aboutsummaryrefslogtreecommitdiff
path: root/arch/x86/kernel/cpu/perf_event_amd.c
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/kernel/cpu/perf_event_amd.c')
-rw-r--r--arch/x86/kernel/cpu/perf_event_amd.c419
1 files changed, 282 insertions, 137 deletions
diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c
index 461f62bbd77..beeb7cc0704 100644
--- a/arch/x86/kernel/cpu/perf_event_amd.c
+++ b/arch/x86/kernel/cpu/perf_event_amd.c
@@ -1,4 +1,11 @@
-#ifdef CONFIG_CPU_SUP_AMD
+#include <linux/perf_event.h>
+#include <linux/export.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <asm/apicdef.h>
+
+#include "perf_event.h"
static __initconst const u64 amd_hw_cache_event_ids
[PERF_COUNT_HW_CACHE_MAX]
@@ -8,7 +15,7 @@ static __initconst const u64 amd_hw_cache_event_ids
[ C(L1D) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */
- [ C(RESULT_MISS) ] = 0x0041, /* Data Cache Misses */
+ [ C(RESULT_MISS) ] = 0x0141, /* Data Cache Misses */
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = 0x0142, /* Data Cache Refills :system */
@@ -89,6 +96,20 @@ static __initconst const u64 amd_hw_cache_event_ids
[ C(RESULT_MISS) ] = -1,
},
},
+ [ C(NODE) ] = {
+ [ C(OP_READ) ] = {
+ [ C(RESULT_ACCESS) ] = 0xb8e9, /* CPU Request to Memory, l+r */
+ [ C(RESULT_MISS) ] = 0x98e9, /* CPU Request to Memory, r */
+ },
+ [ C(OP_WRITE) ] = {
+ [ C(RESULT_ACCESS) ] = -1,
+ [ C(RESULT_MISS) ] = -1,
+ },
+ [ C(OP_PREFETCH) ] = {
+ [ C(RESULT_ACCESS) ] = -1,
+ [ C(RESULT_MISS) ] = -1,
+ },
+ },
};
/*
@@ -96,12 +117,14 @@ static __initconst const u64 amd_hw_cache_event_ids
*/
static const u64 amd_perfmon_event_map[] =
{
- [PERF_COUNT_HW_CPU_CYCLES] = 0x0076,
- [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0,
- [PERF_COUNT_HW_CACHE_REFERENCES] = 0x0080,
- [PERF_COUNT_HW_CACHE_MISSES] = 0x0081,
- [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c2,
- [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c3,
+ [PERF_COUNT_HW_CPU_CYCLES] = 0x0076,
+ [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0,
+ [PERF_COUNT_HW_CACHE_REFERENCES] = 0x0080,
+ [PERF_COUNT_HW_CACHE_MISSES] = 0x0081,
+ [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c2,
+ [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c3,
+ [PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = 0x00d0, /* "Decoder empty" event */
+ [PERF_COUNT_HW_STALLED_CYCLES_BACKEND] = 0x00d1, /* "Dispatch stalls" event */
};
static u64 amd_pmu_event_map(int hw_event)
@@ -109,17 +132,61 @@ static u64 amd_pmu_event_map(int hw_event)
return amd_perfmon_event_map[hw_event];
}
-static int amd_pmu_hw_config(struct perf_event *event)
+/*
+ * Previously calculated offsets
+ */
+static unsigned int event_offsets[X86_PMC_IDX_MAX] __read_mostly;
+static unsigned int count_offsets[X86_PMC_IDX_MAX] __read_mostly;
+
+/*
+ * Legacy CPUs:
+ * 4 counters starting at 0xc0010000 each offset by 1
+ *
+ * CPUs with core performance counter extensions:
+ * 6 counters starting at 0xc0010200 each offset by 2
+ */
+static inline int amd_pmu_addr_offset(int index, bool eventsel)
{
- int ret = x86_pmu_hw_config(event);
+ int offset;
- if (ret)
- return ret;
+ if (!index)
+ return index;
- if (event->attr.type != PERF_TYPE_RAW)
- return 0;
+ if (eventsel)
+ offset = event_offsets[index];
+ else
+ offset = count_offsets[index];
+
+ if (offset)
+ return offset;
+
+ if (!cpu_has_perfctr_core)
+ offset = index;
+ else
+ offset = index << 1;
+
+ if (eventsel)
+ event_offsets[index] = offset;
+ else
+ count_offsets[index] = offset;
+
+ return offset;
+}
- event->hw.config |= event->attr.config & AMD64_RAW_EVENT_MASK;
+static int amd_core_hw_config(struct perf_event *event)
+{
+ if (event->attr.exclude_host && event->attr.exclude_guest)
+ /*
+ * When HO == GO == 1 the hardware treats that as GO == HO == 0
+ * and will count in both modes. We don't want to count in that
+ * case so we emulate no-counting by setting US = OS = 0.
