9 files changed, 68 insertions, 67 deletions

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 45efe0ca88f8..4310b4a311a5 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -250,6 +250,15 @@ config SCHED_SMT
           cost of slightly increased overhead in some places. If unsure say
           N here.
 
+config SCHED_MC
+        bool "Multi-core scheduler support"
+        depends on SMP
+        default y
+        help
+          Multi-core scheduler support improves the CPU scheduler's decision
+          making when dealing with multi-core CPU chips at a cost of slightly
+          increased overhead in some places. If unsure say N here.
+
 source "kernel/Kconfig.preempt"
 
 config NUMA
@@ -325,6 +334,10 @@ config HAVE_ARCH_EARLY_PFN_TO_NID
         def_bool y
         depends on NUMA
 
+config OUT_OF_LINE_PFN_TO_PAGE
+        def_bool y
+        depends on DISCONTIGMEM
+
 config NR_CPUS
         int "Maximum number of CPUs (2-256)"
         range 2 255
diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index 7549a4389fbf..35b2faccdc6c 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -688,6 +688,8 @@ ia32_sys_call_table:
         .quad sys_ni_syscall            /* pselect6 for now */
         .quad sys_ni_syscall            /* ppoll for now */
         .quad sys_unshare               /* 310 */
+        .quad compat_sys_set_robust_list
+        .quad compat_sys_get_robust_list
 ia32_syscall_end:               
         .rept IA32_NR_syscalls-(ia32_syscall_end-ia32_sys_call_table)/8
                 .quad ni_syscall
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 0370720515f1..70dd8e5c6889 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -66,24 +66,17 @@ EXPORT_SYMBOL(boot_option_idle_override);
 void (*pm_idle)(void);
 static DEFINE_PER_CPU(unsigned int, cpu_idle_state);
 
-static struct notifier_block *idle_notifier;
-static DEFINE_SPINLOCK(idle_notifier_lock);
+static ATOMIC_NOTIFIER_HEAD(idle_notifier);
 
 void idle_notifier_register(struct notifier_block *n)
 {
-        unsigned long flags;
-        spin_lock_irqsave(&idle_notifier_lock, flags);
-        notifier_chain_register(&idle_notifier, n);
-        spin_unlock_irqrestore(&idle_notifier_lock, flags);
+        atomic_notifier_chain_register(&idle_notifier, n);
 }
 EXPORT_SYMBOL_GPL(idle_notifier_register);
 
 void idle_notifier_unregister(struct notifier_block *n)
 {
-        unsigned long flags;
-        spin_lock_irqsave(&idle_notifier_lock, flags);
-        notifier_chain_unregister(&idle_notifier, n);
-        spin_unlock_irqrestore(&idle_notifier_lock, flags);
+        atomic_notifier_chain_unregister(&idle_notifier, n);
 }
 EXPORT_SYMBOL(idle_notifier_unregister);
 
@@ -93,13 +86,13 @@ static DEFINE_PER_CPU(enum idle_state, idle_state) = CPU_NOT_IDLE;
 void enter_idle(void)
 {
         __get_cpu_var(idle_state) = CPU_IDLE;
-        notifier_call_chain(&idle_notifier, IDLE_START, NULL);
+        atomic_notifier_call_chain(&idle_notifier, IDLE_START, NULL);
 }
 
 static void __exit_idle(void)
 {
         __get_cpu_var(idle_state) = CPU_NOT_IDLE;
-        notifier_call_chain(&idle_notifier, IDLE_END, NULL);
+        atomic_notifier_call_chain(&idle_notifier, IDLE_END, NULL);
 }
 
 /* Called from interrupts to signify idle end */
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index a57eec8311a7..d1f3e9272c05 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -962,7 +962,6 @@ static void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 
         cpuid(1, &eax, &ebx, &ecx, &edx);
 
-        c->apicid = phys_pkg_id(0);
 
         if (!cpu_has(c, X86_FEATURE_HT) || cpu_has(c, X86_FEATURE_CMP_LEGACY))
                 return;
@@ -1171,6 +1170,8 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
                         c->x86_capability[2] = cpuid_edx(0x80860001);
         }
 
