Release 4.10 arch/x86/events/amd/uncore.c
/*
* Copyright (C) 2013 Advanced Micro Devices, Inc.
*
* Author: Jacob Shin <jacob.shin@amd.com>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*/
#include <linux/perf_event.h>
#include <linux/percpu.h>
#include <linux/types.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/cpu.h>
#include <linux/cpumask.h>
#include <asm/cpufeature.h>
#include <asm/perf_event.h>
#include <asm/msr.h>
#define NUM_COUNTERS_NB 4
#define NUM_COUNTERS_L2 4
#define MAX_COUNTERS NUM_COUNTERS_NB
#define RDPMC_BASE_NB 6
#define RDPMC_BASE_L2 10
#define COUNTER_SHIFT 16
static HLIST_HEAD(uncore_unused_list);
struct amd_uncore {
int id;
int refcnt;
int cpu;
int num_counters;
int rdpmc_base;
u32 msr_base;
cpumask_t *active_mask;
struct pmu *pmu;
struct perf_event *events[MAX_COUNTERS];
struct hlist_node node;
};
static struct amd_uncore * __percpu *amd_uncore_nb;
static struct amd_uncore * __percpu *amd_uncore_l2;
static struct pmu amd_nb_pmu;
static struct pmu amd_l2_pmu;
static cpumask_t amd_nb_active_mask;
static cpumask_t amd_l2_active_mask;
static bool is_nb_event(struct perf_event *event)
{
return event->pmu->type == amd_nb_pmu.type;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 22 | 100.00% | 1 | 100.00% |
| Total | 22 | 100.00% | 1 | 100.00% |
static bool is_l2_event(struct perf_event *event)
{
return event->pmu->type == amd_l2_pmu.type;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 22 | 100.00% | 1 | 100.00% |
| Total | 22 | 100.00% | 1 | 100.00% |
static struct amd_uncore *event_to_amd_uncore(struct perf_event *event)
{
if (is_nb_event(event) && amd_uncore_nb)
return *per_cpu_ptr(amd_uncore_nb, event->cpu);
else if (is_l2_event(event) && amd_uncore_l2)
return *per_cpu_ptr(amd_uncore_l2, event->cpu);
return NULL;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 57 | 100.00% | 1 | 100.00% |
| Total | 57 | 100.00% | 1 | 100.00% |
static void amd_uncore_read(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
u64 prev, new;
s64 delta;
/*
* since we do not enable counter overflow interrupts,
* we do not have to worry about prev_count changing on us
*/
prev = local64_read(&hwc->prev_count);
rdpmcl(hwc->event_base_rdpmc, new);
local64_set(&hwc->prev_count, new);
delta = (new << COUNTER_SHIFT) - (prev << COUNTER_SHIFT);
delta >>= COUNTER_SHIFT;
local64_add(delta, &event->count);
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 87 | 100.00% | 1 | 100.00% |
| Total | 87 | 100.00% | 1 | 100.00% |
static void amd_uncore_start(struct perf_event *event, int flags)
{
struct hw_perf_event *hwc = &event->hw;
if (flags & PERF_EF_RELOAD)
wrmsrl(hwc->event_base, (u64)local64_read(&hwc->prev_count));
hwc->state = 0;
wrmsrl(hwc->config_base, (hwc->config | ARCH_PERFMON_EVENTSEL_ENABLE));
perf_event_update_userpage(event);
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 74 | 100.00% | 1 | 100.00% |
| Total | 74 | 100.00% | 1 | 100.00% |
static void amd_uncore_stop(struct perf_event *event, int flags)
{
struct hw_perf_event *hwc = &event->hw;
wrmsrl(hwc->config_base, hwc->config);
hwc->state |= PERF_HES_STOPPED;
if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
amd_uncore_read(event);
hwc->state |= PERF_HES_UPTODATE;
}
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 71 | 100.00% | 1 | 100.00% |
| Total | 71 | 100.00% | 1 | 100.00% |
static int amd_uncore_add(struct perf_event *event, int flags)
