diff --git a/arch/ia64/include/asm/topology.h b/arch/ia64/include/asm/topology.h index 7b4c8c70b2d1..d0141fbf51d0 100644 --- a/arch/ia64/include/asm/topology.h +++ b/arch/ia64/include/asm/topology.h @@ -61,12 +61,13 @@ void build_cpu_to_node_map(void); .cache_nice_tries = 2, \ .busy_idx = 2, \ .idle_idx = 1, \ - .newidle_idx = 2, \ - .wake_idx = 1, \ - .forkexec_idx = 1, \ + .newidle_idx = 0, \ + .wake_idx = 0, \ + .forkexec_idx = 0, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ + | SD_BALANCE_FORK \ | SD_WAKE_AFFINE, \ .last_balance = jiffies, \ .balance_interval = 1, \ @@ -85,14 +86,14 @@ void build_cpu_to_node_map(void); .cache_nice_tries = 2, \ .busy_idx = 3, \ .idle_idx = 2, \ - .newidle_idx = 2, \ - .wake_idx = 1, \ - .forkexec_idx = 1, \ + .newidle_idx = 0, \ + .wake_idx = 0, \ + .forkexec_idx = 0, \ .flags = SD_LOAD_BALANCE \ + | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ | SD_BALANCE_FORK \ - | SD_SERIALIZE \ - | SD_WAKE_BALANCE, \ + | SD_SERIALIZE, \ .last_balance = jiffies, \ .balance_interval = 64, \ .nr_balance_failed = 0, \ diff --git a/arch/mips/include/asm/mach-ip27/topology.h b/arch/mips/include/asm/mach-ip27/topology.h index 07547231e078..230591707005 100644 --- a/arch/mips/include/asm/mach-ip27/topology.h +++ b/arch/mips/include/asm/mach-ip27/topology.h @@ -48,7 +48,6 @@ extern unsigned char __node_distances[MAX_COMPACT_NODES][MAX_COMPACT_NODES]; .cache_nice_tries = 1, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_EXEC \ - | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ .balance_interval = 1, \ .nr_balance_failed = 0, \ diff --git a/arch/powerpc/include/asm/topology.h b/arch/powerpc/include/asm/topology.h index 054a16d68082..394edcbcce71 100644 --- a/arch/powerpc/include/asm/topology.h +++ b/arch/powerpc/include/asm/topology.h @@ -57,14 +57,13 @@ static inline int pcibus_to_node(struct pci_bus *bus) .cache_nice_tries = 1, \ .busy_idx = 3, \ .idle_idx = 1, \ - .newidle_idx = 2, \ - .wake_idx = 1, \ + .newidle_idx = 0, \ + .wake_idx = 0, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_EXEC \ + | SD_BALANCE_FORK \ | SD_BALANCE_NEWIDLE \ - | SD_WAKE_IDLE \ - | SD_SERIALIZE \ - | SD_WAKE_BALANCE, \ + | SD_SERIALIZE, \ .last_balance = jiffies, \ .balance_interval = 1, \ .nr_balance_failed = 0, \ diff --git a/arch/sh/include/asm/topology.h b/arch/sh/include/asm/topology.h index b69ee850906d..f8c40cc65054 100644 --- a/arch/sh/include/asm/topology.h +++ b/arch/sh/include/asm/topology.h @@ -15,14 +15,14 @@ .cache_nice_tries = 2, \ .busy_idx = 3, \ .idle_idx = 2, \ - .newidle_idx = 2, \ - .wake_idx = 1, \ - .forkexec_idx = 1, \ + .newidle_idx = 0, \ + .wake_idx = 0, \ + .forkexec_idx = 0, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_FORK \ | SD_BALANCE_EXEC \ - | SD_SERIALIZE \ - | SD_WAKE_BALANCE, \ + | SD_BALANCE_NEWIDLE \ + | SD_SERIALIZE, \ .last_balance = jiffies, \ .balance_interval = 1, \ .nr_balance_failed = 0, \ diff --git a/arch/sparc/include/asm/topology_64.h b/arch/sparc/include/asm/topology_64.h index e5ea8d332421..26cd25c08399 100644 --- a/arch/sparc/include/asm/topology_64.h +++ b/arch/sparc/include/asm/topology_64.h @@ -52,13 +52,12 @@ static inline int pcibus_to_node(struct pci_bus *pbus) .busy_idx = 3, \ .idle_idx = 2, \ .newidle_idx = 0, \ - .wake_idx = 1, \ - .forkexec_idx = 1, \ + .wake_idx = 0, \ + .forkexec_idx = 0, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_FORK \ | SD_BALANCE_EXEC \ - | SD_SERIALIZE \ - | SD_WAKE_BALANCE, \ + | SD_SERIALIZE, \ .last_balance = jiffies, \ .balance_interval = 1, \ } diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index 847fee6493a2..9cfc88b97742 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -96,6 +96,7 @@ #define X86_FEATURE_CLFLUSH_MONITOR (3*32+25) /* "" clflush reqd with monitor */ #define X86_FEATURE_EXTD_APICID (3*32+26) /* has extended APICID (8 bits) */ #define X86_FEATURE_AMD_DCM (3*32+27) /* multi-node processor */ +#define X86_FEATURE_APERFMPERF (3*32+28) /* APERFMPERF */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 (4*32+ 0) /* "pni" SSE-3 */ diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 42a3f936dadc..c3429e8b2424 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -27,6 +27,7 @@ struct mm_struct; #include #include #include +#include #include /* @@ -1022,4 +1023,33 @@ extern int set_tsc_mode(unsigned int val); extern int amd_get_nb_id(int cpu); +struct aperfmperf { + u64 aperf, mperf; +}; + +static inline void get_aperfmperf(struct aperfmperf *am) +{ + WARN_ON_ONCE(!boot_cpu_has(X86_FEATURE_APERFMPERF)); + + rdmsrl(MSR_IA32_APERF, am->aperf); + rdmsrl(MSR_IA32_MPERF, am->mperf); +} + +#define APERFMPERF_SHIFT 10 + +static inline +unsigned long calc_aperfmperf_ratio(struct aperfmperf *old, + struct aperfmperf *new) +{ + u64 aperf = new->aperf - old->aperf; + u64 mperf = new->mperf - old->mperf; + unsigned long ratio = aperf; + + mperf >>= APERFMPERF_SHIFT; + if (mperf) + ratio = div64_u64(aperf, mperf); + + return ratio; +} + #endif /* _ASM_X86_PROCESSOR_H */ diff --git a/arch/x86/include/asm/topology.h b/arch/x86/include/asm/topology.h index 26d06e052a18..6f0695d744bf 100644 --- a/arch/x86/include/asm/topology.h +++ b/arch/x86/include/asm/topology.h @@ -116,15 +116,11 @@ extern unsigned long node_remap_size[]; # define SD_CACHE_NICE_TRIES 1 # define SD_IDLE_IDX 1 -# define SD_NEWIDLE_IDX 2 -# define SD_FORKEXEC_IDX 0 #else # define SD_CACHE_NICE_TRIES 2 # define SD_IDLE_IDX 2 -# define SD_NEWIDLE_IDX 2 -# define SD_FORKEXEC_IDX 1 #endif @@ -137,22 +133,20 @@ extern unsigned long node_remap_size[]; .cache_nice_tries = SD_CACHE_NICE_TRIES, \ .busy_idx = 3, \ .idle_idx = SD_IDLE_IDX, \ - .newidle_idx = SD_NEWIDLE_IDX, \ - .wake_idx = 1, \ - .forkexec_idx = SD_FORKEXEC_IDX, \ + .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_WAKE_IDLE \ + | 0*SD_BALANCE_WAKE \ | 1*SD_WAKE_AFFINE \ - | 1*SD_WAKE_BALANCE \ | 0*SD_SHARE_CPUPOWER \ | 0*SD_POWERSAVINGS_BALANCE \ | 0*SD_SHARE_PKG_RESOURCES \ | 1*SD_SERIALIZE \ - | 1*SD_WAKE_IDLE_FAR \ | 0*SD_PREFER_SIBLING \ , \ .last_balance = jiffies, \ diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index c1f253dac155..8dd30638fe44 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -13,7 +13,7 @@ CFLAGS_common.o := $(nostackp) obj-y := intel_cacheinfo.o addon_cpuid_features.o obj-y += proc.o capflags.o powerflags.o common.o -obj-y += vmware.o hypervisor.o +obj-y += vmware.o hypervisor.o sched.o obj-$(CONFIG_X86_32) += bugs.o cmpxchg.o obj-$(CONFIG_X86_64) += bugs_64.o diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index ae9b503220ca..4109679863c1 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -60,7 +60,6 @@ enum { }; #define INTEL_MSR_RANGE (0xffff) -#define CPUID_6_ECX_APERFMPERF_CAPABILITY (0x1) struct acpi_cpufreq_data { struct acpi_processor_performance *acpi_data; @@ -71,11 +70,7 @@ struct acpi_cpufreq_data { static DEFINE_PER_CPU(struct acpi_cpufreq_data *, drv_data); -struct acpi_msr_data { - u64 saved_aperf, saved_mperf; -}; - -static DEFINE_PER_CPU(struct acpi_msr_data, msr_data); +static DEFINE_PER_CPU(struct aperfmperf, old_perf); DEFINE_TRACE(power_mark); @@ -244,23 +239,12 @@ static u32 get_cur_val(const struct cpumask *mask) return cmd.val; } -struct perf_pair { - union { - struct { - u32 lo; - u32 hi; - } split; - u64 whole; - } aperf, mperf; -}; - /* Called via smp_call_function_single(), on the target CPU */ static void read_measured_perf_ctrs(void *_cur) { - struct perf_pair *cur = _cur; + struct aperfmperf *am = _cur; - rdmsr(MSR_IA32_APERF, cur->aperf.split.lo, cur->aperf.split.hi); - rdmsr(MSR_IA32_MPERF, cur->mperf.split.lo, cur->mperf.split.hi); + get_aperfmperf(am); } /* @@ -279,63 +263,17 @@ static void read_measured_perf_ctrs(void *_cur) static unsigned int get_measured_perf(struct cpufreq_policy *policy, unsigned int cpu) { - struct perf_pair readin, cur; - unsigned int perf_percent; + struct aperfmperf perf; + unsigned long ratio; unsigned int retval; - if (smp_call_function_single(cpu, read_measured_perf_ctrs, &readin, 1)) + if (smp_call_function_single(cpu, read_measured_perf_ctrs, &perf, 1)) return 0; - cur.aperf.whole = readin.aperf.whole - - per_cpu(msr_data, cpu).saved_aperf; - cur.mperf.whole = readin.mperf.whole - - per_cpu(msr_data, cpu).saved_mperf; - per_cpu(msr_data, cpu).saved_aperf = readin.aperf.whole; - per_cpu(msr_data, cpu).saved_mperf = readin.mperf.whole; + ratio = calc_aperfmperf_ratio(&per_cpu(old_perf, cpu), &perf); + per_cpu(old_perf, cpu) = perf; -#ifdef __i386__ - /* - * We dont want to do 64 bit divide with 32 bit kernel - * Get an approximate value. Return failure in case we cannot get - * an approximate value. - */ - if (unlikely(cur.aperf.split.hi || cur.mperf.split.hi)) { - int shift_count; - u32 h; - - h = max_t(u32, cur.aperf.split.hi, cur.mperf.split.hi); - shift_count = fls(h); - - cur.aperf.whole >>= shift_count; - cur.mperf.whole >>= shift_count; - } - - if (((unsigned long)(-1) / 100) < cur.aperf.split.lo) { - int shift_count = 7; - cur.aperf.split.lo >>= shift_count; - cur.mperf.split.lo >>= shift_count; - } - - if (cur.aperf.split.lo && cur.mperf.split.lo) - perf_percent = (cur.aperf.split.lo * 100) / cur.mperf.split.lo; - else - perf_percent = 0; - -#else - if (unlikely(((unsigned long)(-1) / 100) < cur.aperf.whole)) { - int shift_count = 7; - cur.aperf.whole >>= shift_count; - cur.mperf.whole >>= shift_count; - } - - if (cur.aperf.whole && cur.mperf.whole) - perf_percent = (cur.aperf.whole * 100) / cur.mperf.whole; - else - perf_percent = 0; - -#endif - - retval = (policy->cpuinfo.max_freq * perf_percent) / 100; + retval = (policy->cpuinfo.max_freq * ratio) >> APERFMPERF_SHIFT; return retval; } @@ -731,12 +669,8 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy) acpi_processor_notify_smm(THIS_MODULE); /* Check for APERF/MPERF support in hardware */ - if (c->x86_vendor == X86_VENDOR_INTEL && c->cpuid_level >= 6) { - unsigned int ecx; - ecx = cpuid_ecx(6); - if (ecx & CPUID_6_ECX_APERFMPERF_CAPABILITY) - acpi_cpufreq_driver.getavg = get_measured_perf; - } + if (cpu_has(c, X86_FEATURE_APERFMPERF)) + acpi_cpufreq_driver.getavg = get_measured_perf; dprintk("CPU%u - ACPI performance management activated.\n", cpu); for (i = 0; i < perf->state_count; i++) diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 80a722a071b5..40e1835b35e8 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -350,6 +350,12 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) set_cpu_cap(c, X86_FEATURE_ARCH_PERFMON); } + if (c->cpuid_level > 6) { + unsigned ecx = cpuid_ecx(6); + if (ecx & 0x01) + set_cpu_cap(c, X86_FEATURE_APERFMPERF); + } + if (cpu_has_xmm2) set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC); if (cpu_has_ds) { diff --git a/arch/x86/kernel/cpu/sched.c b/arch/x86/kernel/cpu/sched.c new file mode 100644 index 000000000000..a640ae5ad201 --- /dev/null +++ b/arch/x86/kernel/cpu/sched.c @@ -0,0 +1,55 @@ +#include +#include +#include +#include + +#include +#include + +#ifdef CONFIG_SMP + +static DEFINE_PER_CPU(struct aperfmperf, old_perf_sched); + +static unsigned long scale_aperfmperf(void) +{ + struct aperfmperf val, *old = &__get_cpu_var(old_perf_sched); + unsigned long ratio, flags; + + local_irq_save(flags); + get_aperfmperf(&val); + local_irq_restore(flags); + + ratio = calc_aperfmperf_ratio(old, &val); + *old = val; + + return ratio; +} + +unsigned long arch_scale_freq_power(struct sched_domain *sd, int cpu) +{ + /* + * do aperf/mperf on the cpu level because it includes things + * like turbo mode, which are relevant to full cores. + */ + if (boot_cpu_has(X86_FEATURE_APERFMPERF)) + return scale_aperfmperf(); + + /* + * maybe have something cpufreq here + */ + + return default_scale_freq_power(sd, cpu); +} + +unsigned long arch_scale_smt_power(struct sched_domain *sd, int cpu) +{ + /* + * aperf/mperf already includes the smt gain + */ + if (boot_cpu_has(X86_FEATURE_APERFMPERF)) + return SCHED_LOAD_SCALE; + + return default_scale_smt_power(sd, cpu); +} + +#endif diff --git a/include/linux/sched.h b/include/linux/sched.h index f3d74bd04d18..8af3d249170e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -190,6 +190,7 @@ extern unsigned long long time_sync_thresh; /* in tsk->state again */ #define TASK_DEAD 64 #define TASK_WAKEKILL 128 +#define TASK_WAKING 256 /* Convenience macros for the sake of set_task_state */ #define TASK_KILLABLE (TASK_WAKEKILL | TASK_UNINTERRUPTIBLE) @@ -802,14 +803,14 @@ enum cpu_idle_type { #define SD_BALANCE_NEWIDLE 0x0002 /* Balance when about to become idle */ #define SD_BALANCE_EXEC 0x0004 /* Balance on exec */ #define SD_BALANCE_FORK 0x0008 /* Balance on fork, clone */ -#define SD_WAKE_IDLE 0x0010 /* Wake to idle CPU on task wakeup */ +#define SD_BALANCE_WAKE 0x0010 /* Balance on wakeup */ #define SD_WAKE_AFFINE 0x0020 /* Wake task to waking CPU */ -#define SD_WAKE_BALANCE 0x0040 /* Perform balancing at task wakeup */ +#define SD_PREFER_LOCAL 0x0040 /* Prefer to keep tasks local to this domain */ #define SD_SHARE_CPUPOWER 0x0080 /* Domain members share cpu power */ #define SD_POWERSAVINGS_BALANCE 0x0100 /* Balance for power savings */ #define SD_SHARE_PKG_RESOURCES 0x0200 /* Domain members share cpu pkg resources */ #define SD_SERIALIZE 0x0400 /* Only a single load balancing instance */ -#define SD_WAKE_IDLE_FAR 0x0800 /* Gain latency sacrificing cache hit */ + #define SD_PREFER_SIBLING 0x1000 /* Prefer to place tasks in a sibling domain */ enum powersavings_balance_level { @@ -991,6 +992,9 @@ static inline int test_sd_parent(struct sched_domain *sd, int flag) return 0; } +unsigned long default_scale_freq_power(struct sched_domain *sd, int cpu); +unsigned long default_scale_smt_power(struct sched_domain *sd, int cpu); + #else /* CONFIG_SMP */ struct sched_domain_attr; @@ -1002,6 +1006,7 @@ partition_sched_domains(int ndoms_new, struct cpumask *doms_new, } #endif /* !CONFIG_SMP */ + struct io_context; /* See blkdev.h */ @@ -1019,6 +1024,12 @@ struct uts_namespace; struct rq; struct sched_domain; +/* + * wake flags + */ +#define WF_SYNC 0x01 /* waker goes to sleep after wakup */ +#define WF_FORK 0x02 /* child wakeup after fork */ + struct sched_class { const struct sched_class *next; @@ -1026,13 +1037,13 @@ struct