Merge branch 'linus' into cont_syslog
[safe/jmp/linux-2.6] / include / linux / topology.h
index 3df1d47..c44df50 100644 (file)
@@ -5,7 +5,7 @@
  *
  * Copyright (C) 2002, IBM Corp.
  *
- * All rights reserved.          
+ * All rights reserved.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -31,6 +31,7 @@
 #include <linux/bitops.h>
 #include <linux/mmzone.h>
 #include <linux/smp.h>
+#include <linux/percpu.h>
 #include <asm/topology.h>
 
 #ifndef node_has_online_mem
 #endif
 
 #ifndef nr_cpus_node
-#define nr_cpus_node(node)                                                     \
-       ({                                                                      \
-               cpumask_t __tmp__;                                              \
-               __tmp__ = node_to_cpumask(node);                                \
-               cpus_weight(__tmp__);                                           \
-       })
+#define nr_cpus_node(node) cpumask_weight(cpumask_of_node(node))
 #endif
 
-#define for_each_node_with_cpus(node)                                          \
-       for_each_online_node(node)                                              \
+#define for_each_node_with_cpus(node)                  \
+       for_each_online_node(node)                      \
                if (nr_cpus_node(node))
 
-#ifndef node_distance
+int arch_update_cpu_topology(void);
+
 /* Conform to ACPI 2.0 SLIT distance definitions */
 #define LOCAL_DISTANCE         10
 #define REMOTE_DISTANCE                20
+#ifndef node_distance
 #define node_distance(from,to) ((from) == (to) ? LOCAL_DISTANCE : REMOTE_DISTANCE)
 #endif
+#ifndef RECLAIM_DISTANCE
+/*
+ * If the distance between nodes in a system is larger than RECLAIM_DISTANCE
+ * (in whatever arch specific measurement units returned by node_distance())
+ * then switch on zone reclaim on boot.
+ */
+#define RECLAIM_DISTANCE 20
+#endif
 #ifndef PENALTY_FOR_NODE_WITH_CPUS
 #define PENALTY_FOR_NODE_WITH_CPUS     (1)
 #endif
@@ -70,7 +76,9 @@
  * by defining their own arch-specific initializer in include/asm/topology.h.
  * A definition there will automagically override these default initializers
  * and allow arch-specific performance tuning of sched_domains.
+ * (Only non-zero and non-null fields need be specified.)
  */
+
 #ifdef CONFIG_SCHED_SMT
 /* MCD - Do we really need this?  It is always on if CONFIG_SCHED_SMT is,
  * so can't we drop this in favor of CONFIG_SCHED_SMT?
 #define ARCH_HAS_SCHED_WAKE_IDLE
 /* Common values for SMT siblings */
 #ifndef SD_SIBLING_INIT
-#define SD_SIBLING_INIT (struct sched_domain) {                \
-       .span                   = CPU_MASK_NONE,        \
-       .parent                 = NULL,                 \
-       .groups                 = NULL,                 \
-       .min_interval           = 1,                    \
-       .max_interval           = 2,                    \
-       .busy_factor            = 8,                    \
-       .imbalance_pct          = 110,                  \
-       .cache_hot_time         = 0,                    \
-       .cache_nice_tries       = 0,                    \
-       .per_cpu_gain           = 25,                   \
-       .busy_idx               = 0,                    \
-       .idle_idx               = 0,                    \
-       .newidle_idx            = 1,                    \
-       .wake_idx               = 0,                    \
-       .forkexec_idx           = 0,                    \
-       .flags                  = SD_LOAD_BALANCE       \
-                               | SD_BALANCE_NEWIDLE    \
-                               | SD_BALANCE_EXEC       \
-                               | SD_WAKE_AFFINE        \
-                               | SD_WAKE_IDLE          \
-                               | SD_SHARE_CPUPOWER,    \
-       .last_balance           = jiffies,              \
-       .balance_interval       = 1,                    \
-       .nr_balance_failed      = 0,                    \
+#define SD_SIBLING_INIT (struct sched_domain) {                                \
+       .min_interval           = 1,                                    \
+       .max_interval           = 2,                                    \
+       .busy_factor            = 64,                                   \
+       .imbalance_pct          = 110,                                  \
+                                                                       \
+       .flags                  = 1*SD_LOAD_BALANCE                     \
+                               | 1*SD_BALANCE_NEWIDLE                  \
+                               | 1*SD_BALANCE_EXEC                     \
+                               | 1*SD_BALANCE_FORK                     \
+                               | 0*SD_BALANCE_WAKE                     \
+                               | 1*SD_WAKE_AFFINE                      \
+                               | 1*SD_SHARE_CPUPOWER                   \
+                               | 0*SD_POWERSAVINGS_BALANCE             \
+                               | 1*SD_SHARE_PKG_RESOURCES              \
+                               | 0*SD_SERIALIZE                        \
+                               | 0*SD_PREFER_SIBLING                   \
+                               ,                                       \
+       .last_balance           = jiffies,                              \
+       .balance_interval       = 1,                                    \
+       .smt_gain               = 1178, /* 15% */                       \
 }
 #endif
 #endif /* CONFIG_SCHED_SMT */
 