{
int i;
struct amd_uncore *uncore = event_to_amd_uncore(event);
struct hw_perf_event *hwc = &event->hw;
/* are we already assigned? */
if (hwc->idx != -1 && uncore->events[hwc->idx] == event)
goto out;
for (i = 0; i < uncore->num_counters; i++) {
if (uncore->events[i] == event) {
hwc->idx = i;
goto out;
}
}
/* if not, take the first available counter */
hwc->idx = -1;
for (i = 0; i < uncore->num_counters; i++) {
if (cmpxchg(&uncore->events[i], NULL, event) == NULL) {
hwc->idx = i;
break;
}
}
out:
if (hwc->idx == -1)
return -EBUSY;
hwc->config_base = uncore->msr_base + (2 * hwc->idx);
hwc->event_base = uncore->msr_base + 1 + (2 * hwc->idx);
hwc->event_base_rdpmc = uncore->rdpmc_base + hwc->idx;
hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
if (flags & PERF_EF_START)
amd_uncore_start(event, PERF_EF_RELOAD);
return 0;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 238 | 100.00% | 1 | 100.00% |
| Total | 238 | 100.00% | 1 | 100.00% |
static void amd_uncore_del(struct perf_event *event, int flags)
{
int i;
struct amd_uncore *uncore = event_to_amd_uncore(event);
struct hw_perf_event *hwc = &event->hw;
amd_uncore_stop(event, PERF_EF_UPDATE);
for (i = 0; i < uncore->num_counters; i++) {
if (cmpxchg(&uncore->events[i], event, NULL) == event)
break;
}
hwc->idx = -1;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 88 | 100.00% | 1 | 100.00% |
| Total | 88 | 100.00% | 1 | 100.00% |
static int amd_uncore_event_init(struct perf_event *event)
{
struct amd_uncore *uncore;
struct hw_perf_event *hwc = &event->hw;
if (event->attr.type != event->pmu->type)
return -ENOENT;
/*
* NB and L2 counters (MSRs) are shared across all cores that share the
* same NB / L2 cache. Interrupts can be directed to a single target
* core, however, event counts generated by processes running on other
* cores cannot be masked out. So we do not support sampling and
* per-thread events.
*/
if (is_sampling_event(event) || event->attach_state & PERF_ATTACH_TASK)
return -EINVAL;
/* NB and L2 counters do not have usr/os/guest/host bits */
if (event->attr.exclude_user || event->attr.exclude_kernel ||
event->attr.exclude_host || event->attr.exclude_guest)
return -EINVAL;
/* and we do not enable counter overflow interrupts */
hwc->config = event->attr.config & AMD64_RAW_EVENT_MASK_NB;
hwc->idx = -1;
if (event->cpu < 0)
return -EINVAL;
uncore = event_to_amd_uncore(event);
if (!uncore)
return -ENODEV;
/*
* since request can come in to any of the shared cores, we will remap
* to a single common cpu.
*/
event->cpu = uncore->cpu;
return 0;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 153 | 100.00% | 1 | 100.00% |
| Total | 153 | 100.00% | 1 | 100.00% |
static ssize_t amd_uncore_attr_show_cpumask(struct device *dev,
struct device_attribute *attr,
char *buf)
{
cpumask_t *active_mask;
struct pmu *pmu = dev_get_drvdata(dev);
if (pmu->type == amd_nb_pmu.type)
active_mask = &amd_nb_active_mask;
else if (pmu->type == amd_l2_pmu.type)
active_mask = &amd_l2_active_mask;
else
return 0;
return cpumap_print_to_pagebuf(true, buf, active_mask);
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 75 | 94.94% | 1 | 50.00% |
sudeep holla | sudeep holla | 4 | 5.06% | 1 | 50.00% |
| Total | 79 | 100.00% | 2 | 100.00% |
static DEVICE_ATTR(cpumask, S_IRUGO, amd_uncore_attr_show_cpumask, NULL);
static struct attribute *amd_uncore_attrs[] = {
