From 6161352142d5fed4cd753b32e5ccde66e705b14e Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Thu, 17 Sep 2009 16:11:28 +0200 Subject: tracing, perf: Convert the power tracer into an event tracer This patch converts the existing power tracer into an event tracer, so that power events (C states and frequency changes) can be tracked via "perf". This also removes the perl script that was used to demo the tracer; its functionality is being replaced entirely with timechart. Signed-off-by: Arjan van de Ven Acked-by: Peter Zijlstra Cc: Paul Mackerras Cc: Frederic Weisbecker LKML-Reference: <20090912130542.6d314860@infradead.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/process.c | 28 +++++++++------------------- 1 file changed, 9 insertions(+), 19 deletions(-) (limited to 'arch/x86/kernel/process.c') diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 071166a4ba83..7b60e3906889 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -9,7 +9,7 @@ #include #include #include -#include +#include #include #include #include @@ -25,9 +25,6 @@ EXPORT_SYMBOL(idle_nomwait); struct kmem_cache *task_xstate_cachep; -DEFINE_TRACE(power_start); -DEFINE_TRACE(power_end); - int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src) { *dst = *src; @@ -299,9 +296,7 @@ static inline int hlt_use_halt(void) void default_idle(void) { if (hlt_use_halt()) { - struct power_trace it; - - trace_power_start(&it, POWER_CSTATE, 1); + trace_power_start(POWER_CSTATE, 1); current_thread_info()->status &= ~TS_POLLING; /* * TS_POLLING-cleared state must be visible before we @@ -314,7 +309,7 @@ void default_idle(void) else local_irq_enable(); current_thread_info()->status |= TS_POLLING; - trace_power_end(&it); + trace_power_end(0); } else { local_irq_enable(); /* loop is done by the caller */ @@ -372,9 +367,7 @@ EXPORT_SYMBOL_GPL(cpu_idle_wait); */ void mwait_idle_with_hints(unsigned long ax, unsigned long cx) { - struct power_trace it; - - trace_power_start(&it, POWER_CSTATE, (ax>>4)+1); + trace_power_start(POWER_CSTATE, (ax>>4)+1); if (!need_resched()) { if (cpu_has(¤t_cpu_data, X86_FEATURE_CLFLUSH_MONITOR)) clflush((void *)¤t_thread_info()->flags); @@ -384,15 +377,14 @@ void mwait_idle_with_hints(unsigned long ax, unsigned long cx) if (!need_resched()) __mwait(ax, cx); } - trace_power_end(&it); + trace_power_end(0); } /* Default MONITOR/MWAIT with no hints, used for default C1 state */ static void mwait_idle(void) { - struct power_trace it; if (!need_resched()) { - trace_power_start(&it, POWER_CSTATE, 1); + trace_power_start(POWER_CSTATE, 1); if (cpu_has(¤t_cpu_data, X86_FEATURE_CLFLUSH_MONITOR)) clflush((void *)¤t_thread_info()->flags); @@ -402,7 +394,7 @@ static void mwait_idle(void) __sti_mwait(0, 0); else local_irq_enable(); - trace_power_end(&it); + trace_power_end(0); } else local_irq_enable(); } @@ -414,13 +406,11 @@ static void mwait_idle(void) */ static void poll_idle(void) { - struct power_trace it; - - trace_power_start(&it, POWER_CSTATE, 0); + trace_power_start(POWER_CSTATE, 0); local_irq_enable(); while (!need_resched()) cpu_relax(); - trace_power_end(&it); + trace_power_end(0); } /* -- cgit v1.2.3 From 288f023e708efd89d77ce9acf977a33a623ae83d Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Sat, 19 Sep 2009 13:35:33 +0200 Subject: tracing, x86, cpuidle: Move the end point of a C state in the power tracer The "end of a C state" trace point currently happens before the code runs that corrects the TSC for having stopped during idle. The result of this is that the timestamp of the end-of-C-state event is garbage on cpus where the TSC stops during idle. This patch moves the end point of the C state to after the timekeeping engine of the kernel has been corrected. Signed-off-by: Arjan van de Ven Cc: Len Brown Cc: fweisbec@gmail.com Cc: peterz@infradead.org Cc: Paul Mackerras LKML-Reference: <20090919133533.139c2a46@infradead.