#include "op_counter.h"
#include "op_x86_model.h"
-DEFINE_PER_CPU(int, switch_index);
-
static struct op_x86_model_spec const *model;
static DEFINE_PER_CPU(struct op_msrs, cpu_msrs);
static DEFINE_PER_CPU(unsigned long, saved_lvtpc);
-static int nmi_start(void);
-static void nmi_stop(void);
-static void nmi_cpu_start(void *dummy);
-static void nmi_cpu_stop(void *dummy);
-static void nmi_cpu_save_mpx_registers(struct op_msrs *msrs);
-static void nmi_cpu_restore_mpx_registers(struct op_msrs *msrs);
-
/* 0 == registered but off, 1 == registered and on */
static int nmi_enabled = 0;
-#ifdef CONFIG_SMP
-static int oprofile_cpu_notifier(struct notifier_block *b, unsigned long action,
- void *data)
-{
- int cpu = (unsigned long)data;
- switch (action) {
- case CPU_DOWN_FAILED:
- case CPU_ONLINE:
- smp_call_function_single(cpu, nmi_cpu_start, NULL, 0);
- break;
- case CPU_DOWN_PREPARE:
- smp_call_function_single(cpu, nmi_cpu_stop, NULL, 1);
- break;
- }
- return NOTIFY_DONE;
-}
-
-static struct notifier_block oprofile_cpu_nb = {
- .notifier_call = oprofile_cpu_notifier
-};
-#endif
-
-#ifdef CONFIG_PM
-
-static int nmi_suspend(struct sys_device *dev, pm_message_t state)
-{
- /* Only one CPU left, just stop that one */
- if (nmi_enabled == 1)
- nmi_cpu_stop(NULL);
- return 0;
-}
-
-static int nmi_resume(struct sys_device *dev)
-{
- if (nmi_enabled == 1)
- nmi_cpu_start(NULL);
- return 0;
-}
-
-static struct sysdev_class oprofile_sysclass = {
- .name = "oprofile",
- .resume = nmi_resume,
- .suspend = nmi_suspend,
-};
-
-static struct sys_device device_oprofile = {
- .id = 0,
- .cls = &oprofile_sysclass,
-};
-
-static int __init init_sysfs(void)
-{
- int error;
+/* common functions */
- error = sysdev_class_register(&oprofile_sysclass);
- if (!error)
- error = sysdev_register(&device_oprofile);
- return error;
-}
-
-static void exit_sysfs(void)
+u64 op_x86_get_ctrl(struct op_x86_model_spec const *model,
+ struct op_counter_config *counter_config)
{
- sysdev_unregister(&device_oprofile);
- sysdev_class_unregister(&oprofile_sysclass);
+ u64 val = 0;
+ u16 event = (u16)counter_config->event;
+
+ val |= ARCH_PERFMON_EVENTSEL_INT;
+ val |= counter_config->user ? ARCH_PERFMON_EVENTSEL_USR : 0;
+ val |= counter_config->kernel ? ARCH_PERFMON_EVENTSEL_OS : 0;
+ val |= (counter_config->unit_mask & 0xFF) << 8;
+ event &= model->event_mask ? model->event_mask : 0xFF;
+ val |= event & 0xFF;
+ val |= (event & 0x0F00) << 24;
+
+ return val;
}
-#else
-#define init_sysfs() do { } while (0)
-#define exit_sysfs() do { } while (0)
-#endif /* CONFIG_PM */
-
-static void nmi_cpu_switch(void *dummy)
-{
- int cpu = smp_processor_id();
- int si = per_cpu(switch_index, cpu);
- struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
-
- nmi_cpu_stop(NULL);
- nmi_cpu_save_mpx_registers(msrs);
-
- /* move to next set */
- si += model->num_hardware_counters;
- if ((si > model->num_counters) || (counter_config[si].count == 0))
- per_cpu(switch_index, smp_processor_id()) = 0;
- else
- per_cpu(switch_index, smp_processor_id()) = si;
-
- nmi_cpu_restore_mpx_registers(msrs);
- model->setup_ctrs(msrs);
- nmi_cpu_start(NULL);
-}
-
-/*
- * Quick check to see if multiplexing is necessary.
- * The check should be sufficient since counters are used
- * in ordre.
