linux/arch/x86/kernel/cpu/perf_event_p6.c

280 lines
6.8 KiB
C
Raw Normal View History

#include <linux/perf_event.h>
#include <linux/types.h>
#include "perf_event.h"
/*
* Not sure about some of these
*/
static const u64 p6_perfmon_event_map[] =
{
[PERF_COUNT_HW_CPU_CYCLES] = 0x0079, /* CPU_CLK_UNHALTED */
[PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, /* INST_RETIRED */
[PERF_COUNT_HW_CACHE_REFERENCES] = 0x0f2e, /* L2_RQSTS:M:E:S:I */
[PERF_COUNT_HW_CACHE_MISSES] = 0x012e, /* L2_RQSTS:I */
[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c4, /* BR_INST_RETIRED */
[PERF_COUNT_HW_BRANCH_MISSES] = 0x00c5, /* BR_MISS_PRED_RETIRED */
[PERF_COUNT_HW_BUS_CYCLES] = 0x0062, /* BUS_DRDY_CLOCKS */
[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = 0x00a2, /* RESOURCE_STALLS */
};
static const u64 __initconst p6_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) ] = 0x0043, /* DATA_MEM_REFS */
[ C(RESULT_MISS) ] = 0x0045, /* DCU_LINES_IN */
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0x0f29, /* L2_LD:M:E:S:I */
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
},
[ C(L1I ) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x0080, /* IFU_IFETCH */
[ C(RESULT_MISS) ] = 0x0f28, /* L2_IFETCH:M:E:S:I */
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
},
[ C(LL ) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0x0025, /* L2_M_LINES_INM */
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
},
[ C(DTLB) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x0043, /* DATA_MEM_REFS */
[ C(RESULT_MISS) ] = 0,
},
[ 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, /* IFU_IFETCH */
[ C(RESULT_MISS) ] = 0x0085, /* ITLB_MISS */
},
[ 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) ] = 0x00c4, /* BR_INST_RETIRED */
[ C(RESULT_MISS) ] = 0x00c5, /* BR_MISS_PRED_RETIRED */
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
},
};
static u64 p6_pmu_event_map(int hw_event)
{
return p6_perfmon_event_map[hw_event];
}
/*
* Event setting that is specified not to count anything.
* We use this to effectively disable a counter.
*
* L2_RQSTS with 0 MESI unit mask.
*/
#define P6_NOP_EVENT 0x0000002EULL
static struct event_constraint p6_event_constraints[] =
{
INTEL_EVENT_CONSTRAINT(0xc1, 0x1), /* FLOPS */
INTEL_EVENT_CONSTRAINT(0x10, 0x1), /* FP_COMP_OPS_EXE */
INTEL_EVENT_CONSTRAINT(0x11, 0x2), /* FP_ASSIST */
INTEL_EVENT_CONSTRAINT(0x12, 0x2), /* MUL */
INTEL_EVENT_CONSTRAINT(0x13, 0x2), /* DIV */
INTEL_EVENT_CONSTRAINT(0x14, 0x1), /* CYCLES_DIV_BUSY */
EVENT_CONSTRAINT_END
};
static void p6_pmu_disable_all(void)
{
u64 val;
/* p6 only has one enable register */
rdmsrl(MSR_P6_EVNTSEL0, val);
val &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
wrmsrl(MSR_P6_EVNTSEL0, val);
}
static void p6_pmu_enable_all(int added)
{
unsigned long val;
/* p6 only has one enable register */
rdmsrl(MSR_P6_EVNTSEL0, val);
val |= ARCH_PERFMON_EVENTSEL_ENABLE;
wrmsrl(MSR_P6_EVNTSEL0, val);
}
static inline void
p6_pmu_disable_event(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
u64 val = P6_NOP_EVENT;
(void)wrmsrl_safe(hwc->config_base, val);
}
static void p6_pmu_enable_event(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
u64 val;
val = hwc->config;
/*
* p6 only has a global event enable, set on PerfEvtSel0
* We "disable" events by programming P6_NOP_EVENT
* and we rely on p6_pmu_enable_all() being called
* to actually enable the events.