sched_class { void (*dequeue_task) (struct rq *rq, struct task_struct *p, int sleep); void (*yield_task) (struct rq *rq); - void (*check_preempt_curr) (struct rq *rq, struct task_struct *p, int sync); + void (*check_preempt_curr) (struct rq *rq, struct task_struct *p, int flags); struct task_struct * (*pick_next_task) (struct rq *rq); void (*put_prev_task) (struct rq *rq, struct task_struct *p); #ifdef CONFIG_SMP - int (*select_task_rq)(struct task_struct *p, int sync); + int (*select_task_rq)(struct task_struct *p, int sd_flag, int flags); unsigned long (*load_balance) (struct rq *this_rq, int this_cpu, struct rq *busiest, unsigned long max_load_move, @@ -1102,6 +1113,8 @@ struct sched_entity { u64 start_runtime; u64 avg_wakeup; + u64 avg_running; + #ifdef CONFIG_SCHEDSTATS u64 wait_start; u64 wait_max; diff --git a/include/linux/topology.h b/include/linux/topology.h index 85e8cf7d393c..809b26c07090 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -95,14 +95,12 @@ int arch_update_cpu_topology(void); | 1*SD_BALANCE_NEWIDLE \ | 1*SD_BALANCE_EXEC \ | 1*SD_BALANCE_FORK \ - | 0*SD_WAKE_IDLE \ + | 0*SD_BALANCE_WAKE \ | 1*SD_WAKE_AFFINE \ - | 1*SD_WAKE_BALANCE \ | 1*SD_SHARE_CPUPOWER \ | 0*SD_POWERSAVINGS_BALANCE \ | 0*SD_SHARE_PKG_RESOURCES \ | 0*SD_SERIALIZE \ - | 0*SD_WAKE_IDLE_FAR \ | 0*SD_PREFER_SIBLING \ , \ .last_balance = jiffies, \ @@ -122,20 +120,19 @@ int arch_update_cpu_topology(void); .imbalance_pct = 125, \ .cache_nice_tries = 1, \ .busy_idx = 2, \ - .wake_idx = 1, \ - .forkexec_idx = 1, \ + .wake_idx = 0, \ + .forkexec_idx = 0, \ \ .flags = 1*SD_LOAD_BALANCE \ | 1*SD_BALANCE_NEWIDLE \ | 1*SD_BALANCE_EXEC \ | 1*SD_BALANCE_FORK \ - | 1*SD_WAKE_IDLE \ + | 0*SD_BALANCE_WAKE \ | 1*SD_WAKE_AFFINE \ - | 1*SD_WAKE_BALANCE \ + | 1*SD_PREFER_LOCAL \ | 0*SD_SHARE_CPUPOWER \ | 1*SD_SHARE_PKG_RESOURCES \ | 0*SD_SERIALIZE \ - | 0*SD_WAKE_IDLE_FAR \ | sd_balance_for_mc_power() \ | sd_power_saving_flags() \ , \ @@ -155,21 +152,20 @@ int arch_update_cpu_topology(void); .cache_nice_tries = 1, \ .busy_idx = 2, \ .idle_idx = 1, \ - .newidle_idx = 2, \ - .wake_idx = 1, \ - .forkexec_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 \ - | 1*SD_WAKE_IDLE \ - | 0*SD_WAKE_AFFINE \ - | 1*SD_WAKE_BALANCE \ + | 0*SD_BALANCE_WAKE \ + | 1*SD_WAKE_AFFINE \ + | 1*SD_PREFER_LOCAL \ | 0*SD_SHARE_CPUPOWER \ | 0*SD_SHARE_PKG_RESOURCES \ | 0*SD_SERIALIZE \ - | 0*SD_WAKE_IDLE_FAR \ | sd_balance_for_package_power() \ | sd_power_saving_flags() \ , \ @@ -191,14 +187,12 @@ int arch_update_cpu_topology(void); | 1*SD_BALANCE_NEWIDLE \ | 0*SD_BALANCE_EXEC \ | 0*SD_BALANCE_FORK \ - | 0*SD_WAKE_IDLE \ - | 1*SD_WAKE_AFFINE \ - | 0*SD_WAKE_BALANCE \ + | 0*SD_BALANCE_WAKE \ + | 0*SD_WAKE_AFFINE \ | 0*SD_SHARE_CPUPOWER \ | 0*SD_POWERSAVINGS_BALANCE \ | 0*SD_SHARE_PKG_RESOURCES \ | 1*SD_SERIALIZE \ - | 1*SD_WAKE_IDLE_FAR \ | 0*SD_PREFER_SIBLING \ , \ .last_balance = jiffies, \ diff --git a/include/linux/wait.h b/include/linux/wait.h index cf3c2f5dba51..a48e16b77d5e 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -26,8 +26,8 @@ #include typedef struct __wait_queue wait_queue_t; -typedef int (*wait_queue_func_t)(wait_queue_t *wait, unsigned mode, int sync, void *key); -int default_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key); +typedef int (*wait_queue_func_t)(wait_queue_t *wait, unsigned mode, int flags, void *key); +int default_wake_function(wait_queue_t *wait, unsigned mode, int flags, void *key); struct __wait_queue { unsigned int flags; diff --git a/kernel/sched.c b/kernel/sched.c index d9db3fb17573..faf4d463bbff 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -119,8 +119,6 @@ */ #define RUNTIME_INF ((u64)~0ULL) -static void double_rq_lock(struct rq *rq1, struct rq *rq2); - static inline int rt_policy(int policy) { if (unlikely(policy == SCHED_FIFO || policy == SCHED_RR)) @@ -378,13 +376,6 @@ static inline void set_task_rq(struct task_struct *p, unsigned int cpu) #else -#ifdef CONFIG_SMP -static int root_task_group_empty(void) -{ - return 1; -} -#endif - static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { } static inline struct task_group *task_group(struct task_struct *p) { @@ -514,14 +505,6 @@ struct root_domain { #ifdef CONFIG_SMP struct cpupri cpupri; #endif -#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) - /* - * Preferred wake up cpu nominated by sched_mc balance that will be - * used when most cpus are idle in the system indicating overall very - * low system utilisation. Triggered at POWERSAVINGS_BALANCE_WAKEUP(2) - */ - unsigned int sched_mc_preferred_wakeup_cpu; -#endif }; /* @@ -646,9 +629,10 @@ struct rq { static DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); -static inline void check_preempt_curr(struct rq *rq, struct task_struct *p, int sync) +static inline +void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) { - rq->curr->sched_class->check_preempt_curr(rq, p, sync); + rq->curr->sched_class->check_preempt_curr(rq, p, flags); } static inline int cpu_of(struct rq *rq) @@ -1509,8 +1493,65 @@ static int tg_nop(struct task_group *tg, void *data) #endif #ifdef CONFIG_SMP -static unsigned long source_load(int cpu, int type); -static unsigned long target_load(int cpu, int type); +/* Used instead of source_load when we know the type == 0 */ +static unsigned long weighted_cpuload(const int cpu) +{ + return cpu_rq(cpu)->load.weight; +} + +/* + * Return a low guess at the load of a migration-source cpu weighted + * according to the scheduling class and "nice" value. + * + * We want to under-estimate the load of migration sources, to + * balance conservatively. + */ +static unsigned long source_load(int cpu, int type) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long total = weighted_cpuload(cpu); + + if (type == 0 || !sched_feat(LB_BIAS)) + return total; + + return min(rq->cpu_load[type-1], total); +} + +/* + * Return a high guess at the load of a migration-target cpu weighted + * according to the scheduling class and "nice" value. + */ +static unsigned long target_load(int cpu, int type) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long total = weighted_cpuload(cpu); + + if (type == 0 || !sched_feat(LB_BIAS)) + return total; + + return max(rq->cpu_load[type-1], total); +} + +static struct sched_group *group_of(int cpu) +{ + struct sched_domain *sd = rcu_dereference(cpu_rq(cpu)->sd); + + if (!sd) + return NULL; + + return sd->groups; +} + +static unsigned long power_of(int cpu) +{ + struct sched_group *group = group_of(cpu); + + if (!group) + return SCHED_LOAD_SCALE; + + return group->cpu_power; +} + static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd); static unsigned long cpu_avg_load_per_task(int cpu) @@ -1695,6 +1736,8 @@ static inline void update_shares_locked(struct rq *rq, struct sched_domain *sd) #ifdef CONFIG_PREEMPT +static void double_rq_lock(struct rq *rq1, struct rq *rq2); + /* * fair double_lock_balance: Safely acquires both rq->locks in a fair * way at the expense of forcing extra atomic operations in all @@ -1959,13 +2002,6 @@ static inline void check_class_changed(struct rq *rq, struct task_struct *p, } #ifdef