+        c->apicid = phys_pkg_id(0);
+
         /*
          * Vendor-specific initialization.  In this section we
          * canonicalize the feature flags, meaning if there are
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 66e98659d077..ea48fa638070 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -68,6 +68,9 @@ u8 phys_proc_id[NR_CPUS] __read_mostly = { [0 ... NR_CPUS-1] = BAD_APICID };
 /* core ID of each logical CPU */
 u8 cpu_core_id[NR_CPUS] __read_mostly = { [0 ... NR_CPUS-1] = BAD_APICID };
 
+/* Last level cache ID of each logical CPU */
+u8 cpu_llc_id[NR_CPUS] __cpuinitdata  = {[0 ... NR_CPUS-1] = BAD_APICID};
+
 /* Bitmask of currently online CPUs */
 cpumask_t cpu_online_map __read_mostly;
 
@@ -445,6 +448,18 @@ void __cpuinit smp_callin(void)
         cpu_set(cpuid, cpu_callin_map);
 }
 
+/* maps the cpu to the sched domain representing multi-core */
+cpumask_t cpu_coregroup_map(int cpu)
+{
+        struct cpuinfo_x86 *c = cpu_data + cpu;
+        /*
+         * For perf, we return last level cache shared map.
+         * TBD: when power saving sched policy is added, we will return
+         *      cpu_core_map when power saving policy is enabled
+         */
+        return c->llc_shared_map;
+}
+
 /* representing cpus for which sibling maps can be computed */
 static cpumask_t cpu_sibling_setup_map;
 
@@ -463,12 +478,16 @@ static inline void set_cpu_sibling_map(int cpu)
                                 cpu_set(cpu, cpu_sibling_map[i]);
                                 cpu_set(i, cpu_core_map[cpu]);
                                 cpu_set(cpu, cpu_core_map[i]);
+                                cpu_set(i, c[cpu].llc_shared_map);
+                                cpu_set(cpu, c[i].llc_shared_map);
                         }
                 }
         } else {
                 cpu_set(cpu, cpu_sibling_map[cpu]);
         }
 
+        cpu_set(cpu, c[cpu].llc_shared_map);
+
         if (current_cpu_data.x86_max_cores == 1) {
                 cpu_core_map[cpu] = cpu_sibling_map[cpu];
                 c[cpu].booted_cores = 1;
@@ -476,6 +495,11 @@ static inline void set_cpu_sibling_map(int cpu)
         }
 
         for_each_cpu_mask(i, cpu_sibling_setup_map) {
+                if (cpu_llc_id[cpu] != BAD_APICID &&
+                    cpu_llc_id[cpu] == cpu_llc_id[i]) {
+                        cpu_set(i, c[cpu].llc_shared_map);
+                        cpu_set(cpu, c[i].llc_shared_map);
+                }
                 if (phys_proc_id[cpu] == phys_proc_id[i]) {
                         cpu_set(i, cpu_core_map[cpu]);
                         cpu_set(cpu, cpu_core_map[i]);
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 7f58fa682491..473b514b66e4 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -504,42 +504,25 @@ unsigned long long sched_clock(void)
 
 static unsigned long get_cmos_time(void)
 {
-        unsigned int timeout = 1000000, year, mon, day, hour, min, sec;
-        unsigned char uip = 0, this = 0;
+        unsigned int year, mon, day, hour, min, sec;
         unsigned long flags;
         unsigned extyear = 0;
 
-/*
- * The Linux interpretation of the CMOS clock register contents: When the
- * Update-In-Progress (UIP) flag goes from 1 to 0, the RTC registers show the
- * second which has precisely just started. Waiting for this can take up to 1
- * second, we timeout approximately after 2.4 seconds on a machine with
- * standard 8.3 MHz ISA bus.
- */
-
         spin_lock_irqsave(&rtc_lock, flags);
 
