AuxXxilium/linux_dsm_epyc7002
1
0
Fork 0
mirror of https://github.com/AuxXxilium/linux_dsm_epyc7002.git synced 2024-12-17 17:46:43 +07:00
Code Issues Actions Packages Projects Releases Wiki Activity
bc1738f6ee
linux_dsm_epyc7002/arch/x86/kernel/cpu/perf_event_amd.c
Robert Richter bc1738f6ee perf, x86: Fix event scheduler for constraints with overlapping counters 
The current x86 event scheduler fails to resolve scheduling problems
of certain combinations of events and constraints. This happens if the
counter mask of such an event is not a subset of any other counter
mask of a constraint with an equal or higher weight, e.g. constraints
of the AMD family 15h pmu:

                        counter mask    weight

 amd_f15_PMC30          0x09            2  <--- overlapping counters
 amd_f15_PMC20          0x07            3
 amd_f15_PMC53          0x38            3

The scheduler does not find then an existing solution. Here is an
example:

 event code     counter         failure         possible solution

 0x02E          PMC[3,0]        0               3
 0x043          PMC[2:0]        1               0
 0x045          PMC[2:0]        2               1
 0x046          PMC[2:0]        FAIL            2

The event scheduler may not select the correct counter in the first
cycle because it needs to know which subsequent events will be
scheduled. It may fail to schedule the events then.

To solve this, we now save the scheduler state of events with
overlapping counter counstraints.  If we fail to schedule the events
we rollback to those states and try to use another free counter.

Constraints with overlapping counters are marked with a new introduced
overlap flag. We set the overlap flag for such constraints to give the
scheduler a hint which events to select for counter rescheduling. The
EVENT_CONSTRAINT_OVERLAP() macro can be used for this.

Care must be taken as the rescheduling algorithm is O(n!) which will
increase scheduling cycles for an over-commited system dramatically.
The number of such EVENT_CONSTRAINT_OVERLAP() macros and its counter
masks must be kept at a minimum. Thus, the current stack is limited to
2 states to limit the number of loops the algorithm takes in the worst
case.

On systems with no overlapping-counter constraints, this
implementation does not increase the loop count compared to the
previous algorithm.

V2:
* Renamed redo -> overlap.
* Reimplementation using perf scheduling helper functions.

V3:
* Added WARN_ON_ONCE() if out of save states.
* Changed function interface of perf_sched_restore_state() to use bool
  as return value.

Signed-off-by: Robert Richter <robert.richter@amd.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1321616122-1533-3-git-send-email-robert.richter@amd.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2011-12-06 08:33:56 +01:00

