Commit de0428a7 authored by Kevin Winchester's avatar Kevin Winchester Committed by Ingo Molnar

x86, perf: Clean up perf_event cpu code

The CPU support for perf events on x86 was implemented via included C files
with #ifdefs.  Clean this up by creating a new header file and compiling
the vendor-specific files as needed.
Signed-off-by: default avatarKevin Winchester <kjwinchester@gmail.com>
Signed-off-by: default avatarPeter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1314747665-2090-1-git-send-email-kjwinchester@gmail.comSigned-off-by: default avatarIngo Molnar <mingo@elte.hu>
parent ed3982cf
......@@ -28,6 +28,11 @@ obj-$(CONFIG_CPU_SUP_UMC_32) += umc.o
obj-$(CONFIG_PERF_EVENTS) += perf_event.o
ifdef CONFIG_PERF_EVENTS
obj-$(CONFIG_CPU_SUP_AMD) += perf_event_amd.o
obj-$(CONFIG_CPU_SUP_INTEL) += perf_event_p6.o perf_event_p4.o perf_event_intel_lbr.o perf_event_intel_ds.o perf_event_intel.o
endif
obj-$(CONFIG_X86_MCE) += mcheck/
obj-$(CONFIG_MTRR) += mtrr/
......
......@@ -32,6 +32,8 @@
#include <asm/smp.h>
#include <asm/alternative.h>
#include "perf_event.h"
#if 0
#undef wrmsrl
#define wrmsrl(msr, val) \
......@@ -43,285 +45,17 @@ do { \
} while (0)
#endif
/*
* | NHM/WSM | SNB |
* register -------------------------------
* | HT | no HT | HT | no HT |
*-----------------------------------------
* offcore | core | core | cpu | core |
* lbr_sel | core | core | cpu | core |
* ld_lat | cpu | core | cpu | core |
*-----------------------------------------
*
* Given that there is a small number of shared regs,
* we can pre-allocate their slot in the per-cpu
* per-core reg tables.
*/
enum extra_reg_type {
EXTRA_REG_NONE = -1, /* not used */
EXTRA_REG_RSP_0 = 0, /* offcore_response_0 */
EXTRA_REG_RSP_1 = 1, /* offcore_response_1 */
EXTRA_REG_MAX /* number of entries needed */
};
struct event_constraint {
union {
unsigned long idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
u64 idxmsk64;
};
u64 code;
u64 cmask;
int weight;
};
struct amd_nb {
int nb_id; /* NorthBridge id */
int refcnt; /* reference count */
struct perf_event *owners[X86_PMC_IDX_MAX];
struct event_constraint event_constraints[X86_PMC_IDX_MAX];
};
struct intel_percore;
#define MAX_LBR_ENTRIES 16
struct cpu_hw_events {
/*
* Generic x86 PMC bits
*/
struct perf_event *events[X86_PMC_IDX_MAX]; /* in counter order */
unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
unsigned long running[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
int enabled;
int n_events;
int n_added;
int n_txn;
int assign[X86_PMC_IDX_MAX]; /* event to counter assignment */
u64 tags[X86_PMC_IDX_MAX];
struct perf_event *event_list[X86_PMC_IDX_MAX]; /* in enabled order */
unsigned int group_flag;
/*
* Intel DebugStore bits
*/
struct debug_store *ds;
u64 pebs_enabled;
/*
* Intel LBR bits
*/
int lbr_users;
void *lbr_context;
struct perf_branch_stack lbr_stack;
struct perf_branch_entry lbr_entries[MAX_LBR_ENTRIES];
/*
* manage shared (per-core, per-cpu) registers
* used on Intel NHM/WSM/SNB
*/
struct intel_shared_regs *shared_regs;
/*
* AMD specific bits
*/
struct amd_nb *amd_nb;
void *kfree_on_online;
};
#define __EVENT_CONSTRAINT(c, n, m, w) {\
{ .idxmsk64 = (n) }, \
.code = (c), \
.cmask = (m), \
.weight = (w), \
}
#define EVENT_CONSTRAINT(c, n, m) \
__EVENT_CONSTRAINT(c, n, m, HWEIGHT(n))
/*
* Constraint on the Event code.
*/
#define INTEL_EVENT_CONSTRAINT(c, n) \
EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT)
/*
* Constraint on the Event code + UMask + fixed-mask
*
* filter mask to validate fixed counter events.
* the following filters disqualify for fixed counters:
* - inv
* - edge
* - cnt-mask
* The other filters are supported by fixed counters.
* The any-thread option is supported starting with v3.