CONFIG_SMP - -/* Used instead of source_load when we know the type == 0 */ -static unsigned long weighted_cpuload(const int cpu) -{ - return cpu_rq(cpu)->load.weight; -} - /* * Is this task likely cache-hot: */ @@ -2239,185 +2275,6 @@ void kick_process(struct task_struct *p) preempt_enable(); } EXPORT_SYMBOL_GPL(kick_process); - -/* - * Return a low guess at the load of a migration-source cpu weighted - * according to the scheduling class and "nice" value. - * - * We want to under-estimate the load of migration sources, to - * balance conservatively. - */ -static unsigned long source_load(int cpu, int type) -{ - struct rq *rq = cpu_rq(cpu); - unsigned long total = weighted_cpuload(cpu); - - if (type == 0 || !sched_feat(LB_BIAS)) - return total; - - return min(rq->cpu_load[type-1], total); -} - -/* - * Return a high guess at the load of a migration-target cpu weighted - * according to the scheduling class and "nice" value. - */ -static unsigned long target_load(int cpu, int type) -{ - struct rq *rq = cpu_rq(cpu); - unsigned long total = weighted_cpuload(cpu); - - if (type == 0 || !sched_feat(LB_BIAS)) - return total; - - return max(rq->cpu_load[type-1], total); -} - -/* - * find_idlest_group finds and returns the least busy CPU group within the - * domain. - */ -static struct sched_group * -find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu) -{ - struct sched_group *idlest = NULL, *this = NULL, *group = sd->groups; - unsigned long min_load = ULONG_MAX, this_load = 0; - int load_idx = sd->forkexec_idx; - int imbalance = 100 + (sd->imbalance_pct-100)/2; - - do { - unsigned long load, avg_load; - int local_group; - int i; - - /* Skip over this group if it has no CPUs allowed */ - if (!cpumask_intersects(sched_group_cpus(group), - &p->cpus_allowed)) - continue; - - local_group = cpumask_test_cpu(this_cpu, - sched_group_cpus(group)); - - /* Tally up the load of all CPUs in the group */ - avg_load = 0; - - for_each_cpu(i, sched_group_cpus(group)) { - /* Bias balancing toward cpus of our domain */ - if (local_group) - load = source_load(i, load_idx); - else - load = target_load(i, load_idx); - - avg_load += load; - } - - /* Adjust by relative CPU power of the group */ - avg_load = (avg_load * SCHED_LOAD_SCALE) / group->cpu_power; - - if (local_group) { - this_load = avg_load; - this = group; - } else if (avg_load < min_load) { - min_load = avg_load; - idlest = group; - } - } while (group = group->next, group != sd->groups); - - if (!idlest || 100*this_load < imbalance*min_load) - return NULL; - return idlest; -} - -/* - * find_idlest_cpu - find the idlest cpu among the cpus in group. - */ -static int -find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) -{ - unsigned long load, min_load = ULONG_MAX; - int idlest = -1; - int i; - - /* Traverse only the allowed CPUs */ - for_each_cpu_and(i, sched_group_cpus(group), &p->cpus_allowed) { - load = weighted_cpuload(i); - - if (load < min_load || (load == min_load && i == this_cpu)) { - min_load = load; - idlest = i; - } - } - - return idlest; -} - -/* - * sched_balance_self: balance the current task (running on cpu) in domains - * that have the 'flag' flag set. In practice, this is SD_BALANCE_FORK and - * SD_BALANCE_EXEC. - * - * Balance, ie. select the least loaded group. - * - * Returns the target CPU number, or the same CPU if no balancing is needed. - * - * preempt must be disabled. - */ -static int sched_balance_self(int cpu, int flag) -{ - struct task_struct *t = current; - struct sched_domain *tmp, *sd = NULL; - - for_each_domain(cpu, tmp) { - /* - * If power savings logic is enabled for a domain, stop there. - */ - if (tmp->flags & SD_POWERSAVINGS_BALANCE) - break; - if (tmp->flags & flag) - sd = tmp; - } - - if (sd) - update_shares(sd); - - while (sd) { - struct sched_group *group; - int new_cpu, weight; - - if (!(sd->flags & flag)) { - sd = sd->child; - continue; - } - - group = find_idlest_group(sd, t, cpu); - if (!group) { - sd = sd->child; - continue; - } - - new_cpu = find_idlest_cpu(group, t, cpu); - if (new_cpu == -1 || new_cpu == cpu) { - /* Now try balancing at a lower domain level of cpu */ - sd = sd->child; - continue; - } - - /* Now try balancing at a lower domain level of new_cpu */ - cpu = new_cpu; - weight = cpumask_weight(sched_domain_span(sd)); - sd = NULL; - for_each_domain(cpu, tmp) { - if (weight <= cpumask_weight(sched_domain_span(tmp))) - break; - if (tmp->flags & flag) - sd = tmp; - } - /* while loop will break here if sd == NULL */ - } - - return cpu; -} - #endif /* CONFIG_SMP */ /** @@ -2455,37 +2312,22 @@ void task_oncpu_function_call(struct task_struct *p, * * returns failure only if the task is already active. */ -static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync) +static int try_to_wake_up(struct task_struct *p, unsigned int state, + int wake_flags) { int cpu, orig_cpu, this_cpu, success = 0; unsigned long flags; - long old_state; struct rq *rq; if (!sched_feat(SYNC_WAKEUPS)) - sync = 0; + wake_flags &= ~WF_SYNC; -#ifdef CONFIG_SMP - if (sched_feat(LB_WAKEUP_UPDATE) && !root_task_group_empty()) { - struct sched_domain *sd; - - this_cpu = raw_smp_processor_id(); - cpu = task_cpu(p); - - for_each_domain(this_cpu, sd) { - if (cpumask_test_cpu(cpu, sched_domain_span(sd))) { - update_shares(sd); - break; - } - } - } -#endif + this_cpu = get_cpu(); smp_wmb(); rq = task_rq_lock(p, &flags); update_rq_clock(rq); - old_state = p->state; - if (!(old_state & state)) + if (!(p->state & state)) goto out; if (p->se.on_rq) @@ -2493,27 +2335,29 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync) cpu = task_cpu(p); orig_cpu = cpu; - this_cpu = smp_processor_id(); #ifdef CONFIG_SMP if (unlikely(task_running(rq, p))) goto out_activate; - cpu = p->sched_class->select_task_rq(p, sync); - if (cpu != orig_cpu) { - set_task_cpu(p, cpu); - task_rq_unlock(rq, &flags); - /* might preempt at this point */ - rq = task_rq_lock(p, &flags); - old_state = p->state; - if (!(old_state & state)) - goto out; - if (p->se.on_rq) - goto out_running; + /* + * In order to handle concurrent wakeups and release the rq->lock + * we put the task in TASK_WAKING state. + * + * First fix up the nr_uninterruptible count: + */ + if (task_contributes_to_load(p)) + rq->nr_uninterruptible--; + p->state = TASK_WAKING; + task_rq_unlock(rq, &flags); - this_cpu = smp_processor_id(); - cpu = task_cpu(p); - } + cpu = p->sched_class->select_task_rq(p, SD_BALANCE_WAKE, wake_flags); + if (cpu != orig_cpu) + set_task_cpu(p, cpu); + + rq = task_rq_lock(p, &flags); + WARN_ON(p->state != TASK_WAKING); + cpu = task_cpu(p); #ifdef CONFIG_SCHEDSTATS schedstat_inc(rq, ttwu_count); @@ -2533,7 +2377,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync) out_activate: #endif /* CONFIG_SMP */ schedstat_inc(p, se.nr_wakeups); - if (sync) + if (wake_flags & WF_SYNC) schedstat_inc(p, se.nr_wakeups_sync); if (orig_cpu != cpu) schedstat_inc(p, se.nr_wakeups_migrate); @@ -2562,7 +2406,7 @@ out_activate: out_running: trace_sched_wakeup(rq, p, success); - check_preempt_curr(rq, p, sync); + check_preempt_curr(rq, p, wake_flags); p->state = TASK_RUNNING; #ifdef CONFIG_SMP @@ -2571,6 +2415,7 @@ out_running: #endif out: task_rq_unlock(rq, &flags); + put_cpu(); return success; } @@ -2613,6 +2458,7 @@ static void __sched_fork(struct task_struct *p) p->se.avg_overlap = 0; p->se.start_runtime = 0; p->se.avg_wakeup = sysctl_sched_wakeup_granularity; + p->se.avg_running = 0; #ifdef CONFIG_SCHEDSTATS p->se.wait_start = 0; @@ -2674,11 +2520,6 @@ void sched_fork(struct task_struct *p, int clone_flags) __sched_fork(p); -#ifdef CONFIG_SMP - cpu = sched_balance_self(cpu, SD_BALANCE_FORK); -#endif - set_task_cpu(p, cpu); - /* * Make sure we do not leak PI boosting priority to the child. */ @@ -2709,6 +2550,11 @@ void sched_fork(struct task_struct *p, int clone_flags) if (!rt_prio(p->prio)) p->sched_class = &fair_sched_class; +#ifdef CONFIG_SMP + cpu = p->sched_class->select_task_rq(p, SD_BALANCE_FORK, 0); +#endif + set_task_cpu(p, cpu); + #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) if (likely(sched_info_on())) memset(&p->sched_info, 0, sizeof(p->sched_info)); @@ -2754,7 +2600,7 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags) inc_nr_running(rq); } trace_sched_wakeup_new(rq, p, 1); - check_preempt_curr(rq, p, 0); + check_preempt_curr(rq, p, WF_FORK); #ifdef CONFIG_SMP if (p->sched_class->task_wake_up) p->sched_class->task_wake_up(rq, p); @@ -3263,7 +3109,7 @@ out: void sched_exec(void) { int new_cpu, this_cpu = get_cpu(); - new_cpu = sched_balance_self(this_cpu, SD_BALANCE_EXEC); + new_cpu = current->sched_class->select_task_rq(current, SD_BALANCE_EXEC, 0); put_cpu(); if (new_cpu != this_cpu) sched_migrate_task(current, new_cpu); @@ -3683,11 +3529,6 @@ static inline int check_power_save_busiest_group(struct sd_lb_stats *sds, *imbalance = sds->min_load_per_task; sds->busiest = sds->group_min; - if (sched_mc_power_savings >= POWERSAVINGS_BALANCE_WAKEUP) { - cpu_rq(this_cpu)->rd->sched_mc_preferred_wakeup_cpu = - group_first_cpu(sds->group_leader); - } - return 1; } @@ -3711,7 +3552,18 @@ static inline int check_power_save_busiest_group(struct sd_lb_stats *sds, } #endif /* CONFIG_SCHED_MC || CONFIG_SCHED_SMT */ -unsigned long __weak arch_scale_smt_power(struct sched_domain *sd, int cpu) + +unsigned long default_scale_freq_power(struct sched_domain *sd, int cpu) +{ + return SCHED_LOAD_SCALE; +} + +unsigned long __weak arch_scale_freq_power(struct sched_domain *sd, int cpu) +{ + return default_scale_freq_power(sd, cpu); +} + +unsigned long default_scale_smt_power(struct sched_domain *sd, int cpu) { unsigned long weight = cpumask_weight(sched_domain_span(sd)); unsigned long smt_gain = sd->smt_gain; @@ -3721,6 +3573,11 @@ unsigned long __weak arch_scale_smt_power(struct sched_domain *sd, int cpu) return smt_gain; } +unsigned long __weak arch_scale_smt_power(struct sched_domain *sd, int cpu) +{ + return default_scale_smt_power(sd, cpu); +} + unsigned long scale_rt_power(int cpu) { struct rq *rq = cpu_rq(cpu); @@ -3745,10 +3602,19 @@ static void update_cpu_power(struct sched_domain *sd, int cpu) unsigned long power = SCHED_LOAD_SCALE; struct sched_group *sdg = sd->groups; - /* here we could scale based on cpufreq */ + if (sched_feat(ARCH_POWER)) + power *= arch_scale_freq_power(sd, cpu); + else + power *= default_scale_freq_power(sd, cpu); + + power >>= SCHED_LOAD_SHIFT; if ((sd->flags & SD_SHARE_CPUPOWER) && weight > 1) { - power *= arch_scale_smt_power(sd, cpu); + if (sched_feat(ARCH_POWER)) + power *= arch_scale_smt_power(sd, cpu); + else + power *= default_scale_smt_power(sd, cpu); + power >>= SCHED_LOAD_SHIFT; } @@ -4161,26 +4027,6 @@ ret: return NULL; } -static struct sched_group *group_of(int cpu) -{ - struct sched_domain *sd = rcu_dereference(cpu_rq(cpu)->sd); - - if (!sd) - return NULL; - - return sd->groups; -} - -static unsigned long power_of(int cpu) -{ - struct sched_group *group = group_of(cpu); - - if (!group) - return SCHED_LOAD_SCALE; - - return group->cpu_power; -} - /* * find_busiest_queue - find the busiest runqueue among the cpus in group. */ @@ -5465,14 +5311,13 @@ static inline void schedule_debug(struct task_struct *prev) #endif } -static void put_prev_task(struct rq *rq, struct task_struct *prev) +static void put_prev_task(struct rq *rq, struct task_struct *p) { - if (prev->state == TASK_RUNNING) { - u64 runtime = prev->se.sum_exec_runtime; + u64 runtime = p->se.sum_exec_runtime - p->se.prev_sum_exec_runtime; - runtime -= prev->se.prev_sum_exec_runtime; - runtime = min_t(u64, runtime, 2*sysctl_sched_migration_cost); + update_avg(&p->se.avg_running, runtime); + if (p->state == TASK_RUNNING) { /* * In order to avoid avg_overlap growing stale when we are * indeed overlapping and hence not getting put to sleep, grow @@ -5482,9 +5327,12 @@ static void put_prev_task(struct rq *rq, struct task_struct *prev) * correlates to the amount of cache footprint a task can * build up. */ - update_avg(&prev->se.avg_overlap, runtime); + runtime = min_t(u64, runtime, 2*sysctl_sched_migration_cost); + update_avg(&p->se.avg_overlap, runtime); + } else { + update_avg(&p->se.avg_running, 0); } - prev->sched_class->put_prev_task(rq, prev); + p->sched_class->put_prev_task(rq, p); } /* @@ -5716,10 +5564,10 @@ asmlinkage void __sched preempt_schedule_irq(void) #endif /* CONFIG_PREEMPT */ -int default_wake_function(wait_queue_t *curr, unsigned mode, int sync, +int default_wake_function(wait_queue_t *curr, unsigned mode, int wake_flags, void *key) { - return try_to_wake_up(curr->private, mode, sync); + return try_to_wake_up(curr->private, mode, wake_flags); } EXPORT_SYMBOL(default_wake_function); @@ -5733,14 +5581,14 @@ EXPORT_SYMBOL(default_wake_function); * zero in this (rare) case, and we handle it by continuing to scan the queue. */ static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, - int nr_exclusive, int sync, void *key) + int nr_exclusive, int wake_flags, void *key) { wait_queue_t *curr, *next; list_for_each_entry_safe(curr, next, &q->task_list, task_list) { unsigned flags = curr->flags; - if (curr->func(curr, mode, sync, key) && + if (curr->func(curr, mode, wake_flags, key) && (flags & WQ_FLAG_EXCLUSIVE) && !