&dev_attr_cpumask.attr,
NULL,
};
static struct attribute_group amd_uncore_attr_group = {
.attrs = amd_uncore_attrs,
};
PMU_FORMAT_ATTR(event, "config:0-7,32-35");
PMU_FORMAT_ATTR(umask, "config:8-15");
static struct attribute *amd_uncore_format_attr[] = {
&format_attr_event.attr,
&format_attr_umask.attr,
NULL,
};
static struct attribute_group amd_uncore_format_group = {
.name = "format",
.attrs = amd_uncore_format_attr,
};
static const struct attribute_group *amd_uncore_attr_groups[] = {
&amd_uncore_attr_group,
&amd_uncore_format_group,
NULL,
};
static struct pmu amd_nb_pmu = {
.task_ctx_nr = perf_invalid_context,
.attr_groups = amd_uncore_attr_groups,
.name = "amd_nb",
.event_init = amd_uncore_event_init,
.add = amd_uncore_add,
.del = amd_uncore_del,
.start = amd_uncore_start,
.stop = amd_uncore_stop,
.read = amd_uncore_read,
};
static struct pmu amd_l2_pmu = {
.task_ctx_nr = perf_invalid_context,
.attr_groups = amd_uncore_attr_groups,
.name = "amd_l2",
.event_init = amd_uncore_event_init,
.add = amd_uncore_add,
.del = amd_uncore_del,
.start = amd_uncore_start,
.stop = amd_uncore_stop,
.read = amd_uncore_read,
};
static struct amd_uncore *amd_uncore_alloc(unsigned int cpu)
{
return kzalloc_node(sizeof(struct amd_uncore), GFP_KERNEL,
cpu_to_node(cpu));
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 29 | 100.00% | 1 | 100.00% |
| Total | 29 | 100.00% | 1 | 100.00% |
static int amd_uncore_cpu_up_prepare(unsigned int cpu)
{
struct amd_uncore *uncore_nb = NULL, *uncore_l2;
if (amd_uncore_nb) {
uncore_nb = amd_uncore_alloc(cpu);
if (!uncore_nb)
goto fail;
uncore_nb->cpu = cpu;
uncore_nb->num_counters = NUM_COUNTERS_NB;
uncore_nb->rdpmc_base = RDPMC_BASE_NB;
uncore_nb->msr_base = MSR_F15H_NB_PERF_CTL;
uncore_nb->active_mask = &amd_nb_active_mask;
uncore_nb->pmu = &amd_nb_pmu;
uncore_nb->id = -1;
*per_cpu_ptr(amd_uncore_nb, cpu) = uncore_nb;
}
if (amd_uncore_l2) {
uncore_l2 = amd_uncore_alloc(cpu);
if (!uncore_l2)
goto fail;
uncore_l2->cpu = cpu;
uncore_l2->num_counters = NUM_COUNTERS_L2;
uncore_l2->rdpmc_base = RDPMC_BASE_L2;
uncore_l2->msr_base = MSR_F16H_L2I_PERF_CTL;
uncore_l2->active_mask = &amd_l2_active_mask;
uncore_l2->pmu = &amd_l2_pmu;
uncore_l2->id = -1;
*per_cpu_ptr(amd_uncore_l2, cpu) = uncore_l2;
}
return 0;
fail:
if (amd_uncore_nb)
*per_cpu_ptr(amd_uncore_nb, cpu) = NULL;
kfree(uncore_nb);
return -ENOMEM;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 119 | 59.50% | 1 | 25.00% |
zhouyi zhou | zhouyi zhou | 53 | 26.50% | 1 | 25.00% |
thomas gleixner | thomas gleixner | 14 | 7.00% | 1 | 25.00% |
sebastian andrzej siewior | sebastian andrzej siewior | 14 | 7.00% | 1 | 25.00% |
| Total | 200 | 100.00% | 4 | 100.00% |
static struct amd_uncore *
amd_uncore_find_online_sibling(struct amd_uncore *this,
struct amd_uncore * __percpu *uncores)
{
unsigned int cpu;
struct amd_uncore *that;
for_each_online_cpu(cpu) {
that = *per_cpu_ptr(uncores, cpu);
if (!that)
continue;
if (this == that)
continue;
if (this->id == that->id) {
hlist_add_head(&this->node, &uncore_unused_list);
this = that;
break;
}
}
this->refcnt++;
return this;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 85 | 90.43% | 1 | 50.00% |
sebastian andrzej siewior | sebastian andrzej siewior | 9 | 9.57% | 1 | 50.00% |
| Total | 94 | 100.00% | 2 | 100.00% |
static int amd_uncore_cpu_starting(unsigned int cpu)
{
unsigned int eax, ebx, ecx, edx;