*/
#define FIXED_EVENT_CONSTRAINT(c, n) \
EVENT_CONSTRAINT(c, (1ULL << (32+n)), X86_RAW_EVENT_MASK)
/*
* Constraint on the Event code + UMask
*/
#define INTEL_UEVENT_CONSTRAINT(c, n) \
EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK)
#define EVENT_CONSTRAINT_END \
EVENT_CONSTRAINT(0, 0, 0)
#define for_each_event_constraint(e, c) \
for ((e) = (c); (e)->weight; (e)++)
/*
* Per register state.
*/
struct er_account {
raw_spinlock_t lock; /* per-core: protect structure */
u64 config; /* extra MSR config */
u64 reg; /* extra MSR number */
atomic_t ref; /* reference count */
};
/*
* Extra registers for specific events.
*
* Some events need large masks and require external MSRs.
* Those extra MSRs end up being shared for all events on
* a PMU and sometimes between PMU of sibling HT threads.
* In either case, the kernel needs to handle conflicting
* accesses to those extra, shared, regs. The data structure
* to manage those registers is stored in cpu_hw_event.
*/
struct extra_reg {
unsigned int event;
unsigned int msr;
u64 config_mask;
u64 valid_mask;
int idx; /* per_xxx->regs[] reg index */
};
#define EVENT_EXTRA_REG(e, ms, m, vm, i) { \
.event = (e), \
.msr = (ms), \
.config_mask = (m), \
.valid_mask = (vm), \
.idx = EXTRA_REG_##i \
}
#define INTEL_EVENT_EXTRA_REG(event, msr, vm, idx) \
EVENT_EXTRA_REG(event, msr, ARCH_PERFMON_EVENTSEL_EVENT, vm, idx)
struct x86_pmu x86_pmu __read_mostly;
#define EVENT_EXTRA_END EVENT_EXTRA_REG(0, 0, 0, 0, RSP_0)
union perf_capabilities {
struct {
u64 lbr_format : 6;
u64 pebs_trap : 1;
u64 pebs_arch_reg : 1;
u64 pebs_format : 4;
u64 smm_freeze : 1;
};
u64 capabilities;
};
/*
* struct x86_pmu - generic x86 pmu
*/
struct x86_pmu {
/*
* Generic x86 PMC bits
*/
const char *name;
int version;
int (*handle_irq)(struct pt_regs *);
void (*disable_all)(void);
void (*enable_all)(int added);
void (*enable)(struct perf_event *);
void (*disable)(struct perf_event *);
int (*hw_config)(struct perf_event *event);
int (*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign);
unsigned eventsel;
unsigned perfctr;
u64 (*event_map)(int);
int max_events;
int num_counters;
int num_counters_fixed;
int cntval_bits;
u64 cntval_mask;
int apic;
u64 max_period;
struct event_constraint *
(*get_event_constraints)(struct cpu_hw_events *cpuc,
struct perf_event *event);
void (*put_event_constraints)(struct cpu_hw_events *cpuc,
struct perf_event *event);
struct event_constraint *event_constraints;
void (*quirks)(void);
int perfctr_second_write;
int (*cpu_prepare)(int cpu);
void (*cpu_starting)(int cpu);
void (*cpu_dying)(int cpu);
void (*cpu_dead)(int cpu);
/*
* Intel Arch Perfmon v2+
*/
u64 intel_ctrl;
union perf_capabilities intel_cap;
/*
* Intel DebugStore bits
*/
int bts, pebs;
int bts_active, pebs_active;
int pebs_record_size;
void (*drain_pebs)(struct pt_regs *regs);
struct event_constraint *pebs_constraints;
/*
* Intel LBR
*/
unsigned long lbr_tos, lbr_from, lbr_to; /* MSR base regs */
int lbr_nr; /* hardware stack size */
/*
* Extra registers for events
*/
struct extra_reg *extra_regs;
unsigned int er_flags;
};
#define ERF_NO_HT_SHARING 1
#define ERF_HAS_RSP_1 2
static struct x86_pmu x86_pmu __read_mostly;
static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
.enabled = 1,
};
static int x86_perf_event_set_period(struct perf_event *event);
/*
* Generalized hw caching related hw_event table, filled
* in on a per model basis. A value of 0 means
* 'not supported', -1 means 'hw_event makes no sense on
* this CPU', any other value means the raw hw_event
* ID.
*/
#define C(x) PERF_COUNT_HW_CACHE_##x
static u64 __read_mostly hw_cache_event_ids
u64 __read_mostly hw_cache_event_ids
[PERF_COUNT_HW_CACHE_MAX]
[PERF_COUNT_HW_CACHE_OP_MAX]
[PERF_COUNT_HW_CACHE_RESULT_MAX];
static u64 __read_mostly hw_cache_extra_regs
u64 __read_mostly hw_cache_extra_regs
[PERF_COUNT_HW_CACHE_MAX]
[PERF_COUNT_HW_CACHE_OP_MAX]
[PERF_COUNT_HW_CACHE_RESULT_MAX];
......@@ -331,8 +65,7 @@ static u64 __read_mostly hw_cache_extra_regs
* Can only be executed on the CPU where the event is active.