624 lines
16 KiB
C
Raw Blame History

#include <linux/perf_event.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]
[PERF_COUNT_HW_CACHE_OP_MAX]
[PERF_COUNT_HW_CACHE_RESULT_MAX] =
{
[ C(L1D) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */
[ C(RESULT_MISS) ] = 0x0141, /* Data Cache Misses */
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = 0x0142, /* Data Cache Refills :system */
[ C(RESULT_MISS) ] = 0,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = 0x0267, /* Data Prefetcher :attempts */
[ C(RESULT_MISS) ] = 0x0167, /* Data Prefetcher :cancelled */
},
},
[ C(L1I ) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x0080, /* Instruction cache fetches */
[ C(RESULT_MISS) ] = 0x0081, /* Instruction cache misses */
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = 0x014B, /* Prefetch Instructions :Load */
[ C(RESULT_MISS) ] = 0,
},
},
[ C(LL ) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x037D, /* Requests to L2 Cache :IC+DC */
[ C(RESULT_MISS) ] = 0x037E, /* L2 Cache Misses : IC+DC */
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = 0x017F, /* L2 Fill/Writeback */
[ C(RESULT_MISS) ] = 0,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
},
[ C(DTLB) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */
[ C(RESULT_MISS) ] = 0x0746, /* L1_DTLB_AND_L2_DLTB_MISS.ALL */
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
},
[ C(ITLB) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x0080, /* Instruction fecthes */
[ C(RESULT_MISS) ] = 0x0385, /* L1_ITLB_AND_L2_ITLB_MISS.ALL */
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
},
[ C(BPU ) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x00c2, /* Retired Branch Instr. */
[ C(RESULT_MISS) ] = 0x00c3, /* Retired Mispredicted BI */
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ 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,
},
},
};
/*
* AMD Performance Monitor K7 and later.
*/
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_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)
{
return amd_perfmon_event_map[hw_event];
}
static int amd_pmu_hw_config(struct perf_event *event)
{
int ret = x86_pmu_hw_config(event);
if (ret)
return ret;
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 |= AMD_PERFMON_EVENTSEL_GUESTONLY;
else if (event->attr.exclude_guest)
event->hw.config |= AMD_PERFMON_EVENTSEL_HOSTONLY;
if (event->attr.type != PERF_TYPE_RAW)
return 0;
event->hw.config |= event->attr.config & AMD64_RAW_EVENT_MASK;
return 0;
}
/*
* AMD64 events are detected based on their event codes.
*/
static inline unsigned int amd_get_event_code(struct hw_perf_event *hwc)
{
return ((hwc->config >> 24) & 0x0f00) | (hwc->config & 0x00ff);
}
static inline int amd_is_nb_event(struct hw_perf_event *hwc)
{
return (hwc->config & 0xe0) == 0xe0;
}
static inline int amd_has_nb(struct cpu_hw_events *cpuc)
{
struct amd_nb *nb = cpuc->amd_nb;
return nb && nb->nb_id != -1;
}
static void amd_put_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
*
* no race condition possible because event can only
* be removed on one CPU at a time AND PMU is disabled
* when we come here
*/
for (i = 0; i < x86_pmu.num_counters; i++) {
if (nb->owners[i] == event) {
cmpxchg(nb->owners+i, event, NULL);
break;
}
}
}
/*
* AMD64 NorthBridge events need special treatment because
* counter access needs to be synchronized across all cores
* of a package. Refer to BKDG section 3.12
*
* NB events are events measuring L3 cache, Hypertransport
* traffic. They are identified by an event code >= 0xe00.
* They measure events on the NorthBride which is shared
* by all cores on a package. NB events are counted on a
* shared set of counters. When a NB event is programmed
* in a counter, the data actually comes from a shared
* counter. Thus, access to those counters needs to be
* synchronized.
*
* We implement the synchronization such that no two cores
* can be measuring NB events using the same counters. Thus,
* we maintain a per-NB allocation table. The available slot
* is propagated using the event_constraint structure.
*
* We provide only one choice for each NB event based on
* the fact that only NB events have restrictions. Consequently,
* if a counter is available, there is a guarantee the NB event
* will be assigned to it. If no slot is available, an empty
* constraint is returned and scheduling will eventually fail
* for this event.
*
* Note that all cores attached the same NB compete for the same
* counters to host NB events, this is why we use atomic ops. Some
* multi-chip CPUs may have more than one NB.
*
* Given that resources are allocated (cmpxchg), they must be
* eventually freed for others to use. This is accomplished by
* calling amd_put_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)
{
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;
/*
* if not NB event or no NB, then no constraints
*/
if (!(amd_has_nb(cpuc) && amd_is_nb_event(hwc)))
return &unconstrained;
/*
* detect if already present, if so reuse
*
* cannot merge with actual allocation
* because of possible holes
*
* event can already be present yet not assigned (in hwc->idx)