+ */
+ event->hw.config &= ~(ARCH_PERFMON_EVENTSEL_USR |
+ ARCH_PERFMON_EVENTSEL_OS);
+ else if (event->attr.exclude_host)
+ event->hw.config |= AMD64_EVENTSEL_GUESTONLY;
+ else if (event->attr.exclude_guest)
+ event->hw.config |= AMD64_EVENTSEL_HOSTONLY;
return 0;
}
@@ -144,20 +211,34 @@ static inline int amd_has_nb(struct cpu_hw_events *cpuc)
return nb && nb->nb_id != -1;
}
-static void amd_put_event_constraints(struct cpu_hw_events *cpuc,
- struct perf_event *event)
+static int amd_pmu_hw_config(struct perf_event *event)
+{
+ int ret;
+
+ /* pass precise event sampling to ibs: */
+ if (event->attr.precise_ip && get_ibs_caps())
+ return -ENOENT;
+
+ if (has_branch_stack(event))
+ return -EOPNOTSUPP;
+
+ ret = x86_pmu_hw_config(event);
+ if (ret)
+ return ret;
+
+ if (event->attr.type == PERF_TYPE_RAW)
+ event->hw.config |= event->attr.config & AMD64_RAW_EVENT_MASK;
+
+ return amd_core_hw_config(event);
+}
+
+static void __amd_put_nb_event_constraints(struct cpu_hw_events *cpuc,
+ struct perf_event *event)
{
- struct hw_perf_event *hwc = &event->hw;
struct amd_nb *nb = cpuc->amd_nb;
int i;
/*
- * only care about NB events
- */
- if (!(amd_has_nb(cpuc) && amd_is_nb_event(hwc)))
- return;
-
- /*
* need to scan whole list because event may not have
* been assigned during scheduling
*
@@ -166,10 +247,8 @@ static void amd_put_event_constraints(struct cpu_hw_events *cpuc,
* when we come here
*/
for (i = 0; i < x86_pmu.num_counters; i++) {
- if (nb->owners[i] == event) {
- cmpxchg(nb->owners+i, event, NULL);
+ if (cmpxchg(nb->owners + i, event, NULL) == event)
break;
- }
}
}
@@ -205,24 +284,24 @@ static void amd_put_event_constraints(struct cpu_hw_events *cpuc,
*
* Given that resources are allocated (cmpxchg), they must be
* eventually freed for others to use. This is accomplished by
- * calling amd_put_event_constraints().
+ * calling __amd_put_nb_event_constraints()
*
* Non NB events are not impacted by this restriction.