--nr_exclusive) break; } @@ -5801,16 +5649,16 @@ void __wake_up_sync_key(wait_queue_head_t *q, unsigned int mode, int nr_exclusive, void *key) { unsigned long flags; - int sync = 1; + int wake_flags = WF_SYNC; if (unlikely(!q)) return; if (unlikely(!nr_exclusive)) - sync = 0; + wake_flags = 0; spin_lock_irqsave(&q->lock, flags); - __wake_up_common(q, mode, nr_exclusive, sync, key); + __wake_up_common(q, mode, nr_exclusive, wake_flags, key); spin_unlock_irqrestore(&q->lock, flags); } EXPORT_SYMBOL_GPL(__wake_up_sync_key); @@ -8000,9 +7848,7 @@ static int sd_degenerate(struct sched_domain *sd) } /* Following flags don't use groups */ - if (sd->flags & (SD_WAKE_IDLE | - SD_WAKE_AFFINE | - SD_WAKE_BALANCE)) + if (sd->flags & (SD_WAKE_AFFINE)) return 0; return 1; @@ -8019,10 +7865,6 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent) if (!cpumask_equal(sched_domain_span(sd), sched_domain_span(parent))) return 0; - /* Does parent contain flags not in child? */ - /* WAKE_BALANCE is a subset of WAKE_AFFINE */ - if (cflags & SD_WAKE_AFFINE) - pflags &= ~SD_WAKE_BALANCE; /* Flags needing groups don't count if only 1 group in parent */ if (parent->groups == parent->groups->next) { pflags &= ~(SD_LOAD_BALANCE | @@ -8708,10 +8550,10 @@ static void set_domain_attribute(struct sched_domain *sd, request = attr->relax_domain_level; if (request < sd->level) { /* turn off idle balance on this domain */ - sd->flags &= ~(SD_WAKE_IDLE|SD_BALANCE_NEWIDLE); + sd->flags &= ~(SD_BALANCE_WAKE|SD_BALANCE_NEWIDLE); } else { /* turn on idle balance on this domain */ - sd->flags |= (SD_WAKE_IDLE_FAR|SD_BALANCE_NEWIDLE); + sd->flags |= (SD_BALANCE_WAKE|SD_BALANCE_NEWIDLE); } } diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 5ddbd0891267..efb84409bc43 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -395,6 +395,7 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) PN(se.sum_exec_runtime); PN(se.avg_overlap); PN(se.avg_wakeup); + PN(se.avg_running); nr_switches = p->nvcsw + p->nivcsw; diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index aa7f84121016..10d218ab69f2 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -711,7 +711,7 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) if (!initial) { /* sleeps upto a single latency don't count. */ - if (sched_feat(NEW_FAIR_SLEEPERS)) { + if (sched_feat(FAIR_SLEEPERS)) { unsigned long thresh = sysctl_sched_latency; /* @@ -725,6 +725,13 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) task_of(se)->policy != SCHED_IDLE)) thresh = calc_delta_fair(thresh, se); + /* + * Halve their sleep time's effect, to allow + * for a gentler effect of sleepers: + */ + if (sched_feat(GENTLE_FAIR_SLEEPERS)) + thresh >>= 1; + vruntime -= thresh; } } @@ -757,10 +764,10 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int wakeup) static void __clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se) { - if (cfs_rq->last == se) + if (!se || cfs_rq->last == se) cfs_rq->last = NULL; - if (cfs_rq->next == se) + if (!se || cfs_rq->next == se) cfs_rq->next = NULL; } @@ -1062,83 +1069,6 @@ static void yield_task_fair(struct rq *rq) se->vruntime = rightmost->vruntime + 1; } -/* - * wake_idle() will wake a task on an idle cpu if task->cpu is - * not idle and an idle cpu is available. The span of cpus to - * search starts with cpus closest then further out as needed, - * so we always favor a closer, idle cpu. - * Domains may include CPUs that are not usable for migration, - * hence we need to mask them out (rq->rd->online) - * - * Returns the CPU we should wake onto. - */ -#if defined(ARCH_HAS_SCHED_WAKE_IDLE) - -#define cpu_rd_active(cpu, rq) cpumask_test_cpu(cpu, rq->rd->online) - -static int wake_idle(int cpu, struct task_struct *p) -{ - struct sched_domain *sd; - int i; - unsigned int chosen_wakeup_cpu; - int this_cpu; - struct rq *task_rq = task_rq(p); - - /* - * At POWERSAVINGS_BALANCE_WAKEUP level, if both this_cpu and prev_cpu - * are idle and this is not a kernel thread and this task's affinity - * allows it to be moved to preferred cpu, then just move! - */ - - this_cpu = smp_processor_id(); - chosen_wakeup_cpu = - cpu_rq(this_cpu)->rd->sched_mc_preferred_wakeup_cpu; - - if (sched_mc_power_savings >= POWERSAVINGS_BALANCE_WAKEUP && - idle_cpu(cpu) && idle_cpu(this_cpu) && - p->mm && !(p->flags & PF_KTHREAD) && - cpu_isset(chosen_wakeup_cpu, p->cpus_allowed)) - return chosen_wakeup_cpu; - - /* - * If it is idle, then it is the best cpu to run this task. - * - * This cpu is also the best, if it has more than one task already. - * Siblings must be also busy(in most cases) as they didn't already - * pickup the extra load from this cpu and hence we need not check - * sibling runqueue info. This will avoid the checks and cache miss - * penalities associated with that. - */ - if (idle_cpu(cpu) || cpu_rq(cpu)->cfs.nr_running > 1) - return cpu; - - for_each_domain(cpu, sd) { - if ((sd->flags & SD_WAKE_IDLE) - || ((sd->flags & SD_WAKE_IDLE_FAR) - && !task_hot(p, task_rq->clock, sd))) { - for_each_cpu_and(i, sched_domain_span(sd), - &p->cpus_allowed) { - if (cpu_rd_active(i, task_rq) && idle_cpu(i)) { - if (i != task_cpu(p)) { - schedstat_inc(p, - se.nr_wakeups_idle); - } - return i; - } - } - } else { - break; - } - } - return cpu; -} -#else /* !ARCH_HAS_SCHED_WAKE_IDLE*/ -static inline int wake_idle(int cpu, struct task_struct *p) -{ - return cpu; -} -#endif - #ifdef CONFIG_SMP #ifdef CONFIG_FAIR_GROUP_SCHED @@ -1225,25 +1155,34 @@ static inline unsigned long effective_load(struct task_group *tg, int cpu, #endif -static int -wake_affine(struct sched_domain *this_sd, struct rq *this_rq, - struct task_struct *p, int prev_cpu, int this_cpu, int sync, - int idx, unsigned long load, unsigned long this_load, - unsigned int imbalance) +static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync) { - struct task_struct *curr = this_rq->curr; - struct task_group *tg; - unsigned long tl = this_load; + struct task_struct *curr = current; + unsigned long this_load, load; + int idx, this_cpu, prev_cpu; unsigned long tl_per_task; + unsigned int imbalance; + struct task_group *tg; unsigned long weight; int balanced; - if (!(this_sd->flags & SD_WAKE_AFFINE) || !sched_feat(AFFINE_WAKEUPS)) - return 0; + idx = sd->wake_idx; + this_cpu = smp_processor_id(); + prev_cpu = task_cpu(p); + load = source_load(prev_cpu, idx); + this_load = target_load(this_cpu, idx); - if (sync && (curr->se.avg_overlap > sysctl_sched_migration_cost || - p->se.avg_overlap > sysctl_sched_migration_cost)) - sync = 0; + if (sync) { + if (sched_feat(SYNC_LESS) && + (curr->se.avg_overlap > sysctl_sched_migration_cost || + p->se.avg_overlap > sysctl_sched_migration_cost)) + sync = 0; + } else { + if (sched_feat(SYNC_MORE) && + (curr->se.avg_overlap < sysctl_sched_migration_cost && + p->se.avg_overlap < sysctl_sched_migration_cost)) + sync = 1; + } /* * If sync wakeup then subtract the (maximum possible) @@ -1254,24 +1193,26 @@ wake_affine(struct sched_domain *this_sd, struct rq *this_rq, tg = task_group(current); weight = current->se.load.weight; - tl += effective_load(tg, this_cpu, -weight, -weight); + this_load += effective_load(tg, this_cpu, -weight, -weight); load += effective_load(tg, prev_cpu, 0, -weight); } tg = task_group(p); weight = p->se.load.weight; + imbalance = 100 + (sd->imbalance_pct - 100) / 2; + /* * In low-load situations, where prev_cpu is idle and this_cpu is idle - * due to the sync cause above having dropped tl to 0, we'll always have - * an imbalance, but there's really nothing you can do about that, so - * that's good too. + * due to the sync cause above having dropped this_load to 0, we'll + * always have an imbalance, but there's really nothing you can do + * about that, so that's good too. * * Otherwise check if either cpus are near enough in load to allow this * task to be woken on this_cpu. */ - balanced = !tl || - 100*(tl + effective_load(tg, this_cpu, weight, weight)) <= + balanced = !this_load || + 100*(this_load + effective_load(tg, this_cpu, weight, weight)) <= imbalance*(load + effective_load(tg, prev_cpu, 0, weight)); /* @@ -1285,14 +1226,15 @@ wake_affine(struct sched_domain *this_sd, struct rq *this_rq, schedstat_inc(p, se.nr_wakeups_affine_attempts); tl_per_task = cpu_avg_load_per_task(this_cpu); - if (balanced || (tl <= load && tl + target_load(prev_cpu, idx) <= - tl_per_task)) { + if (balanced || + (this_load <= load && + this_load + target_load(prev_cpu, idx) <= tl_per_task)) { /* * This domain has SD_WAKE_AFFINE and * p is cache cold in this domain, and * there is no bad imbalance. */ - schedstat_inc(this_sd, ttwu_move_affine); + schedstat_inc(sd, ttwu_move_affine); schedstat_inc(p, se.nr_wakeups_affine); return 1; @@ -1300,65 +1242,215 @@ wake_affine(struct sched_domain *this_sd, struct rq *this_rq, return 0; } -static int select_task_rq_fair(struct task_struct *p, int sync) +/* + * find_idlest_group finds and returns the least busy CPU group within the + * domain. + */ +static struct sched_group * +find_idlest_group(struct sched_domain *sd, struct task_struct *p, + int this_cpu, int load_idx) { - struct sched_domain *sd, *this_sd = NULL; - int prev_cpu, this_cpu, new_cpu; - unsigned long load, this_load; - struct rq *this_rq; - unsigned int imbalance; - int idx; + struct sched_group *idlest = NULL, *this = NULL, *group = sd->groups; + unsigned long min_load = ULONG_MAX, this_load = 0; + int imbalance = 100 + (sd->imbalance_pct-100)/2; - prev_cpu = task_cpu(p); - this_cpu = smp_processor_id(); - this_rq = cpu_rq(this_cpu); - new_cpu = prev_cpu; + do { + unsigned long load, avg_load; + int local_group; + int i; - /* - * 'this_sd' is the first domain that both - * this_cpu and prev_cpu are present in: - */ - for_each_domain(this_cpu, sd) { - if (cpumask_test_cpu(prev_cpu, sched_domain_span(sd))) { - this_sd = sd; - break; + /* Skip over this group if it has no CPUs allowed */ + if (!cpumask_intersects(sched_group_cpus(group), + &p->cpus_allowed)) + continue; + + local_group = cpumask_test_cpu(this_cpu, + sched_group_cpus(group)); + + /* Tally up the load of all CPUs in the group */ + avg_load = 0; + + for_each_cpu(i, sched_group_cpus(group)) { + /* Bias balancing toward cpus of our domain */ + if (local_group) + load = source_load(i, load_idx); + else + load = target_load(i, load_idx); + + avg_load += load; + } + + /* Adjust by relative CPU power of the group */ + avg_load = (avg_load * SCHED_LOAD_SCALE) / group->cpu_power; + + if (local_group) { + this_load = avg_load; + this = group; + } else if (avg_load < min_load) { + min_load = avg_load; + idlest = group; + } + } while (group = group->next, group != sd->groups); + + if (!idlest || 100*this_load < imbalance*min_load) + return NULL; + return idlest; +} + +/* + * find_idlest_cpu - find the idlest cpu among the cpus in group. + */ +static int +find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) +{ + unsigned long load, min_load = ULONG_MAX; + int idlest = -1; + int i; + + /* Traverse only the allowed CPUs */ + for_each_cpu_and(i, sched_group_cpus(group), &p->cpus_allowed) { + load = weighted_cpuload(i); + + if (load < min_load || (load == min_load && i == this_cpu)) { + min_load = load; + idlest = i; } } - if (unlikely(!cpumask_test_cpu(this_cpu, &p->cpus_allowed))) - goto out; + return idlest; +} - /* - * Check for affine wakeup and passive balancing possibilities. - */ - if (!this_sd) - goto out; +/* + * sched_balance_self: balance the current task (running on cpu) in domains + * that have the 'flag' flag set. In practice, this is SD_BALANCE_FORK and + * SD_BALANCE_EXEC. + * + * Balance, ie. select the least loaded group. + * + * Returns the target CPU number, or the same CPU if no balancing is needed. + * + * preempt must be disabled. + */ +static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags) +{ + struct sched_domain *tmp, *affine_sd = NULL, *sd = NULL; + int cpu = smp_processor_id(); + int prev_cpu = task_cpu(p); + int new_cpu = cpu; + int want_affine = 0; + int want_sd = 1; + int sync = wake_flags & WF_SYNC; - idx = this_sd->wake_idx; + if (sd_flag & SD_BALANCE_WAKE) { + if (sched_feat(AFFINE_WAKEUPS)) + want_affine = 1; + new_cpu = prev_cpu; + } - imbalance = 100 + (this_sd->imbalance_pct - 100) / 2; + rcu_read_lock(); + for_each_domain(cpu, tmp) { + /* + * If power savings logic is enabled for a domain, see if we + * are not overloaded, if so, don't balance wider. + */ + if (tmp->flags & (SD_POWERSAVINGS_BALANCE|SD_PREFER_LOCAL)) { + unsigned long power = 0; + unsigned long nr_running = 0; + unsigned long capacity; + int i; - load = source_load(prev_cpu, idx); - this_load = target_load(this_cpu, idx); + for_each_cpu(i, sched_domain_span(tmp)) { + power += power_of(i); + nr_running += cpu_rq(i)->cfs.nr_running; + } - if (wake_affine(this_sd, this_rq, p, prev_cpu, this_cpu, sync, idx, - load, this_load, imbalance)) - return this_cpu; + capacity = DIV_ROUND_CLOSEST(power, SCHED_LOAD_SCALE); - /* - * Start passive balancing when half the imbalance_pct - * limit is reached. - */ - if (this_sd->flags & SD_WAKE_BALANCE) { - if (imbalance*this_load <= 100*load) { - schedstat_inc(this_sd, ttwu_move_balance); - schedstat_inc(p, se.nr_wakeups_passive); - return this_cpu; + if (tmp->flags & SD_POWERSAVINGS_BALANCE) + nr_running /= 2; + + if (nr_running < capacity) + want_sd = 0; } + + if (want_affine && (tmp->flags & SD_WAKE_AFFINE) && + cpumask_test_cpu(prev_cpu, sched_domain_span(tmp))) { + + affine_sd = tmp; + want_affine = 0; + } + + if (!want_sd && !want_affine) + break; + + if (!(tmp->flags & sd_flag)) + continue; + + if (want_sd) + sd = tmp; + } + + if (sched_feat(LB_SHARES_UPDATE)) { + /* + * Pick the largest domain to update shares over + */ + tmp = sd; + if (affine_sd && (!tmp || + cpumask_weight(sched_domain_span(affine_sd)) > + cpumask_weight(sched_domain_span(sd)))) + tmp = affine_sd; + + if (tmp) + update_shares(tmp); + } + + if (affine_sd && wake_affine(affine_sd, p, sync)) { + new_cpu = cpu; + goto out; + } + + while (sd) { + int load_idx = sd->forkexec_idx; + struct sched_group *group; + int weight; + + if (!(sd->flags & sd_flag)) { + sd = sd->child; + continue; + } + + if (sd_flag & SD_BALANCE_WAKE) + load_idx = sd->wake_idx; + + group = find_idlest_group(sd, p, cpu, load_idx); + if (!group) { + sd = sd->child; + continue; + } + + new_cpu = find_idlest_cpu(group, p, cpu); + if (new_cpu == -1 || new_cpu == cpu) { + /* Now try balancing at a lower domain level of cpu */ + sd = sd->child; + continue; + } + + /* Now try balancing at a lower domain level of new_cpu */ + cpu = new_cpu; + weight = cpumask_weight(sched_domain_span(sd)); + sd = NULL; + for_each_domain(cpu, tmp) { + if (weight <= cpumask_weight(sched_domain_span(tmp))) + break; + if (tmp->flags & sd_flag) + sd = tmp; + } + /* while loop will break here if sd == NULL */ } out: - return wake_idle(new_cpu, p); + rcu_read_unlock(); + return new_cpu; } #endif /* CONFIG_SMP */ @@ -1471,11 +1563,12 @@ static void set_next_buddy(struct sched_entity *se) /* * Preempt the current task with a newly woken task if needed: */ -static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int sync) +static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_flags) { struct task_struct *curr = rq->curr; struct sched_entity *se = &curr->se, *pse = &p->se; struct cfs_rq *cfs_rq = task_cfs_rq(curr); + int sync = wake_flags & WF_SYNC; update_curr(cfs_rq); @@ -1501,7 +1594,8 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int sync) */ if (sched_feat(LAST_BUDDY) && likely(se->on_rq && curr != rq->idle)) set_last_buddy(se); - set_next_buddy(pse); + if (sched_feat(NEXT_BUDDY) && !