- */
-static int nmi_multiplex_on(void)
-{
- return counter_config[model->num_hardware_counters].count ? 0 : -EINVAL;
-}
-
-static int nmi_switch_event(void)
-{
- if (nmi_multiplex_on() < 0)
- return -EINVAL;
-
- on_each_cpu(nmi_cpu_switch, NULL, 1);
-
- return 0;
-}
static int profile_exceptions_notify(struct notifier_block *self,
unsigned long val, void *data)
switch (val) {
case DIE_NMI:
- if (model->check_ctrs(args->regs, &per_cpu(cpu_msrs, cpu)))
- ret = NOTIFY_STOP;
+ case DIE_NMI_IPI:
+ model->check_ctrs(args->regs, &per_cpu(cpu_msrs, cpu));
+ ret = NOTIFY_STOP;
break;
default:
break;
static void nmi_cpu_save_registers(struct op_msrs *msrs)
{
- unsigned int const nr_ctrs = model->num_counters;
- unsigned int const nr_ctrls = model->num_controls;
struct op_msr *counters = msrs->counters;
struct op_msr *controls = msrs->controls;
unsigned int i;
- for (i = 0; i < nr_ctrs; ++i) {
- if (counters[i].addr) {
- rdmsr(counters[i].addr,
- counters[i].saved.low,
- counters[i].saved.high);
- }
+ for (i = 0; i < model->num_counters; ++i) {
+ if (counters[i].addr)
+ rdmsrl(counters[i].addr, counters[i].saved);
}
- for (i = 0; i < nr_ctrls; ++i) {
- if (controls[i].addr) {
- rdmsr(controls[i].addr,
- controls[i].saved.low,
- controls[i].saved.high);
- }
+ for (i = 0; i < model->num_controls; ++i) {
+ if (controls[i].addr)
+ rdmsrl(controls[i].addr, controls[i].saved);
}
}
static int allocate_msrs(void)
{
- int i, success = 1;
+ int success = 1;
size_t controls_size = sizeof(struct op_msr) * model->num_controls;
size_t counters_size = sizeof(struct op_msr) * model->num_counters;
+ int i;
for_each_possible_cpu(i) {
per_cpu(cpu_msrs, i).counters = kmalloc(counters_size,
GFP_KERNEL);
success = 0;
break;
}
- per_cpu(cpu_msrs, i).controls =
- kmalloc(controls_size, GFP_KERNEL);
+ per_cpu(cpu_msrs, i).controls = kmalloc(controls_size,
+ GFP_KERNEL);
if (!per_cpu(cpu_msrs, i).controls) {
success = 0;
break;
int cpu = smp_processor_id();
struct op_msrs *msrs = &per_cpu(cpu_msrs, cpu);
spin_lock(&oprofilefs_lock);
- model->setup_ctrs(msrs);
+ model->setup_ctrs(model, msrs);
spin_unlock(&oprofilefs_lock);
per_cpu(saved_lvtpc, cpu) = apic_read(APIC_LVTPC);
apic_write(APIC_LVTPC, APIC_DM_NMI);
static struct notifier_block profile_exceptions_nb = {
.notifier_call = profile_exceptions_notify,
.next = NULL,
- .priority = 0
+ .priority = 2
};
static int nmi_setup(void)
return err;
}
- /*
- * We need to serialize save and setup for HT because the subset
+ /* We need to serialize save and setup for HT because the subset
* of msrs are distinct for save and setup operations
*/
per_cpu(cpu_msrs, 0).controls,
sizeof(struct op_msr) * model->num_controls);
}
+
}
on_each_cpu(nmi_save_registers, NULL, 1);
on_each_cpu(nmi_cpu_setup, NULL, 1);
return 0;
}
-static void nmi_cpu_save_mpx_registers(struct op_msrs *msrs)
-{
- unsigned int si = __get_cpu_var(switch_index);
- unsigned int const nr_ctrs = model->num_hardware_counters;
- struct op_msr *counters = &msrs->counters[si];
- unsigned int i;
-
- for (i = 0; i < nr_ctrs; ++i) {
- int offset = i + si;
- if (counters[offset].addr) {
- rdmsr(counters[offset].addr,
- counters[offset].multiplex.low,