* Returns the delta events processed.
*/
static u64
x86_perf_event_update(struct perf_event *event)
u64 x86_perf_event_update(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
int shift = 64 - x86_pmu.cntval_bits;
......@@ -375,30 +108,6 @@ x86_perf_event_update(struct perf_event *event)
return new_raw_count;
}
static inline int x86_pmu_addr_offset(int index)
{
int offset;
/* offset = X86_FEATURE_PERFCTR_CORE ? index << 1 : index */
alternative_io(ASM_NOP2,
"shll $1, %%eax",
X86_FEATURE_PERFCTR_CORE,
"=a" (offset),
"a" (index));
return offset;
}
static inline unsigned int x86_pmu_config_addr(int index)
{
return x86_pmu.eventsel + x86_pmu_addr_offset(index);
}
static inline unsigned int x86_pmu_event_addr(int index)
{
return x86_pmu.perfctr + x86_pmu_addr_offset(index);
}
/*
* Find and validate any extra registers to set up.
*/
......@@ -534,9 +243,6 @@ static bool check_hw_exists(void)
return false;
}
static void reserve_ds_buffers(void);
static void release_ds_buffers(void);
static void hw_perf_event_destroy(struct perf_event *event)
{
if (atomic_dec_and_mutex_lock(&active_events, &pmc_reserve_mutex)) {
......@@ -585,7 +291,7 @@ set_ext_hw_attr(struct hw_perf_event *hwc, struct perf_event *event)
return x86_pmu_extra_regs(val, event);
}
static int x86_setup_perfctr(struct perf_event *event)
int x86_setup_perfctr(struct perf_event *event)
{
struct perf_event_attr *attr = &event->attr;
struct hw_perf_event *hwc = &event->hw;
......@@ -649,7 +355,7 @@ static int x86_setup_perfctr(struct perf_event *event)
return 0;
}
static int x86_pmu_hw_config(struct perf_event *event)
int x86_pmu_hw_config(struct perf_event *event)
{
if (event->attr.precise_ip) {
int precise = 0;
......@@ -725,7 +431,7 @@ static int __x86_pmu_event_init(struct perf_event *event)
return x86_pmu.hw_config(event);
}
static void x86_pmu_disable_all(void)
void x86_pmu_disable_all(void)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
int idx;
......@@ -760,15 +466,7 @@ static void x86_pmu_disable(struct pmu *pmu)
x86_pmu.disable_all();
}
static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc,
u64 enable_mask)
{
if (hwc->extra_reg.reg)
wrmsrl(hwc->extra_reg.reg, hwc->extra_reg.config);
wrmsrl(hwc->config_base, hwc->config | enable_mask);
}
static void x86_pmu_enable_all(int added)
void x86_pmu_enable_all(int added)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
int idx;
......@@ -790,7 +488,7 @@ static inline int is_x86_event(struct perf_event *event)
return event->pmu == &pmu;
}
static int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
{
struct event_constraint *c, *constraints[X86_PMC_IDX_MAX];
unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
......@@ -961,7 +659,6 @@ static inline int match_prev_assignment(struct hw_perf_event *hwc,
}
static void x86_pmu_start(struct perf_event *event, int flags);
static void x86_pmu_stop(struct perf_event *event, int flags);
static void x86_pmu_enable(struct pmu *pmu)
{
......@@ -1033,21 +730,13 @@ static void x86_pmu_enable(struct pmu *pmu)
x86_pmu.enable_all(added);
}
static inline void x86_pmu_disable_event(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
wrmsrl(hwc->config_base, hwc->config);
}
static DEFINE_PER_CPU(u64 [X86_PMC_IDX_MAX], pmc_prev_left);
/*
* Set the next IRQ period, based on the hwc->period_left value.