*/
static struct event_constraint *
-amd_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
+__amd_get_nb_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event,
+ struct event_constraint *c)
{
struct hw_perf_event *hwc = &event->hw;
struct amd_nb *nb = cpuc->amd_nb;
- struct perf_event *old = NULL;
- int max = x86_pmu.num_counters;
- int i, j, k = -1;
+ struct perf_event *old;
+ int idx, new = -1;
- /*
- * if not NB event or no NB, then no constraints
- */
- if (!(amd_has_nb(cpuc) && amd_is_nb_event(hwc)))
- return &unconstrained;
+ if (!c)
+ c = &unconstrained;
+
+ if (cpuc->is_fake)
+ return c;
/*
* detect if already present, if so reuse
@@ -234,48 +313,33 @@ amd_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
* because of successive calls to x86_schedule_events() from
* hw_perf_group_sched_in() without hw_perf_enable()
*/
- for (i = 0; i < max; i++) {
- /*
- * keep track of first free slot
- */
- if (k == -1 && !nb->owners[i])
- k = i;
+ for_each_set_bit(idx, c->idxmsk, x86_pmu.num_counters) {
+ if (new == -1 || hwc->idx == idx)
+ /* assign free slot, prefer hwc->idx */
+ old = cmpxchg(nb->owners + idx, NULL, event);
+ else if (nb->owners[idx] == event)
+ /* event already present */
+ old = event;
+ else
+ continue;
+
+ if (old && old != event)
+ continue;
+
+ /* reassign to this slot */
+ if (new != -1)
+ cmpxchg(nb->owners + new, event, NULL);
+ new = idx;
/* already present, reuse */
- if (nb->owners[i] == event)
- goto done;
- }
- /*
- * not present, so grab a new slot
- * starting either at:
- */
- if (hwc->idx != -1) {
- /* previous assignment */
- i = hwc->idx;
- } else if (k != -1) {
- /* start from free slot found */
- i = k;
- } else {
- /*
- * event not found, no slot found in
- * first pass, try again from the
- * beginning
- */
- i = 0;
- }
- j = i;
- do {
- old = cmpxchg(nb->owners+i, NULL, event);
- if (!old)
+ if (old == event)
break;
- if (++i == max)
- i = 0;
- } while (i != j);
-done:
- if (!old)
- return &nb->event_constraints[i];
-
- return &emptyconstraint;
+ }
+
+ if (new == -1)
+ return &emptyconstraint;
+
+ return &nb->event_constraints[new];
}
static struct amd_nb *amd_alloc_nb(int cpu)
@@ -283,8 +347,7 @@ static struct amd_nb *amd_alloc_nb(int cpu)
struct amd_nb *nb;
int i;
- nb = kmalloc_node(sizeof(struct amd_nb), GFP_KERNEL | __GFP_ZERO,
- cpu_to_node(cpu));
+ nb = kzalloc_node(sizeof(struct amd_nb), GFP_KERNEL, cpu_to_node(cpu));
if (!nb)
return NULL;
@@ -322,6 +385,8 @@ static void amd_pmu_cpu_starting(int cpu)
struct amd_nb *nb;
int i, nb_id;
+ cpuc->perf_ctr_virt_mask = AMD64_EVENTSEL_HOSTONLY;
+
if (boot_cpu_data.x86_max_cores < 2)
return;
@@ -334,7 +399,7 @@ static void amd_pmu_cpu_starting(int cpu)
continue;
if (nb->nb_id == nb_id) {
- kfree(cpuc->amd_nb);
+ cpuc->kfree_on_online = cpuc->amd_nb;
cpuc->amd_nb = nb;
break;
}
@@ -363,31 +428,38 @@ static void amd_pmu_cpu_dead(int cpu)
}
}
-static __initconst const struct x86_pmu amd_pmu = {
- .name = "AMD",
- .handle_irq = x86_pmu_handle_irq,
- .disable_all = x86_pmu_disable_all,
- .enable_all = x86_pmu_enable_all,
- .enable = x86_pmu_enable_event,