(wake_flags & WF_FORK)) + set_next_buddy(pse); /* * We can come here with TIF_NEED_RESCHED already set from new task @@ -1523,16 +1617,25 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int sync) return; } - if (!sched_feat(WAKEUP_PREEMPT)) - return; - - if (sched_feat(WAKEUP_OVERLAP) && (sync || - (se->avg_overlap < sysctl_sched_migration_cost && - pse->avg_overlap < sysctl_sched_migration_cost))) { + if ((sched_feat(WAKEUP_SYNC) && sync) || + (sched_feat(WAKEUP_OVERLAP) && + (se->avg_overlap < sysctl_sched_migration_cost && + pse->avg_overlap < sysctl_sched_migration_cost))) { resched_task(curr); return; } + if (sched_feat(WAKEUP_RUNNING)) { + if (pse->avg_running < se->avg_running) { + set_next_buddy(pse); + resched_task(curr); + return; + } + } + + if (!sched_feat(WAKEUP_PREEMPT)) + return; + find_matching_se(&se, &pse); BUG_ON(!pse); @@ -1555,8 +1658,13 @@ static struct task_struct *pick_next_task_fair(struct rq *rq) /* * If se was a buddy, clear it so that it will have to earn * the favour again. + * + * If se was not a buddy, clear the buddies because neither + * was elegible to run, let them earn it again. + * + * IOW. unconditionally clear buddies. */ - __clear_buddies(cfs_rq, se); + __clear_buddies(cfs_rq, NULL); set_next_entity(cfs_rq, se); cfs_rq = group_cfs_rq(se); } while (cfs_rq); diff --git a/kernel/sched_features.h b/kernel/sched_features.h index e2dc63a5815d..0d94083582c7 100644 --- a/kernel/sched_features.h +++ b/kernel/sched_features.h @@ -1,17 +1,123 @@ -SCHED_FEAT(NEW_FAIR_SLEEPERS, 0) +/* + * Disregards a certain amount of sleep time (sched_latency_ns) and + * considers the task to be running during that period. This gives it + * a service deficit on wakeup, allowing it to run sooner. + */ +SCHED_FEAT(FAIR_SLEEPERS, 1) + +/* + * Only give sleepers 50% of their service deficit. This allows + * them to run sooner, but does not allow tons of sleepers to + * rip the spread apart. + */ +SCHED_FEAT(GENTLE_FAIR_SLEEPERS, 1) + +/* + * By not normalizing the sleep time, heavy tasks get an effective + * longer period, and lighter task an effective shorter period they + * are considered running. + */ SCHED_FEAT(NORMALIZED_SLEEPER, 0) -SCHED_FEAT(ADAPTIVE_GRAN, 1) -SCHED_FEAT(WAKEUP_PREEMPT, 1) + +/* + * Place new tasks ahead so that they do not starve already running + * tasks + */ SCHED_FEAT(START_DEBIT, 1) -SCHED_FEAT(AFFINE_WAKEUPS, 1) -SCHED_FEAT(CACHE_HOT_BUDDY, 1) + +/* + * Should wakeups try to preempt running tasks. + */ +SCHED_FEAT(WAKEUP_PREEMPT, 1) + +/* + * Compute wakeup_gran based on task behaviour, clipped to + * [0, sched_wakeup_gran_ns] + */ +SCHED_FEAT(ADAPTIVE_GRAN, 1) + +/* + * When converting the wakeup granularity to virtual time, do it such + * that heavier tasks preempting a lighter task have an edge. + */ +SCHED_FEAT(ASYM_GRAN, 1) + +/* + * Always wakeup-preempt SYNC wakeups, see SYNC_WAKEUPS. + */ +SCHED_FEAT(WAKEUP_SYNC, 0) + +/* + * Wakeup preempt based on task behaviour. Tasks that do not overlap + * don't get preempted. + */ +SCHED_FEAT(WAKEUP_OVERLAP, 0) + +/* + * Wakeup preemption towards tasks that run short + */ +SCHED_FEAT(WAKEUP_RUNNING, 0) + +/* + * Use the SYNC wakeup hint, pipes and the likes use this to indicate + * the remote end is likely to consume the data we just wrote, and + * therefore has cache benefit from being placed on the same cpu, see + * also AFFINE_WAKEUPS. + */ SCHED_FEAT(SYNC_WAKEUPS, 1) + +/* + * Based on load and program behaviour, see if it makes sense to place + * a newly woken task on the same cpu as the task that woke it -- + * improve cache locality. Typically used with SYNC wakeups as + * generated by pipes and the like, see also SYNC_WAKEUPS. + */ +SCHED_FEAT(AFFINE_WAKEUPS, 1) + +/* + * Weaken SYNC hint based on overlap + */ +SCHED_FEAT(SYNC_LESS, 1) + +/* + * Add SYNC hint based on overlap + */ +SCHED_FEAT(SYNC_MORE, 0) + +/* + * Prefer to schedule the task we woke last (assuming it failed + * wakeup-preemption), since its likely going to consume data we + * touched, increases cache locality. + */ +SCHED_FEAT(NEXT_BUDDY, 0) + +/* + * Prefer to schedule the task that ran last (when we did + * wake-preempt) as that likely will touch the same data, increases + * cache locality. + */ +SCHED_FEAT(LAST_BUDDY, 1) + +/* + * Consider buddies to be cache hot, decreases the likelyness of a + * cache buddy being migrated away, increases cache locality. + */ +SCHED_FEAT(CACHE_HOT_BUDDY, 1) + +/* + * Use arch dependent cpu power functions + */ +SCHED_FEAT(ARCH_POWER, 0) + SCHED_FEAT(HRTICK, 0) SCHED_FEAT(DOUBLE_TICK, 0) -SCHED_FEAT(ASYM_GRAN, 1) SCHED_FEAT(LB_BIAS, 1) -SCHED_FEAT(LB_WAKEUP_UPDATE, 1) +SCHED_FEAT(LB_SHARES_UPDATE, 1) SCHED_FEAT(ASYM_EFF_LOAD, 1) -SCHED_FEAT(WAKEUP_OVERLAP, 0) -SCHED_FEAT(LAST_BUDDY, 1) + +/* + * Spin-wait on mutex acquisition when the mutex owner is running on + * another cpu -- assumes that when the owner is running, it will soon + * release the lock. Decreases scheduling overhead. + */ SCHED_FEAT(OWNER_SPIN, 1) diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c index 499672c10cbd..a8b448af004b 100644 --- a/kernel/sched_idletask.c +++ b/kernel/sched_idletask.c @@ -6,7 +6,7 @@ */ #ifdef CONFIG_SMP -static int select_task_rq_idle(struct task_struct *p, int sync) +static int select_task_rq_idle(struct task_struct *p, int sd_flag, int flags) { return task_cpu(p); /* IDLE tasks as never migrated */ } @@ -14,7 +14,7 @@ static int select_task_rq_idle(struct task_struct *p, int sync) /* * Idle tasks are unconditionally rescheduled: */ -static void check_preempt_curr_idle(struct rq *rq, struct task_struct *p, int sync) +static void check_preempt_curr_idle(struct rq *rq, struct task_struct *p, int flags) { resched_task(rq->idle); } diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 2eb4bd6a526c..13de7126a6ab 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -938,10 +938,13 @@ static void yield_task_rt(struct rq *rq) #ifdef CONFIG_SMP static int find_lowest_rq(struct task_struct *task); -static int select_task_rq_rt(struct task_struct *p, int sync) +static int select_task_rq_rt(struct task_struct *p, int sd_flag, int flags) { struct rq *rq = task_rq(p); + if (sd_flag != SD_BALANCE_WAKE) + return smp_processor_id(); + /* * If the current task is an RT task, then * try to see if we can wake this RT task up on another @@ -999,7 +1002,7 @@ static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p) /* * Preempt the current task with a newly woken task if needed: */ -static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p, int sync) +static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p, int flags) { if (p->prio < rq->curr->prio) { resched_task(rq->curr);