+#ifdef CONFIG_SCHED_MC
+/* Common values for MC siblings. for now mostly derived from SD_CPU_INIT */
+#ifndef SD_MC_INIT
+#define SD_MC_INIT (struct sched_domain) {                             \
+       .min_interval           = 1,                                    \
+       .max_interval           = 4,                                    \
+       .busy_factor            = 64,                                   \
+       .imbalance_pct          = 125,                                  \
+       .cache_nice_tries       = 1,                                    \
+       .busy_idx               = 2,                                    \
+       .wake_idx               = 0,                                    \
+       .forkexec_idx           = 0,                                    \
+                                                                       \
+       .flags                  = 1*SD_LOAD_BALANCE                     \
+                               | 1*SD_BALANCE_NEWIDLE                  \
+                               | 1*SD_BALANCE_EXEC                     \
+                               | 1*SD_BALANCE_FORK                     \
+                               | 0*SD_BALANCE_WAKE                     \
+                               | 1*SD_WAKE_AFFINE                      \
+                               | 0*SD_PREFER_LOCAL                     \
+                               | 0*SD_SHARE_CPUPOWER                   \
+                               | 1*SD_SHARE_PKG_RESOURCES              \
+                               | 0*SD_SERIALIZE                        \
+                               | sd_balance_for_mc_power()             \
+                               | sd_power_saving_flags()               \
+                               ,                                       \
+       .last_balance           = jiffies,                              \
+       .balance_interval       = 1,                                    \
+}
+#endif
+#endif /* CONFIG_SCHED_MC */
+
 /* Common values for CPUs */
 #ifndef SD_CPU_INIT
-#define SD_CPU_INIT (struct sched_domain) {            \
-       .span                   = CPU_MASK_NONE,        \
-       .parent                 = NULL,                 \
-       .groups                 = NULL,                 \
-       .min_interval           = 1,                    \
-       .max_interval           = 4,                    \
-       .busy_factor            = 64,                   \
-       .imbalance_pct          = 125,                  \
-       .cache_hot_time         = (5*1000000/2),        \
-       .cache_nice_tries       = 1,                    \
-       .per_cpu_gain           = 100,                  \
-       .busy_idx               = 2,                    \
-       .idle_idx               = 1,                    \
-       .newidle_idx            = 2,                    \
-       .wake_idx               = 1,                    \
-       .forkexec_idx           = 1,                    \
-       .flags                  = SD_LOAD_BALANCE       \
-                               | SD_BALANCE_NEWIDLE    \
-                               | SD_BALANCE_EXEC       \
-                               | SD_WAKE_AFFINE,       \
-       .last_balance           = jiffies,              \
-       .balance_interval       = 1,                    \
-       .nr_balance_failed      = 0,                    \
+#define SD_CPU_INIT (struct sched_domain) {                            \
+       .min_interval           = 1,                                    \
+       .max_interval           = 4,                                    \
+       .busy_factor            = 64,                                   \
+       .imbalance_pct          = 125,                                  \
+       .cache_nice_tries       = 1,                                    \
+       .busy_idx               = 2,                                    \
+       .idle_idx               = 1,                                    \
+       .newidle_idx            = 0,                                    \
+       .wake_idx               = 0,                                    \
+       .forkexec_idx           = 0,                                    \
+                                                                       \
+       .flags                  = 1*SD_LOAD_BALANCE                     \
+                               | 1*SD_BALANCE_NEWIDLE                  \
+                               | 1*SD_BALANCE_EXEC                     \
+                               | 1*SD_BALANCE_FORK                     \
+                               | 0*SD_BALANCE_WAKE                     \
+                               | 1*SD_WAKE_AFFINE                      \
+                               | 0*SD_PREFER_LOCAL                     \
+                               | 0*SD_SHARE_CPUPOWER                   \
+                               | 0*SD_SHARE_PKG_RESOURCES              \
+                               | 0*SD_SERIALIZE                        \
+                               | sd_balance_for_package_power()        \
+                               | sd_power_saving_flags()               \
+                               ,                                       \
+       .last_balance           = jiffies,                              \
+       .balance_interval       = 1,                                    \
 }
 #endif
 