- counters[offset].multiplex.high);
- }
- }
-}
-
-static void nmi_cpu_restore_mpx_registers(struct op_msrs *msrs)
+static void nmi_restore_registers(struct op_msrs *msrs)
{
- unsigned int si = __get_cpu_var(switch_index);
- unsigned int const nr_ctrs = model->num_hardware_counters;
- struct op_msr *counters = &msrs->counters[si];
- unsigned int i;
-
- for (i = 0; i < nr_ctrs; ++i) {
- int offset = i + si;
- if (counters[offset].addr) {
- wrmsr(counters[offset].addr,
- counters[offset].multiplex.low,
- counters[offset].multiplex.high);
- }
- }
-}
-
-static void nmi_cpu_restore_registers(struct op_msrs *msrs)
-{
- unsigned int const nr_ctrs = model->num_counters;
- unsigned int const nr_ctrls = model->num_controls;
struct op_msr *counters = msrs->counters;
struct op_msr *controls = msrs->controls;
unsigned int i;
- for (i = 0; i < nr_ctrls; ++i) {
- if (controls[i].addr) {
- wrmsr(controls[i].addr,
- controls[i].saved.low,
- controls[i].saved.high);
- }
+ for (i = 0; i < model->num_controls; ++i) {
+ if (controls[i].addr)
+ wrmsrl(controls[i].addr, controls[i].saved);
}
- for (i = 0; i < nr_ctrs; ++i) {
- if (counters[i].addr) {
- wrmsr(counters[i].addr,
- counters[i].saved.low,
- counters[i].saved.high);
- }
+ for (i = 0; i < model->num_counters; ++i) {
+ if (counters[i].addr)
+ wrmsrl(counters[i].addr, counters[i].saved);
}
}
apic_write(APIC_LVTERR, v | APIC_LVT_MASKED);
apic_write(APIC_LVTPC, per_cpu(saved_lvtpc, cpu));
apic_write(APIC_LVTERR, v);
- nmi_cpu_restore_registers(msrs);
- __get_cpu_var(switch_index) = 0;
+ nmi_restore_registers(msrs);
}
static void nmi_shutdown(void)
oprofilefs_create_ulong(sb, dir, "unit_mask", &counter_config[i].unit_mask);
oprofilefs_create_ulong(sb, dir, "kernel", &counter_config[i].kernel);
oprofilefs_create_ulong(sb, dir, "user", &counter_config[i].user);
- counter_config[i].save_count_low = 0;
}
return 0;
}
-static int p4force;
-module_param(p4force, int, 0);
+#ifdef CONFIG_SMP
+static int oprofile_cpu_notifier(struct notifier_block *b, unsigned long action,
+ void *data)
+{
+ int cpu = (unsigned long)data;
+ switch (action) {
+ case CPU_DOWN_FAILED:
+ case CPU_ONLINE:
+ smp_call_function_single(cpu, nmi_cpu_start, NULL, 0);
+ break;
+ case CPU_DOWN_PREPARE:
+ smp_call_function_single(cpu, nmi_cpu_stop, NULL, 1);
+ break;
+ }
+ return NOTIFY_DONE;
+}
+
+static struct notifier_block oprofile_cpu_nb = {
+ .notifier_call = oprofile_cpu_notifier
+};
+#endif
+
+#ifdef CONFIG_PM
+
+static int nmi_suspend(struct sys_device *dev, pm_message_t state)
+{
+ /* Only one CPU left, just stop that one */
+ if (nmi_enabled == 1)
+ nmi_cpu_stop(NULL);
+ return 0;
+}
+
+static int nmi_resume(struct sys_device *dev)
+{
+ if (nmi_enabled == 1)
+ nmi_cpu_start(NULL);
+ return 0;
+}
+
+static struct sysdev_class oprofile_sysclass = {
+ .name = "oprofile",
+ .resume = nmi_resume,
+ .suspend = nmi_suspend,
+};
+
+static struct sys_device device_oprofile = {
+ .id = 0,
+ .cls = &oprofile_sysclass,
+};
+
+static int __init init_sysfs(void)
+{
+ int error;
+
+ error = sysdev_class_register(&oprofile_sysclass);
+ if (!error)
+ error = sysdev_register(&device_oprofile);
+ return error;
+}
+
+static void exit_sysfs(void)