* 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;
/* 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)
break;
if (++i == max)
i = 0;
} while (i != j);
done:
if (!old)
return &nb->event_constraints[i];
return &emptyconstraint;
}
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));
if (!nb)
return NULL;
nb->nb_id = -1;
/*
* initialize all possible NB constraints
*/
for (i = 0; i < x86_pmu.num_counters; i++) {
__set_bit(i, nb->event_constraints[i].idxmsk);
nb->event_constraints[i].weight = 1;
}
return nb;
}
static int amd_pmu_cpu_prepare(int cpu)
{
struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
WARN_ON_ONCE(cpuc->amd_nb);
if (boot_cpu_data.x86_max_cores < 2)
return NOTIFY_OK;
cpuc->amd_nb = amd_alloc_nb(cpu);
if (!cpuc->amd_nb)
return NOTIFY_BAD;
return NOTIFY_OK;
}
static void amd_pmu_cpu_starting(int cpu)
{
struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
struct amd_nb *nb;
int i, nb_id;
if (boot_cpu_data.x86_max_cores < 2)
return;
nb_id = amd_get_nb_id(cpu);
WARN_ON_ONCE(nb_id == BAD_APICID);
for_each_online_cpu(i) {
nb = per_cpu(cpu_hw_events, i).amd_nb;
if (WARN_ON_ONCE(!nb))
continue;
if (nb->nb_id == nb_id) {
cpuc->kfree_on_online = cpuc->amd_nb;
cpuc->amd_nb = nb;
break;
}
}
cpuc->amd_nb->nb_id = nb_id;
cpuc->amd_nb->refcnt++;
}
static void amd_pmu_cpu_dead(int cpu)
{
struct cpu_hw_events *cpuhw;
if (boot_cpu_data.x86_max_cores < 2)
return;
cpuhw = &per_cpu(cpu_hw_events, cpu);
if (cpuhw->amd_nb) {
struct amd_nb *nb = cpuhw->amd_nb;
if (nb->nb_id == -1 || --nb->refcnt == 0)
kfree(nb);
cpuhw->amd_nb = NULL;
}
}
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 = 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,
.put_event_constraints = amd_put_event_constraints,
.cpu_prepare = amd_pmu_cpu_prepare,
.cpu_starting = amd_pmu_cpu_starting,
.cpu_dead = amd_pmu_cpu_dead,
};
/* AMD Family 15h */
#define AMD_EVENT_TYPE_MASK 0x000000F0ULL
#define AMD_EVENT_FP 0x00000000ULL ... 0x00000010ULL
#define AMD_EVENT_LS 0x00000020ULL ... 0x00000030ULL
#define AMD_EVENT_DC 0x00000040ULL ... 0x00000050ULL
#define AMD_EVENT_CU 0x00000060ULL ... 0x00000070ULL
#define AMD_EVENT_IC_DE 0x00000080ULL ... 0x00000090ULL
#define AMD_EVENT_EX_LS 0x000000C0ULL
#define AMD_EVENT_DE 0x000000D0ULL
#define AMD_EVENT_NB 0x000000E0ULL ... 0x000000F0ULL
/*
* AMD family 15h event code/PMC mappings:
*
* type = event_code & 0x0F0:
*
* 0x000 FP PERF_CTL[5:3]
* 0x010 FP PERF_CTL[5:3]
* 0x020 LS PERF_CTL[5:0]
* 0x030 LS PERF_CTL[5:0]
* 0x040 DC PERF_CTL[5:0]
* 0x050 DC PERF_CTL[5:0]
* 0x060 CU PERF_CTL[2:0]
* 0x070 CU PERF_CTL[2:0]
* 0x080 IC/DE PERF_CTL[2:0]
* 0x090 IC/DE PERF_CTL[2:0]
* 0x0A0 ---
* 0x0B0 ---
* 0x0C0 EX/LS PERF_CTL[5:0]
* 0x0D0 DE PERF_CTL[2:0]
* 0x0E0 NB NB_PERF_CTL[3:0]
* 0x0F0 NB NB_PERF_CTL[3:0]
*
* 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]
* 0x043 CU PERF_CTL[2:0]
* 0x045 CU PERF_CTL[2:0]
* 0x046 CU PERF_CTL[2:0]
* 0x054 CU PERF_CTL[2:0]
* 0x055 CU PERF_CTL[2:0]
* 0x08F IC PERF_CTL[0]
* 0x187 DE PERF_CTL[0]
* 0x188 DE PERF_CTL[0]
* 0x0DB EX PERF_CTL[5:0]
* 0x0DC LS PERF_CTL[5:0]
* 0x0DD LS PERF_CTL[5:0]
* 0x0DE LS PERF_CTL[5:0]
* 0x0DF LS PERF_CTL[5:0]
* 0x1D6 EX PERF_CTL[5:0]
* 0x1D8 EX PERF_CTL[5:0]
*
* (*) depending on the umask all FPU counters may be used
*/
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_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)
{
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;
}
return &amd_f15_PMC53;
case AMD_EVENT_LS:
case AMD_EVENT_DC:
case AMD_EVENT_EX_LS:
switch (event_code) {
case 0x023:
case 0x043:
case 0x045:
case 0x046:
case 0x054:
case 0x055:
return &amd_f15_PMC20;
case 0x02D:
return &amd_f15_PMC3;
case 0x02E:
return &amd_f15_PMC30;
default:
return &amd_f15_PMC50;
}
case AMD_EVENT_CU:
case AMD_EVENT_IC_DE:
case AMD_EVENT_DE:
switch (event_code) {
case 0x08F:
case 0x187:
case 0x188:
return &amd_f15_PMC0;
case 0x0DB ... 0x0DF:
case 0x1D6:
case 0x1D8:
return &amd_f15_PMC50;
default:
return &amd_f15_PMC20;
}
case AMD_EVENT_NB:
/* not yet implemented */
return &emptyconstraint;
default:
return &emptyconstraint;
}
}
static __initconst const struct x86_pmu amd_pmu_f15h = {
.name = "AMD Family 15h",
.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_F15H_PERF_CTL,
.perfctr = MSR_F15H_PERF_CTR,
.event_map = amd_pmu_event_map,
.max_events = ARRAY_SIZE(amd_perfmon_event_map),
.num_counters = AMD64_NUM_COUNTERS_F15H,
.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
.put_event_constraints = amd_put_event_constraints,
.cpu_prepare = amd_pmu_cpu_prepare,
.cpu_starting = amd_pmu_cpu_starting,
.cpu_dead = amd_pmu_cpu_dead,
#endif
};
__init int amd_pmu_init(void)
{
/* Performance-monitoring supported from K7 and later: */
if (boot_cpu_data.x86 < 6)
return -ENODEV;
/*
* 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;
break;
default:
if (cpu_has_perfctr_core)
return -ENODEV;
x86_pmu = amd_pmu;
break;
}
/* Events are common for all AMDs */
memcpy(hw_cache_event_ids, amd_hw_cache_event_ids,
sizeof(hw_cache_event_ids));
return 0;
}
Reference in New Issue View Git Blame Copy Permalink