
         if (irq == 0)
                 return -1;
@@ -3228,7 +3338,7 @@ int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc)
                 goto error;
 no_ir:
 #endif
-        ret = setup_msi_irq(dev, desc, irq);
+        ret = setup_msi_irq(dev, msidesc, irq);
         if (ret < 0) {
                 destroy_irq(irq);
                 return ret;
@@ -3246,7 +3356,7 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 {
         unsigned int irq;
         int ret, sub_handle;
-        struct msi_desc *desc;
+        struct msi_desc *msidesc;
         unsigned int irq_want;
 
 #ifdef CONFIG_INTR_REMAP
@@ -3254,10 +3364,11 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
         int index = 0;
 #endif
 
-        irq_want = build_irq_for_pci_dev(dev) + 0x100;
+        irq_want = nr_irqs_gsi;
         sub_handle = 0;
-        list_for_each_entry(desc, &dev->msi_list, list) {
-                irq = create_irq_nr(irq_want--);
+        list_for_each_entry(msidesc, &dev->msi_list, list) {
+                irq = create_irq_nr(irq_want);
+                irq_want++;
                 if (irq == 0)
                         return -1;
 #ifdef CONFIG_INTR_REMAP
@@ -3289,7 +3400,7 @@ int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
                 }
 no_ir:
 #endif
-                ret = setup_msi_irq(dev, desc, irq);
+                ret = setup_msi_irq(dev, msidesc, irq);
                 if (ret < 0)
                         goto error;
                 sub_handle++;
@@ -3310,19 +3421,21 @@ void arch_teardown_msi_irq(unsigned int irq)
 #ifdef CONFIG_SMP
 static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
+        struct irq_desc *desc = irq_to_desc(irq);
         struct irq_cfg *cfg;
         struct msi_msg msg;
         unsigned int dest;
         cpumask_t tmp;
-        struct irq_desc *desc;
 
         if (!cpumask_intersects(mask, cpu_online_mask))
                 return;
 
-        if (assign_irq_vector(irq, *mask))
+        cfg = desc->chip_data;
+        if (assign_irq_vector(irq, cfg, *mask))
                 return;
 
-        cfg = irq_cfg(irq);
+        set_extra_move_desc(desc, *mask);
+
         cpumask_and(&tmp, &cfg->domain, mask);
         dest = cpu_mask_to_apicid(tmp);
 
@@ -3334,9 +3447,9 @@ static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
         msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
         dmar_msi_write(irq, &msg);
-        desc = irq_to_desc(irq);
         cpumask_copy(&desc->affinity, mask);
 }
+
 #endif /* CONFIG_SMP */
 
 struct irq_chip dmar_msi_type = {
@@ -3370,8 +3483,8 @@ int arch_setup_dmar_msi(unsigned int irq)
 #ifdef CONFIG_SMP
 static void hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
+        struct irq_desc *desc = irq_to_desc(irq);
         struct irq_cfg *cfg;
-        struct irq_desc *desc;
         struct msi_msg msg;
         unsigned int dest;
         cpumask_t tmp;
@@ -3379,10 +3492,12 @@ static void hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
         if (!cpumask_intersects(mask, cpu_online_mask))
                 return;
 
-        if (assign_irq_vector(irq, *mask))
+        cfg = desc->chip_data;
+        if (assign_irq_vector(irq, cfg, *mask))
                 return;
 
-        cfg = irq_cfg(irq);
+        set_extra_move_desc(desc, *mask);
+
         cpumask_and(&tmp, &cfg->domain, mask);
         dest = cpu_mask_to_apicid(tmp);
 
@@ -3394,9 +3509,9 @@ static void hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
         msg.address_lo |= MSI_ADDR_DEST_ID(dest);
 
         hpet_msi_write(irq, &msg);
-        desc = irq_to_desc(irq);
         cpumask_copy(&desc->affinity, mask);
 }
+
 #endif /* CONFIG_SMP */
 
 struct irq_chip hpet_msi_type = {
@@ -3451,25 +3566,27 @@ static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector)
 
 static void set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask)
 {
+        struct irq_desc *desc = irq_to_desc(irq);
         struct irq_cfg *cfg;
         unsigned int dest;
         cpumask_t tmp;
-        struct irq_desc *desc;
 
         if (!cpumask_intersects(mask, cpu_online_mask))
                 return;
 
-        if (assign_irq_vector(irq, *mask))
+        cfg = desc->chip_data;
+        if (assign_irq_vector(irq, cfg, *mask))
                 return;
 
-        cfg = irq_cfg(irq);
+        set_extra_move_desc(desc, *mask);
+
         cpumask_and(&tmp, &cfg->domain, mask);
         dest = cpu_mask_to_apicid(tmp);
 
         target_ht_irq(irq, dest, cfg->vector);
-        desc = irq_to_desc(irq);
         cpumask_copy(&desc->affinity, mask);
 }
+
 #endif
 
 static struct irq_chip ht_irq_chip = {
@@ -3489,13 +3606,13 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
         int err;
         cpumask_t tmp;
 