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/process.c | 3 --- drivers/cpuidle/cpuidle.c | 2 ++ 2 files changed, 2 insertions(+), 3 deletions(-) (limited to 'arch/x86/kernel/process.c') diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 7b60e3906889..847ab4160315 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -309,7 +309,6 @@ void default_idle(void) else local_irq_enable(); current_thread_info()->status |= TS_POLLING; - trace_power_end(0); } else { local_irq_enable(); /* loop is done by the caller */ @@ -377,7 +376,6 @@ void mwait_idle_with_hints(unsigned long ax, unsigned long cx) if (!need_resched()) __mwait(ax, cx); } - trace_power_end(0); } /* Default MONITOR/MWAIT with no hints, used for default C1 state */ @@ -394,7 +392,6 @@ static void mwait_idle(void) __sti_mwait(0, 0); else local_irq_enable(); - trace_power_end(0); } else local_irq_enable(); } diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index 8504a2108557..ad41f19b8e3f 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -17,6 +17,7 @@ #include #include #include +#include #include "cpuidle.h" @@ -91,6 +92,7 @@ static void cpuidle_idle_call(void) /* give the governor an opportunity to reflect on the outcome */ if (cpuidle_curr_governor->reflect) cpuidle_curr_governor->reflect(dev); + trace_power_end(0); } /** -- cgit v1.2.3 From 79f5599772ac2f138d7a75b8f3f06a93f09c75f7 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Mon, 15 Jun 2009 14:58:26 +0800 Subject: cpumask: use zalloc_cpumask_var() where possible Remove open-coded zalloc_cpumask_var() and zalloc_cpumask_var_node(). Signed-off-by: Li Zefan Signed-off-by: Rusty Russell --- arch/x86/kernel/apic/io_apic.c | 7 ++----- arch/x86/kernel/process.c | 6 ++---- arch/x86/kernel/smpboot.c | 9 +++------ drivers/acpi/processor_perflib.c | 3 +-- drivers/acpi/processor_throttling.c | 3 +-- drivers/net/sfc/efx.c | 3 +-- drivers/oprofile/buffer_sync.c | 3 +-- kernel/trace/trace.c | 7 ++----- virt/kvm/kvm_main.c | 3 +-- 9 files changed, 14 insertions(+), 30 deletions(-) (limited to 'arch/x86/kernel/process.c') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 64970b9885f2..dc69f28489f5 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -227,17 +227,14 @@ static struct irq_cfg *get_one_free_irq_cfg(int node) cfg = kzalloc_node(sizeof(*cfg), GFP_ATOMIC, node); if (cfg) { - if (!alloc_cpumask_var_node(&cfg->domain, GFP_ATOMIC, node)) { + if (!zalloc_cpumask_var_node(&cfg->domain, GFP_ATOMIC, node)) { kfree(cfg); cfg = NULL; - } else if (!alloc_cpumask_var_node(&cfg->old_domain, + } else if (!zalloc_cpumask_var_node(&cfg->old_domain, GFP_ATOMIC, node)) { free_cpumask_var(cfg->domain); kfree(cfg); cfg = NULL; - } else { - cpumask_clear(cfg->domain); - cpumask_clear(cfg->old_domain); } } diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 847ab4160315..5284cd2b5776 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -555,10 +555,8 @@ void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) void __init init_c1e_mask(void) { /* If we're using c1e_idle, we need to allocate c1e_mask. */ - if (pm_idle == c1e_idle) { - alloc_cpumask_var(&c1e_mask, GFP_KERNEL); - cpumask_clear(c1e_mask); - } + if (pm_idle == c1e_idle) + zalloc_cpumask_var(&c1e_mask, GFP_KERNEL); } static int __init idle_setup(char *str) diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 09c5e077dff7..565ebc65920e 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1059,12 +1059,9 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus) #endif current_thread_info()->cpu = 0; /* needed? */ for_each_possible_cpu(i) { - alloc_cpumask_var(&per_cpu(cpu_sibling_map, i), GFP_KERNEL); - alloc_cpumask_var(&per_cpu(cpu_core_map, i), GFP_KERNEL); - alloc_cpumask_var(&cpu_data(i).llc_shared_map, GFP_KERNEL); - cpumask_clear(per_cpu(cpu_core_map, i)); - cpumask_clear(per_cpu(cpu_sibling_map, i)); - cpumask_clear(cpu_data(i).llc_shared_map); + zalloc_cpumask_var(&per_cpu(cpu_sibling_map, i), GFP_KERNEL); + zalloc_cpumask_var(&per_cpu(cpu_core_map, i), GFP_KERNEL); + zalloc_cpumask_var(&cpu_data(i).llc_shared_map, GFP_KERNEL); } set_cpu_sibling_map(0); diff --git