- .disable = x86_pmu_disable_event,
- .hw_config = amd_pmu_hw_config,
- .schedule_events = x86_schedule_events,
- .eventsel = MSR_K7_EVNTSEL0,
- .perfctr = MSR_K7_PERFCTR0,
- .event_map = amd_pmu_event_map,
- .max_events = ARRAY_SIZE(amd_perfmon_event_map),
- .num_counters = 4,
- .cntval_bits = 48,
- .cntval_mask = (1ULL << 48) - 1,
- .apic = 1,
- /* use highest bit to detect overflow */
- .max_period = (1ULL << 47) - 1,
- .get_event_constraints = amd_get_event_constraints,
- .put_event_constraints = amd_put_event_constraints,
+static struct event_constraint *
+amd_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
+{
+ /*
+ * if not NB event or no NB, then no constraints
+ */
+ if (!(amd_has_nb(cpuc) && amd_is_nb_event(&event->hw)))
+ return &unconstrained;
- .cpu_prepare = amd_pmu_cpu_prepare,
- .cpu_starting = amd_pmu_cpu_starting,
- .cpu_dead = amd_pmu_cpu_dead,
+ return __amd_get_nb_event_constraints(cpuc, event, NULL);
+}
+
+static void amd_put_event_constraints(struct cpu_hw_events *cpuc,
+ struct perf_event *event)
+{
+ if (amd_has_nb(cpuc) && amd_is_nb_event(&event->hw))
+ __amd_put_nb_event_constraints(cpuc, event);
+}
+
+PMU_FORMAT_ATTR(event, "config:0-7,32-35");
+PMU_FORMAT_ATTR(umask, "config:8-15" );
+PMU_FORMAT_ATTR(edge, "config:18" );
+PMU_FORMAT_ATTR(inv, "config:23" );
+PMU_FORMAT_ATTR(cmask, "config:24-31" );
+
+static struct attribute *amd_format_attr[] = {
+ &format_attr_event.attr,
+ &format_attr_umask.attr,
+ &format_attr_edge.attr,
+ &format_attr_inv.attr,
+ &format_attr_cmask.attr,
+ NULL,
};
/* AMD Family 15h */
@@ -427,12 +499,15 @@ static __initconst const struct x86_pmu amd_pmu = {
*
* Exceptions:
*
+ * 0x000 FP PERF_CTL[3], PERF_CTL[5:3] (*)
* 0x003 FP PERF_CTL[3]
+ * 0x004 FP PERF_CTL[3], PERF_CTL[5:3] (*)
* 0x00B FP PERF_CTL[3]
* 0x00D FP PERF_CTL[3]
* 0x023 DE PERF_CTL[2:0]
* 0x02D LS PERF_CTL[3]
* 0x02E LS PERF_CTL[3,0]
+ * 0x031 LS PERF_CTL[2:0] (**)
* 0x043 CU PERF_CTL[2:0]
* 0x045 CU PERF_CTL[2:0]
* 0x046 CU PERF_CTL[2:0]
@@ -446,32 +521,46 @@ static __initconst const struct x86_pmu amd_pmu = {
* 0x0DD LS PERF_CTL[5:0]
* 0x0DE LS PERF_CTL[5:0]
* 0x0DF LS PERF_CTL[5:0]
+ * 0x1C0 EX PERF_CTL[5:3]
* 0x1D6 EX PERF_CTL[5:0]
* 0x1D8 EX PERF_CTL[5:0]
+ *
+ * (*) depending on the umask all FPU counters may be used
+ * (**) only one unitmask enabled at a time
*/
static struct event_constraint amd_f15_PMC0 = EVENT_CONSTRAINT(0, 0x01, 0);
static struct event_constraint amd_f15_PMC20 = EVENT_CONSTRAINT(0, 0x07, 0);
static struct event_constraint amd_f15_PMC3 = EVENT_CONSTRAINT(0, 0x08, 0);
-static struct event_constraint amd_f15_PMC30 = EVENT_CONSTRAINT(0, 0x09, 0);
+static struct event_constraint amd_f15_PMC30 = EVENT_CONSTRAINT_OVERLAP(0, 0x09, 0);
static struct event_constraint amd_f15_PMC50 = EVENT_CONSTRAINT(0, 0x3F, 0);
static struct event_constraint amd_f15_PMC53 = EVENT_CONSTRAINT(0, 0x38, 0);
static struct event_constraint *
amd_get_event_constraints_f15h(struct cpu_hw_events *cpuc, struct perf_event *event)
{
- unsigned int event_code = amd_get_event_code(&event->hw);
+ struct hw_perf_event *hwc = &event->hw;