 /* sched_domains SD_ALLNODES_INIT for NUMA machines */
-#define SD_ALLNODES_INIT (struct sched_domain) {       \
-       .span                   = CPU_MASK_NONE,        \
-       .parent                 = NULL,                 \
-       .groups                 = NULL,                 \
-       .min_interval           = 64,                   \
-       .max_interval           = 64*num_online_cpus(), \
-       .busy_factor            = 128,                  \
-       .imbalance_pct          = 133,                  \
-       .cache_hot_time         = (10*1000000),         \
-       .cache_nice_tries       = 1,                    \
-       .busy_idx               = 3,                    \
-       .idle_idx               = 3,                    \
-       .newidle_idx            = 0, /* unused */       \
-       .wake_idx               = 0, /* unused */       \
-       .forkexec_idx           = 0, /* unused */       \
-       .per_cpu_gain           = 100,                  \
-       .flags                  = SD_LOAD_BALANCE,      \
-       .last_balance           = jiffies,              \
-       .balance_interval       = 64,                   \
-       .nr_balance_failed      = 0,                    \
+#define SD_ALLNODES_INIT (struct sched_domain) {                       \
+       .min_interval           = 64,                                   \
+       .max_interval           = 64*num_online_cpus(),                 \
+       .busy_factor            = 128,                                  \
+       .imbalance_pct          = 133,                                  \
+       .cache_nice_tries       = 1,                                    \
+       .busy_idx               = 3,                                    \
+       .idle_idx               = 3,                                    \
+       .flags                  = 1*SD_LOAD_BALANCE                     \
+                               | 1*SD_BALANCE_NEWIDLE                  \
+                               | 0*SD_BALANCE_EXEC                     \
+                               | 0*SD_BALANCE_FORK                     \
+                               | 0*SD_BALANCE_WAKE                     \
+                               | 0*SD_WAKE_AFFINE                      \
+                               | 0*SD_SHARE_CPUPOWER                   \
+                               | 0*SD_POWERSAVINGS_BALANCE             \
+                               | 0*SD_SHARE_PKG_RESOURCES              \
+                               | 1*SD_SERIALIZE                        \
+                               | 0*SD_PREFER_SIBLING                   \
+                               ,                                       \
+       .last_balance           = jiffies,                              \
+       .balance_interval       = 64,                                   \
 }
 