a/drivers/acpi/processor_perflib.c b/drivers/acpi/processor_perflib.c index 11088cf10319..8ba0ed0b9ddb 100644 --- a/drivers/acpi/processor_perflib.c +++ b/drivers/acpi/processor_perflib.c @@ -511,7 +511,7 @@ int acpi_processor_preregister_performance( struct acpi_processor *match_pr; struct acpi_psd_package *match_pdomain; - if (!alloc_cpumask_var(&covered_cpus, GFP_KERNEL)) + if (!zalloc_cpumask_var(&covered_cpus, GFP_KERNEL)) return -ENOMEM; mutex_lock(&performance_mutex); @@ -558,7 +558,6 @@ int acpi_processor_preregister_performance( * Now that we have _PSD data from all CPUs, lets setup P-state * domain info. */ - cpumask_clear(covered_cpus); for_each_possible_cpu(i) { pr = per_cpu(processors, i); if (!pr) diff --git a/drivers/acpi/processor_throttling.c b/drivers/acpi/processor_throttling.c index ce7cf3bc5101..4c6c14c1e307 100644 --- a/drivers/acpi/processor_throttling.c +++ b/drivers/acpi/processor_throttling.c @@ -77,7 +77,7 @@ static int acpi_processor_update_tsd_coord(void) struct acpi_tsd_package *pdomain, *match_pdomain; struct acpi_processor_throttling *pthrottling, *match_pthrottling; - if (!alloc_cpumask_var(&covered_cpus, GFP_KERNEL)) + if (!zalloc_cpumask_var(&covered_cpus, GFP_KERNEL)) return -ENOMEM; /* @@ -105,7 +105,6 @@ static int acpi_processor_update_tsd_coord(void) if (retval) goto err_ret; - cpumask_clear(covered_cpus); for_each_possible_cpu(i) { pr = per_cpu(processors, i); if (!pr) diff --git a/drivers/net/sfc/efx.c b/drivers/net/sfc/efx.c index 07a7e4b8f8fc..cc4b2f99989d 100644 --- a/drivers/net/sfc/efx.c +++ b/drivers/net/sfc/efx.c @@ -884,13 +884,12 @@ static int efx_wanted_rx_queues(void) int count; int cpu; - if (unlikely(!alloc_cpumask_var(&core_mask, GFP_KERNEL))) { + if (unlikely(!zalloc_cpumask_var(&core_mask, GFP_KERNEL))) { printk(KERN_WARNING "sfc: RSS disabled due to allocation failure\n"); return 1; } - cpumask_clear(core_mask); count = 0; for_each_online_cpu(cpu) { if (!cpumask_test_cpu(cpu, core_mask)) { diff --git a/drivers/oprofile/buffer_sync.c b/drivers/oprofile/buffer_sync.c index 8574622e36a5..c9e2ae90f195 100644 --- a/drivers/oprofile/buffer_sync.c +++ b/drivers/oprofile/buffer_sync.c @@ -154,9 +154,8 @@ int sync_start(void) { int err; - if (!alloc_cpumask_var(&marked_cpus, GFP_KERNEL)) + if (!zalloc_cpumask_var(&marked_cpus, GFP_KERNEL)) return -ENOMEM; - cpumask_clear(marked_cpus); start_cpu_work(); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 6c0f6a8a22eb..411af37f4be4 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1984,11 +1984,9 @@ __tracing_open(struct inode *inode, struct file *file) if (current_trace) *iter->trace = *current_trace; - if (!alloc_cpumask_var(&iter->started, GFP_KERNEL)) + if (!zalloc_cpumask_var(&iter->started, GFP_KERNEL)) goto fail; - cpumask_clear(iter->started); - if (current_trace && current_trace->print_max) iter->tr = &max_tr; else @@ -4389,7 +4387,7 @@ __init static int tracer_alloc_buffers(void) if (!alloc_cpumask_var(&tracing_cpumask, GFP_KERNEL)) goto out_free_buffer_mask; - if (!alloc_cpumask_var(&tracing_reader_cpumask, GFP_KERNEL)) + if (!zalloc_cpumask_var(&tracing_reader_cpumask, GFP_KERNEL)) goto out_free_tracing_cpumask; /* To save memory, keep the ring buffer size to its minimum */ @@ -4400,7 +4398,6 @@ __init static int tracer_alloc_buffers(void) cpumask_copy(tracing_buffer_mask, cpu_possible_mask); cpumask_copy(tracing_cpumask, cpu_all_mask); - cpumask_clear(tracing_reader_cpumask); /* TODO: make the number of buffers hot pluggable with CPUS */ global_trace.buffer = ring_buffer_alloc(ring_buf_size, diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 897bff3b7df9..034a798b0431 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -738,8 +738,7 @@ static bool make_all_cpus_request(struct kvm *kvm, unsigned int req) bool called = true; struct kvm_vcpu *vcpu; - if (alloc_cpumask_var(&cpus, GFP_ATOMIC)) - cpumask_clear(cpus); + zalloc_cpumask_var(&cpus, GFP_ATOMIC); spin_lock(&kvm->requests_lock); me = smp_processor_id(); -- cgit v1.2.3