-        while (timeout && (!uip || this)) {
-                uip |= this;
-                this = CMOS_READ(RTC_FREQ_SELECT) & RTC_UIP;
-                timeout--;
-        }
-
-        /*
-         * Here we are safe to assume the registers won't change for a whole
-         * second, so we just go ahead and read them.
-         */
-        sec = CMOS_READ(RTC_SECONDS);
-        min = CMOS_READ(RTC_MINUTES);
-        hour = CMOS_READ(RTC_HOURS);
-        day = CMOS_READ(RTC_DAY_OF_MONTH);
-        mon = CMOS_READ(RTC_MONTH);
-        year = CMOS_READ(RTC_YEAR);
-
+        do {
+                sec = CMOS_READ(RTC_SECONDS);
+                min = CMOS_READ(RTC_MINUTES);
+                hour = CMOS_READ(RTC_HOURS);
+                day = CMOS_READ(RTC_DAY_OF_MONTH);
+                mon = CMOS_READ(RTC_MONTH);
+                year = CMOS_READ(RTC_YEAR);
 #ifdef CONFIG_ACPI
-        if (acpi_fadt.revision >= FADT2_REVISION_ID && acpi_fadt.century)
-                extyear = CMOS_READ(acpi_fadt.century);
+                if (acpi_fadt.revision >= FADT2_REVISION_ID &&
+                                        acpi_fadt.century)
+                        extyear = CMOS_READ(acpi_fadt.century);
 #endif
+        } while (sec != CMOS_READ(RTC_SECONDS));
 
         spin_unlock_irqrestore(&rtc_lock, flags);
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 7b148309c529..edaa9fe654dc 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -69,20 +69,20 @@ asmlinkage void alignment_check(void);
 asmlinkage void machine_check(void);
 asmlinkage void spurious_interrupt_bug(void);
 
-struct notifier_block *die_chain;
-static DEFINE_SPINLOCK(die_notifier_lock);
+ATOMIC_NOTIFIER_HEAD(die_chain);
 
 int register_die_notifier(struct notifier_block *nb)
 {
-        int err = 0;
-        unsigned long flags;
-
         vmalloc_sync_all();
-        spin_lock_irqsave(&die_notifier_lock, flags);
-        err = notifier_chain_register(&die_chain, nb);
-        spin_unlock_irqrestore(&die_notifier_lock, flags);
-        return err;
+        return atomic_notifier_chain_register(&die_chain, nb);
+}
+EXPORT_SYMBOL(register_die_notifier);
+
+int unregister_die_notifier(struct notifier_block *nb)
+{
+        return atomic_notifier_chain_unregister(&die_chain, nb);
 }
+EXPORT_SYMBOL(unregister_die_notifier);
 
 static inline void conditional_sti(struct pt_regs *regs)
 {
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index b04415625442..e5f7f1c34462 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -72,7 +72,7 @@ void show_mem(void)
         show_free_areas();
         printk(KERN_INFO "Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
 
-        for_each_pgdat(pgdat) {
+        for_each_online_pgdat(pgdat) {
                for (i = 0; i < pgdat->node_spanned_pages; ++i) {
                         page = pfn_to_page(pgdat->node_start_pfn + i);
                         total++;
diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index 63c72641b737..4be82d6e2b48 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -377,21 +377,6 @@ EXPORT_SYMBOL(node_data);
  * Should do that.
  */
 
-/* Requires pfn_valid(pfn) to be true */
-struct page *pfn_to_page(unsigned long pfn)
-{
-        int nid = phys_to_nid(((unsigned long)(pfn)) << PAGE_SHIFT);
-        return (pfn - node_start_pfn(nid)) + NODE_DATA(nid)->node_mem_map;
-}
-EXPORT_SYMBOL(pfn_to_page);
-
-unsigned long page_to_pfn(struct page *page)
-{
-        return (long)(((page) - page_zone(page)->zone_mem_map) +
-                      page_zone(page)->zone_start_pfn);
-}
-EXPORT_SYMBOL(page_to_pfn);
-
 int pfn_valid(unsigned long pfn)
 {
         unsigned nid;