+        cfg = irq_cfg(irq);
         tmp = TARGET_CPUS;
-        err = assign_irq_vector(irq, tmp);
+        err = assign_irq_vector(irq, cfg, tmp);
         if (!err) {
                 struct ht_irq_msg msg;
                 unsigned dest;
 
-                cfg = irq_cfg(irq);
                 cpus_and(tmp, cfg->domain, tmp);
                 dest = cpu_mask_to_apicid(tmp);
 
@@ -3541,7 +3658,9 @@ int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
         unsigned long flags;
         int err;
 
-        err = assign_irq_vector(irq, *eligible_cpu);
+        cfg = irq_cfg(irq);
+
+        err = assign_irq_vector(irq, cfg, *eligible_cpu);
         if (err != 0)
                 return err;
 
@@ -3550,8 +3669,6 @@ int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade,
                                       irq_name);
         spin_unlock_irqrestore(&vector_lock, flags);
 
-        cfg = irq_cfg(irq);
-
         mmr_value = 0;
         entry = (struct uv_IO_APIC_route_entry *)&mmr_value;
         BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long));
@@ -3603,9 +3720,16 @@ int __init io_apic_get_redir_entries (int ioapic)
         return reg_01.bits.entries;
 }
 
-int __init probe_nr_irqs(void)
+void __init probe_nr_irqs_gsi(void)
 {
-        return NR_IRQS;
+        int idx;
+        int nr = 0;
+
+        for (idx = 0; idx < nr_ioapics; idx++)
+                nr += io_apic_get_redir_entries(idx) + 1;
+
+        if (nr > nr_irqs_gsi)
+                nr_irqs_gsi = nr;
 }
 
 /* --------------------------------------------------------------------------
@@ -3704,19 +3828,31 @@ int __init io_apic_get_version(int ioapic)
 
 int io_apic_set_pci_routing (int ioapic, int pin, int irq, int triggering, int polarity)
 {
+        struct irq_desc *desc;
+        struct irq_cfg *cfg;
+        int cpu = boot_cpu_id;
+
         if (!IO_APIC_IRQ(irq)) {
                 apic_printk(APIC_QUIET,KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n",
                         ioapic);
                 return -EINVAL;
         }
 
+        desc = irq_to_desc_alloc_cpu(irq, cpu);
+        if (!desc) {
+                printk(KERN_INFO "can not get irq_desc %d\n", irq);
+                return 0;
+        }
+
         /*
          * IRQs < 16 are already in the irq_2_pin[] map
          */
-        if (irq >= 16)
-                add_pin_to_irq(irq, ioapic, pin);
+        if (irq >= NR_IRQS_LEGACY) {
+                cfg = desc->chip_data;
+                add_pin_to_irq_cpu(cfg, cpu, ioapic, pin);
+        }
 
-        setup_IO_APIC_irq(ioapic, pin, irq, triggering, polarity);
+        setup_IO_APIC_irq(ioapic, pin, irq, desc, triggering, polarity);
 
         return 0;
 }
@@ -3770,9 +3906,10 @@ void __init setup_ioapic_dest(void)
                          * when you have too many devices, because at that time only boot
                          * cpu is online.
                          */
-                        cfg = irq_cfg(irq);
+                        desc = irq_to_desc(irq);
+                        cfg = desc->chip_data;
                         if (!cfg->vector) {
-                                setup_IO_APIC_irq(ioapic, pin, irq,
+                                setup_IO_APIC_irq(ioapic, pin, irq, desc,
                                                   irq_trigger(irq_entry),
                                                   irq_polarity(irq_entry));
                                 continue;
@@ -3782,7 +3919,6 @@ void __init setup_ioapic_dest(void)
                         /*
                          * Honour affinities which have been set in early boot
                          */
-                        desc = irq_to_desc(irq);
                         if (desc->status &
                             (IRQ_NO_BALANCING | IRQ_AFFINITY_SET))
                                 mask = desc->affinity;
@@ -3791,10 +3927,10 @@ void __init setup_ioapic_dest(void)
 
 #ifdef CONFIG_INTR_REMAP
                         if (intr_remapping_enabled)
-                                set_ir_ioapic_affinity_irq(irq, &mask);
+                                set_ir_ioapic_affinity_irq_desc(desc, &mask);
                         else
 #endif
-                                set_ioapic_affinity_irq(irq, &mask);
+                                set_ioapic_affinity_irq_desc(desc, &mask);
                 }
 
         }
@@ -3843,7 +3979,6 @@ void __init ioapic_init_mappings(void)
         struct resource *ioapic_res;
         int i;
 
-        irq_2_pin_init();
         ioapic_res = ioapic_setup_resources();
         for (i = 0; i < nr_ioapics; i++) {
                 if (smp_found_config) {