struct amd_uncore *uncore;
if (amd_uncore_nb) {
uncore = *per_cpu_ptr(amd_uncore_nb, cpu);
cpuid(0x8000001e, &eax, &ebx, &ecx, &edx);
uncore->id = ecx & 0xff;
uncore = amd_uncore_find_online_sibling(uncore, amd_uncore_nb);
*per_cpu_ptr(amd_uncore_nb, cpu) = uncore;
}
if (amd_uncore_l2) {
unsigned int apicid = cpu_data(cpu).apicid;
unsigned int nshared;
uncore = *per_cpu_ptr(amd_uncore_l2, cpu);
cpuid_count(0x8000001d, 2, &eax, &ebx, &ecx, &edx);
nshared = ((eax >> 14) & 0xfff) + 1;
uncore->id = apicid - (apicid % nshared);
uncore = amd_uncore_find_online_sibling(uncore, amd_uncore_l2);
*per_cpu_ptr(amd_uncore_l2, cpu) = uncore;
}
return 0;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 179 | 97.81% | 1 | 50.00% |
richard cochran | richard cochran | 4 | 2.19% | 1 | 50.00% |
| Total | 183 | 100.00% | 2 | 100.00% |
static void uncore_clean_online(void)
{
struct amd_uncore *uncore;
struct hlist_node *n;
hlist_for_each_entry_safe(uncore, n, &uncore_unused_list, node) {
hlist_del(&uncore->node);
kfree(uncore);
}
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
sebastian andrzej siewior | sebastian andrzej siewior | 43 | 100.00% | 1 | 100.00% |
| Total | 43 | 100.00% | 1 | 100.00% |
static void uncore_online(unsigned int cpu,
struct amd_uncore * __percpu *uncores)
{
struct amd_uncore *uncore = *per_cpu_ptr(uncores, cpu);
uncore_clean_online();
if (cpu == uncore->cpu)
cpumask_set_cpu(cpu, uncore->active_mask);
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 48 | 96.00% | 1 | 50.00% |
sebastian andrzej siewior | sebastian andrzej siewior | 2 | 4.00% | 1 | 50.00% |
| Total | 50 | 100.00% | 2 | 100.00% |
static int amd_uncore_cpu_online(unsigned int cpu)
{
if (amd_uncore_nb)
uncore_online(cpu, amd_uncore_nb);
if (amd_uncore_l2)
uncore_online(cpu, amd_uncore_l2);
return 0;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 31 | 88.57% | 1 | 50.00% |
richard cochran | richard cochran | 4 | 11.43% | 1 | 50.00% |
| Total | 35 | 100.00% | 2 | 100.00% |
static void uncore_down_prepare(unsigned int cpu,
struct amd_uncore * __percpu *uncores)
{
unsigned int i;
struct amd_uncore *this = *per_cpu_ptr(uncores, cpu);
if (this->cpu != cpu)
return;
/* this cpu is going down, migrate to a shared sibling if possible */
for_each_online_cpu(i) {
struct amd_uncore *that = *per_cpu_ptr(uncores, i);
if (cpu == i)
continue;
if (this == that) {
perf_pmu_migrate_context(this->pmu, cpu, i);
cpumask_clear_cpu(cpu, that->active_mask);
cpumask_set_cpu(i, that->active_mask);
that->cpu = i;
break;
}
}
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 114 | 100.00% | 1 | 100.00% |
| Total | 114 | 100.00% | 1 | 100.00% |
static int amd_uncore_cpu_down_prepare(unsigned int cpu)
{
if (amd_uncore_nb)
uncore_down_prepare(cpu, amd_uncore_nb);
if (amd_uncore_l2)
uncore_down_prepare(cpu, amd_uncore_l2);
return 0;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 31 | 88.57% | 1 | 50.00% |
richard cochran | richard cochran | 4 | 11.43% | 1 | 50.00% |
| Total | 35 | 100.00% | 2 | 100.00% |
static void uncore_dead(unsigned int cpu, struct amd_uncore * __percpu *uncores)
{
struct amd_uncore *uncore = *per_cpu_ptr(uncores, cpu);
if (cpu == uncore->cpu)
cpumask_clear_cpu(cpu, uncore->active_mask);
if (!--uncore->refcnt)
kfree(uncore);
*per_cpu_ptr(uncores, cpu) = NULL;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 69 | 98.57% | 1 | 50.00% |