 #ifdef CONFIG_NUMA
 #ifndef SD_NODE_INIT
 #error Please define an appropriate SD_NODE_INIT in include/asm/topology.h!!!
 #endif
+
 #endif /* CONFIG_NUMA */
 
+#ifdef CONFIG_USE_PERCPU_NUMA_NODE_ID
+DECLARE_PER_CPU(int, numa_node);
+
+#ifndef numa_node_id
+/* Returns the number of the current Node. */
+static inline int numa_node_id(void)
+{
+       return __this_cpu_read(numa_node);
+}
+#endif
+
+#ifndef cpu_to_node
+static inline int cpu_to_node(int cpu)
+{
+       return per_cpu(numa_node, cpu);
+}
+#endif
+
+#ifndef set_numa_node
+static inline void set_numa_node(int node)
+{
+       percpu_write(numa_node, node);
+}
+#endif
+
+#ifndef set_cpu_numa_node
+static inline void set_cpu_numa_node(int cpu, int node)
+{
+       per_cpu(numa_node, cpu) = node;
+}
+#endif
+
+#else  /* !CONFIG_USE_PERCPU_NUMA_NODE_ID */
+
+/* Returns the number of the current Node. */
+#ifndef numa_node_id
+static inline int numa_node_id(void)
+{
+       return cpu_to_node(raw_smp_processor_id());
+}
+#endif
+
+#endif /* [!]CONFIG_USE_PERCPU_NUMA_NODE_ID */
+
+#ifdef CONFIG_HAVE_MEMORYLESS_NODES
+
+/*
+ * N.B., Do NOT reference the '_numa_mem_' per cpu variable directly.
+ * It will not be defined when CONFIG_HAVE_MEMORYLESS_NODES is not defined.
+ * Use the accessor functions set_numa_mem(), numa_mem_id() and cpu_to_mem().
+ */
+DECLARE_PER_CPU(int, _numa_mem_);
+
+#ifndef set_numa_mem
+static inline void set_numa_mem(int node)
+{
+       percpu_write(_numa_mem_, node);
+}
+#endif
+
+#ifndef numa_mem_id
+/* Returns the number of the nearest Node with memory */
+static inline int numa_mem_id(void)
+{
+       return __this_cpu_read(_numa_mem_);
+}
+#endif
+
+#ifndef cpu_to_mem
+static inline int cpu_to_mem(int cpu)
+{
+       return per_cpu(_numa_mem_, cpu);
+}
+#endif
+
+#ifndef set_cpu_numa_mem
+static inline void set_cpu_numa_mem(int cpu, int node)
+{
+       per_cpu(_numa_mem_, cpu) = node;
+}
+#endif
+
+#else  /* !CONFIG_HAVE_MEMORYLESS_NODES */
+
+static inline void set_numa_mem(int node) {}
+
+static inline void set_cpu_numa_mem(int cpu, int node) {}
+
+#ifndef numa_mem_id
+/* Returns the number of the nearest Node with memory */
+static inline int numa_mem_id(void)
+{
+       return numa_node_id();
+}
+#endif
+
+#ifndef cpu_to_mem
+static inline int cpu_to_mem(int cpu)
+{
+       return cpu_to_node(cpu);
+}
+#endif
+
+#endif /* [!]CONFIG_HAVE_MEMORYLESS_NODES */
+
+#ifndef topology_physical_package_id
+#define topology_physical_package_id(cpu)      ((void)(cpu), -1)
+#endif
+#ifndef topology_core_id
+#define topology_core_id(cpu)                  ((void)(cpu), 0)
+#endif
+#ifndef topology_thread_cpumask
+#define topology_thread_cpumask(cpu)           cpumask_of(cpu)
+#endif
+#ifndef topology_core_cpumask
+#define topology_core_cpumask(cpu)             cpumask_of(cpu)
+#endif
+
 #endif /* _LINUX_TOPOLOGY_H */