+ unsigned int event_code = amd_get_event_code(hwc);
switch (event_code & AMD_EVENT_TYPE_MASK) {
case AMD_EVENT_FP:
switch (event_code) {
+ case 0x000:
+ if (!(hwc->config & 0x0000F000ULL))
+ break;
+ if (!(hwc->config & 0x00000F00ULL))
+ break;
+ return &amd_f15_PMC3;
+ case 0x004:
+ if (hweight_long(hwc->config & ARCH_PERFMON_EVENTSEL_UMASK) <= 1)
+ break;
+ return &amd_f15_PMC3;
case 0x003:
case 0x00B:
case 0x00D:
return &amd_f15_PMC3;
- default:
- return &amd_f15_PMC53;
}
+ return &amd_f15_PMC53;
case AMD_EVENT_LS:
case AMD_EVENT_DC:
case AMD_EVENT_EX_LS:
@@ -487,6 +576,12 @@ amd_get_event_constraints_f15h(struct cpu_hw_events *cpuc, struct perf_event *ev
return &amd_f15_PMC3;
case 0x02E:
return &amd_f15_PMC30;
+ case 0x031:
+ if (hweight_long(hwc->config & ARCH_PERFMON_EVENTSEL_UMASK) <= 1)
+ return &amd_f15_PMC20;
+ return &emptyconstraint;
+ case 0x1C0:
+ return &amd_f15_PMC53;
default:
return &amd_f15_PMC50;
}
@@ -506,15 +601,23 @@ amd_get_event_constraints_f15h(struct cpu_hw_events *cpuc, struct perf_event *ev
return &amd_f15_PMC20;
}
case AMD_EVENT_NB:
- /* not yet implemented */
+ /* moved to perf_event_amd_uncore.c */
return &emptyconstraint;
default:
return &emptyconstraint;
}
}
-static __initconst const struct x86_pmu amd_pmu_f15h = {
- .name = "AMD Family 15h",
+static ssize_t amd_event_sysfs_show(char *page, u64 config)
+{
+ u64 event = (config & ARCH_PERFMON_EVENTSEL_EVENT) |
+ (config & AMD64_EVENTSEL_EVENT) >> 24;
+
+ return x86_event_sysfs_show(page, config, event);
+}
+
+static __initconst const struct x86_pmu amd_pmu = {
+ .name = "AMD",
.handle_irq = x86_pmu_handle_irq,
.disable_all = x86_pmu_disable_all,
.enable_all = x86_pmu_enable_all,
@@ -522,50 +625,71 @@ static __initconst const struct x86_pmu amd_pmu_f15h = {
.disable = x86_pmu_disable_event,
.hw_config = amd_pmu_hw_config,
.schedule_events = x86_schedule_events,
- .eventsel = MSR_F15H_PERF_CTL,
- .perfctr = MSR_F15H_PERF_CTR,
+ .eventsel = MSR_K7_EVNTSEL0,
+ .perfctr = MSR_K7_PERFCTR0,
+ .addr_offset = amd_pmu_addr_offset,
.event_map = amd_pmu_event_map,
.max_events = ARRAY_SIZE(amd_perfmon_event_map),
- .num_counters = 6,
+ .num_counters = AMD64_NUM_COUNTERS,
.cntval_bits = 48,
.cntval_mask = (1ULL << 48) - 1,
.apic = 1,
/* use highest bit to detect overflow */
.max_period = (1ULL << 47) - 1,
- .get_event_constraints = amd_get_event_constraints_f15h,
- /* nortbridge counters not yet implemented: */
-#if 0
+ .get_event_constraints = amd_get_event_constraints,
.put_event_constraints = amd_put_event_constraints,
+ .format_attrs = amd_format_attr,
+ .events_sysfs_show = amd_event_sysfs_show,
+
.cpu_prepare = amd_pmu_cpu_prepare,
.cpu_starting = amd_pmu_cpu_starting,
.cpu_dead = amd_pmu_cpu_dead,
-#endif
};
-static __init int amd_pmu_init(void)
+static int __init amd_core_pmu_init(void)
{
- /* Performance-monitoring supported from K7 and later: */
- if (boot_cpu_data.x86 < 6)
- return -ENODEV;
+ if (!cpu_has_perfctr_core)
+ return 0;
- /*
- * If core performance counter extensions exists, it must be
- * family 15h, otherwise fail. See x86_pmu_addr_offset().