zhouyi zhou | zhouyi zhou | 1 | 1.43% | 1 | 50.00% |
| Total | 70 | 100.00% | 2 | 100.00% |
static int amd_uncore_cpu_dead(unsigned int cpu)
{
if (amd_uncore_nb)
uncore_dead(cpu, amd_uncore_nb);
if (amd_uncore_l2)
uncore_dead(cpu, amd_uncore_l2);
return 0;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 32 | 91.43% | 1 | 33.33% |
richard cochran | richard cochran | 2 | 5.71% | 1 | 33.33% |
zhouyi zhou | zhouyi zhou | 1 | 2.86% | 1 | 33.33% |
| Total | 35 | 100.00% | 3 | 100.00% |
static int __init amd_uncore_init(void)
{
int ret = -ENODEV;
if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD)
goto fail_nodev;
if (!boot_cpu_has(X86_FEATURE_TOPOEXT))
goto fail_nodev;
if (boot_cpu_has(X86_FEATURE_PERFCTR_NB)) {
amd_uncore_nb = alloc_percpu(struct amd_uncore *);
if (!amd_uncore_nb) {
ret = -ENOMEM;
goto fail_nb;
}
ret = perf_pmu_register(&amd_nb_pmu, amd_nb_pmu.name, -1);
if (ret)
goto fail_nb;
pr_info("perf: AMD NB counters detected\n");
ret = 0;
}
if (boot_cpu_has(X86_FEATURE_PERFCTR_L2)) {
amd_uncore_l2 = alloc_percpu(struct amd_uncore *);
if (!amd_uncore_l2) {
ret = -ENOMEM;
goto fail_l2;
}
ret = perf_pmu_register(&amd_l2_pmu, amd_l2_pmu.name, -1);
if (ret)
goto fail_l2;
pr_info("perf: AMD L2I counters detected\n");
ret = 0;
}
/*
* Install callbacks. Core will call them for each online cpu.
*/
if (cpuhp_setup_state(CPUHP_PERF_X86_AMD_UNCORE_PREP,
"perf/x86/amd/uncore:prepare",
amd_uncore_cpu_up_prepare, amd_uncore_cpu_dead))
goto fail_l2;
if (cpuhp_setup_state(CPUHP_AP_PERF_X86_AMD_UNCORE_STARTING,
"perf/x86/amd/uncore:starting",
amd_uncore_cpu_starting, NULL))
goto fail_prep;
if (cpuhp_setup_state(CPUHP_AP_PERF_X86_AMD_UNCORE_ONLINE,
"perf/x86/amd/uncore:online",
amd_uncore_cpu_online,
amd_uncore_cpu_down_prepare))
goto fail_start;
return 0;
fail_start:
cpuhp_remove_state(CPUHP_AP_PERF_X86_AMD_UNCORE_STARTING);
fail_prep:
cpuhp_remove_state(CPUHP_PERF_X86_AMD_UNCORE_PREP);
fail_l2:
if (boot_cpu_has(X86_FEATURE_PERFCTR_NB))
perf_pmu_unregister(&amd_nb_pmu);
if (amd_uncore_l2)
free_percpu(amd_uncore_l2);
fail_nb:
if (amd_uncore_nb)
free_percpu(amd_uncore_nb);
fail_nodev:
return ret;
}
Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 106 | 39.11% | 2 | 28.57% |
zhouyi zhou | zhouyi zhou | 104 | 38.38% | 1 | 14.29% |
richard cochran | richard cochran | 38 | 14.02% | 1 | 14.29% |
borislav petkov | borislav petkov | 18 | 6.64% | 1 | 14.29% |
thomas gleixner | thomas gleixner | 3 | 1.11% | 1 | 14.29% |
chen yucong | chen yucong | 2 | 0.74% | 1 | 14.29% |
| Total | 271 | 100.00% | 7 | 100.00% |
device_initcall(amd_uncore_init);
Overall Contributors
| Person | Tokens | Prop | Commits | CommitProp |
jacob shin | jacob shin | 2070 | 85.96% | 2 | 18.18% |
zhouyi zhou | zhouyi zhou | 159 | 6.60% | 1 | 9.09% |
sebastian andrzej siewior | sebastian andrzej siewior | 76 | 3.16% | 1 | 9.09% |
richard cochran | richard cochran | 52 | 2.16% | 1 | 9.09% |
borislav petkov | borislav petkov | 18 | 0.75% | 1 | 9.09% |
thomas gleixner | thomas gleixner | 17 | 0.71% | 2 | 18.18% |
peter zijlstra | peter zijlstra | 10 | 0.42% | 1 | 9.09% |
sudeep holla | sudeep holla | 4 | 0.17% | 1 | 9.09% |
chen yucong | chen yucong | 2 | 0.08% | 1 | 9.09% |
| Total | 2408 | 100.00% | 11 | 100.00% |
Information contained on this website is for historical information purposes only and does not indicate or represent copyright ownership.