+{
+ sysdev_unregister(&device_oprofile);
+ sysdev_class_unregister(&oprofile_sysclass);
+}
+
+#else
+#define init_sysfs() do { } while (0)
+#define exit_sysfs() do { } while (0)
+#endif /* CONFIG_PM */
static int __init p4_init(char **cpu_type)
{
__u8 cpu_model = boot_cpu_data.x86_model;
- if (!p4force && (cpu_model > 6 || cpu_model == 5))
+ if (cpu_model > 6 || cpu_model == 5)
return 0;
#ifndef CONFIG_SMP
return 0;
}
+static int force_arch_perfmon;
+static int force_cpu_type(const char *str, struct kernel_param *kp)
+{
+ if (!strcmp(str, "archperfmon")) {
+ force_arch_perfmon = 1;
+ printk(KERN_INFO "oprofile: forcing architectural perfmon\n");
+ }
+
+ return 0;
+}
+module_param_call(cpu_type, force_cpu_type, NULL, NULL, 0);
+
static int __init ppro_init(char **cpu_type)
{
__u8 cpu_model = boot_cpu_data.x86_model;
+ if (force_arch_perfmon && cpu_has_arch_perfmon)
+ return 0;
+
switch (cpu_model) {
case 0 ... 2:
*cpu_type = "i386/ppro";
*cpu_type = "i386/pii";
break;
case 6 ... 8:
+ case 10 ... 11:
*cpu_type = "i386/piii";
break;
case 9:
+ case 13:
*cpu_type = "i386/p6_mobile";
break;
- case 10 ... 13:
- *cpu_type = "i386/p6";
- break;
case 14:
*cpu_type = "i386/core";
break;
*cpu_type = "i386/core_2";
break;
case 26:
- *cpu_type = "i386/core_2";
+ model = &op_arch_perfmon_spec;
+ *cpu_type = "i386/core_i7";
+ break;
+ case 28:
+ *cpu_type = "i386/atom";
break;
default:
/* Unknown */
{
__u8 vendor = boot_cpu_data.x86_vendor;
__u8 family = boot_cpu_data.x86;
- char *cpu_type;
+ char *cpu_type = NULL;
int ret = 0;
if (!cpu_has_apic)
/* Needs to be at least an Athlon (or hammer in 32bit mode) */
switch (family) {
- default:
- return -ENODEV;
case 6:
- model = &op_amd_spec;
cpu_type = "i386/athlon";
break;
case 0xf:
- model = &op_amd_spec;
- /* Actually it could be i386/hammer too, but give
- user space an consistent name. */
+ /*
+ * Actually it could be i386/hammer too, but
+ * give user space an consistent name.
+ */
cpu_type = "x86-64/hammer";
break;
case 0x10:
- model = &op_amd_spec;
cpu_type = "x86-64/family10";
break;
case 0x11:
- model = &op_amd_spec;
cpu_type = "x86-64/family11h";
break;
+ default:
+ return -ENODEV;
}
+ model = &op_amd_spec;
break;
case X86_VENDOR_INTEL:
switch (family) {
/* Pentium IV */
case 0xf:
- if (!p4_init(&cpu_type))
- return -ENODEV;
+ p4_init(&cpu_type);
break;
/* A P6-class processor */
case 6:
- if (!ppro_init(&cpu_type))
- return -ENODEV;
+ ppro_init(&cpu_type);
break;
default:
- return -ENODEV;
+ break;
}
+
+ if (cpu_type)
+ break;
+
+ if (!cpu_has_arch_perfmon)
+ return -ENODEV;
+
+ /* use arch perfmon as fallback */
+ cpu_type = "i386/arch_perfmon";
+ model = &op_arch_perfmon_spec;
break;
default:
register_cpu_notifier(&oprofile_cpu_nb);
#endif
/* default values, can be overwritten by model */
- __raw_get_cpu_var(switch_index) = 0;
ops->create_files = nmi_create_files;
ops->setup = nmi_setup;
ops->shutdown = nmi_shutdown;
ops->start = nmi_start;
ops->stop = nmi_stop;
ops->cpu_type = cpu_type;
- ops->switch_events = nmi_switch_event;
if (model->init)
ret = model->init(ops);
#ifdef CONFIG_SMP
unregister_cpu_notifier(&oprofile_cpu_nb);
#endif
+ }
if (model->exit)
model->exit();
- }
}