- */
switch (boot_cpu_data.x86) {
case 0x15:
- if (!cpu_has_perfctr_core)
- return -ENODEV;
- x86_pmu = amd_pmu_f15h;
+ pr_cont("Fam15h ");
+ x86_pmu.get_event_constraints = amd_get_event_constraints_f15h;
break;
+
default:
- if (cpu_has_perfctr_core)
- return -ENODEV;
- x86_pmu = amd_pmu;
- break;
+ pr_err("core perfctr but no constraints; unknown hardware!\n");
+ return -ENODEV;
}
+ /*
+ * If core performance counter extensions exists, we must use
+ * MSR_F15H_PERF_CTL/MSR_F15H_PERF_CTR msrs. See also
+ * amd_pmu_addr_offset().
+ */
+ x86_pmu.eventsel = MSR_F15H_PERF_CTL;
+ x86_pmu.perfctr = MSR_F15H_PERF_CTR;
+ x86_pmu.num_counters = AMD64_NUM_COUNTERS_CORE;
+
+ pr_cont("core perfctr, ");
+ return 0;
+}
+
+__init int amd_pmu_init(void)
+{
+ int ret;
+
+ /* Performance-monitoring supported from K7 and later: */
+ if (boot_cpu_data.x86 < 6)
+ return -ENODEV;
+
+ x86_pmu = amd_pmu;
+
+ ret = amd_core_pmu_init();
+ if (ret)
+ return ret;
+
/* Events are common for all AMDs */
memcpy(hw_cache_event_ids, amd_hw_cache_event_ids,
sizeof(hw_cache_event_ids));
@@ -573,11 +697,32 @@ static __init int amd_pmu_init(void)
return 0;
}
-#else /* CONFIG_CPU_SUP_AMD */
-
-static int amd_pmu_init(void)
+void amd_pmu_enable_virt(void)
{
- return 0;
+ struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+ cpuc->perf_ctr_virt_mask = 0;
+
+ /* Reload all events */
+ x86_pmu_disable_all();
+ x86_pmu_enable_all(0);
}
+EXPORT_SYMBOL_GPL(amd_pmu_enable_virt);
+
+void amd_pmu_disable_virt(void)
+{
+ struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-#endif
+ /*
+ * We only mask out the Host-only bit so that host-only counting works
+ * when SVM is disabled. If someone sets up a guest-only counter when
+ * SVM is disabled the Guest-only bits still gets set and the counter
+ * will not count anything.
+ */
+ cpuc->perf_ctr_virt_mask = AMD64_EVENTSEL_HOSTONLY;
+
+ /* Reload all events */
+ x86_pmu_disable_all();
+ x86_pmu_enable_all(0);
+}
+EXPORT_SYMBOL_GPL(amd_pmu_disable_virt);