* To be called with the event disabled in hw:
*/
static int
x86_perf_event_set_period(struct perf_event *event)
int x86_perf_event_set_period(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
s64 left = local64_read(&hwc->period_left);
......@@ -1107,7 +796,7 @@ x86_perf_event_set_period(struct perf_event *event)
return ret;
}
static void x86_pmu_enable_event(struct perf_event *event)
void x86_pmu_enable_event(struct perf_event *event)
{
if (__this_cpu_read(cpu_hw_events.enabled))
__x86_pmu_enable_event(&event->hw,
......@@ -1246,7 +935,7 @@ void perf_event_print_debug(void)
local_irq_restore(flags);
}
static void x86_pmu_stop(struct perf_event *event, int flags)
void x86_pmu_stop(struct perf_event *event, int flags)
{
struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
struct hw_perf_event *hwc = &event->hw;
......@@ -1299,7 +988,7 @@ static void x86_pmu_del(struct perf_event *event, int flags)
perf_event_update_userpage(event);
}
static int x86_pmu_handle_irq(struct pt_regs *regs)
int x86_pmu_handle_irq(struct pt_regs *regs)
{
struct perf_sample_data data;
struct cpu_hw_events *cpuc;
......@@ -1439,30 +1128,8 @@ static __read_mostly struct notifier_block perf_event_nmi_notifier = {
.priority = NMI_LOCAL_LOW_PRIOR,
};
static struct event_constraint unconstrained;
static struct event_constraint emptyconstraint;
static struct event_constraint *
x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
{
struct event_constraint *c;
if (x86_pmu.event_constraints) {
for_each_event_constraint(c, x86_pmu.event_constraints) {
if ((event->hw.config & c->cmask) == c->code)
return c;
}
}
return &unconstrained;
}
#include "perf_event_amd.c"
#include "perf_event_p6.c"
#include "perf_event_p4.c"
#include "perf_event_intel_lbr.c"
#include "perf_event_intel_ds.c"
#include "perf_event_intel.c"
struct event_constraint emptyconstraint;
struct event_constraint unconstrained;
static int __cpuinit
x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
......
/*
* Performance events x86 architecture header
*
* Copyright (C) 2008 Thomas Gleixner <tglx@linutronix.de>
* Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar
* Copyright (C) 2009 Jaswinder Singh Rajput
* Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter
* Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
* Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com>
* Copyright (C) 2009 Google, Inc., Stephane Eranian
*
* For licencing details see kernel-base/COPYING
*/
#include <linux/perf_event.h>
/*
* | NHM/WSM | SNB |
* register -------------------------------
* | HT | no HT | HT | no HT |
*-----------------------------------------
* offcore | core | core | cpu | core |
* lbr_sel | core | core | cpu | core |
* ld_lat | cpu | core | cpu | core |
*-----------------------------------------
*
* Given that there is a small number of shared regs,
* we can pre-allocate their slot in the per-cpu
* per-core reg tables.
*/
enum extra_reg_type {
EXTRA_REG_NONE = -1, /* not used */
EXTRA_REG_RSP_0 = 0, /* offcore_response_0 */
EXTRA_REG_RSP_1 = 1, /* offcore_response_1 */
EXTRA_REG_MAX /* number of entries needed */
};
struct event_constraint {
union {
unsigned long idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
u64 idxmsk64;
};
u64 code;
u64 cmask;
int weight;
};
struct amd_nb {
int nb_id; /* NorthBridge id */
int refcnt; /* reference count */
struct perf_event *owners[X86_PMC_IDX_MAX];
struct event_constraint event_constraints[X86_PMC_IDX_MAX];
};
/* The maximal number of PEBS events: */
#define MAX_PEBS_EVENTS 4
/*
* A debug store configuration.
*
* We only support architectures that use 64bit fields.
*/
struct debug_store {
u64 bts_buffer_base;
u64 bts_index;
u64 bts_absolute_maximum;
u64 bts_interrupt_threshold;
u64 pebs_buffer_base;
u64 pebs_index;
u64 pebs_absolute_maximum;
u64 pebs_interrupt_threshold;
u64 pebs_event_reset[MAX_PEBS_EVENTS];
};
/*
* Per register state.
*/
struct er_account {
raw_spinlock_t lock; /* per-core: protect structure */
u64 config; /* extra MSR config */
u64 reg; /* extra MSR number */
atomic_t ref; /* reference count */
};
/*
* Per core/cpu state
*
* Used to coordinate shared registers between HT threads or
* among events on a single PMU.
*/
struct intel_shared_regs {
struct er_account regs[EXTRA_REG_MAX];
int refcnt; /* per-core: #HT threads */
unsigned core_id; /* per-core: core id */
};
#define MAX_LBR_ENTRIES 16
struct cpu_hw_events {
/*
* Generic x86 PMC bits
*/
struct perf_event *events[X86_PMC_IDX_MAX]; /* in counter order */
unsigned long active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
unsigned long running[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
int enabled;
int n_events;
int n_added;
int n_txn;
int assign[X86_PMC_IDX_MAX]; /* event to counter assignment */
u64 tags[X86_PMC_IDX_MAX];
struct perf_event *event_list[X86_PMC_IDX_MAX]; /* in enabled order */
unsigned int group_flag;
/*
* Intel DebugStore bits
*/
struct debug_store *ds;
u64 pebs_enabled;
/*
* Intel LBR bits
*/
int lbr_users;
void *lbr_context;
struct perf_branch_stack lbr_stack;
struct perf_branch_entry lbr_entries[MAX_LBR_ENTRIES];
/*
* manage shared (per-core, per-cpu) registers
* used on Intel NHM/WSM/SNB