*/
(void)wrmsrl_safe(hwc->config_base, val);
}
PMU_FORMAT_ATTR(event, "config:0-7" );
PMU_FORMAT_ATTR(umask, "config:8-15" );
PMU_FORMAT_ATTR(edge, "config:18" );
PMU_FORMAT_ATTR(pc, "config:19" );
PMU_FORMAT_ATTR(inv, "config:23" );
PMU_FORMAT_ATTR(cmask, "config:24-31" );
static struct attribute *intel_p6_formats_attr[] = {
&format_attr_event.attr,
&format_attr_umask.attr,
&format_attr_edge.attr,
&format_attr_pc.attr,
&format_attr_inv.attr,
&format_attr_cmask.attr,
NULL,
};
static __initconst const struct x86_pmu p6_pmu = {
.name = "p6",
.handle_irq = x86_pmu_handle_irq,
.disable_all = p6_pmu_disable_all,
.enable_all = p6_pmu_enable_all,
.enable = p6_pmu_enable_event,
.disable = p6_pmu_disable_event,
.hw_config = x86_pmu_hw_config,
perf, x86: Implement initial P4 PMU driver The netburst PMU is way different from the "architectural perfomance monitoring" specification that current CPUs use. P4 uses a tuple of ESCR+CCCR+COUNTER MSR registers to handle perfomance monitoring events. A few implementational details: 1) We need a separate x86_pmu::hw_config helper in struct x86_pmu since register bit-fields are quite different from P6, Core and later cpu series. 2) For the same reason is a x86_pmu::schedule_events helper introduced. 3) hw_perf_event::config consists of packed ESCR+CCCR values. It's allowed since in reality both registers only use a half of their size. Of course before making a real write into a particular MSR we need to unpack the value and extend it to a proper size. 4) The tuple of packed ESCR+CCCR in hw_perf_event::config doesn't describe the memory address of ESCR MSR register so that we need to keep a mapping between these tuples used and available ESCR (various P4 events may use same ESCRs but not simultaneously), for this sake every active event has a per-cpu map of hw_perf_event::idx <--> ESCR addresses. 5) Since hw_perf_event::idx is an offset to counter/control register we need to lift X86_PMC_MAX_GENERIC up, otherwise kernel strips it down to 8 registers and event armed may never be turned off (ie the bit in active_mask is set but the loop never reaches this index to check), thanks to Peter Zijlstra Restrictions: - No cascaded counters support (do we ever need them?) - No dependent events support (so PERF_COUNT_HW_INSTRUCTIONS doesn't work for now) - There are events with same counters which can't work simultaneously (need to use intersected ones due to broken counter 1) - No PERF_COUNT_HW_CACHE_ events yet Todo: - Implement dependent events - Need proper hashing for event opcodes (no linear search, good for debugging stage but not in real loads) - Some events counted during a clock cycle -- need to set threshold for them and count every clock cycle just to get summary statistics (ie to behave the same way as other PMUs do) - Need to swicth to use event_constraints - To support RAW events we need to encode a global list of P4 events into p4_templates - Cache events need to be added Event support status matrix: Event status ----------------------------- cycles works cache-references works cache-misses works branch-misses works bus-cycles partially (does not work on 64bit cpu with HT enabled) instruction doesnt work (needs dependent event [mop tagging]) branches doesnt work Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org> Signed-off-by: Lin Ming <ming.m.lin@intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Stephane Eranian <eranian@google.com> Cc: Robert Richter <robert.richter@amd.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> LKML-Reference: <20100311165439.GB5129@lenovo> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-03-11 17:54:39 +01:00
.schedule_events = x86_schedule_events,
.eventsel = MSR_P6_EVNTSEL0,
.perfctr = MSR_P6_PERFCTR0,
.event_map = p6_pmu_event_map,
.max_events = ARRAY_SIZE(p6_perfmon_event_map),
.apic = 1,
.max_period = (1ULL << 31) - 1,
.version = 0,
.num_counters = 2,
/*
* Events have 40 bits implemented. However they are designed such
* that bits [32-39] are sign extensions of bit 31. As such the
* effective width of a event for P6-like PMU is 32 bits only.
*
* See IA-32 Intel Architecture Software developer manual Vol 3B
*/
.cntval_bits = 32,
.cntval_mask = (1ULL << 32) - 1,
.get_event_constraints = x86_get_event_constraints,
.event_constraints = p6_event_constraints,
.format_attrs = intel_p6_formats_attr,
.events_sysfs_show = intel_event_sysfs_show,
};
static __init void p6_pmu_rdpmc_quirk(void)
{
if (boot_cpu_data.x86_mask < 9) {
/*
* PPro erratum 26; fixed in stepping 9 and above.
*/
pr_warn("Userspace RDPMC support disabled due to a CPU erratum\n");
x86_pmu.attr_rdpmc_broken = 1;
x86_pmu.attr_rdpmc = 0;
}
}
__init int p6_pmu_init(void)
{
x86_pmu = p6_pmu;
switch (boot_cpu_data.x86_model) {
case 1: /* Pentium Pro */
x86_add_quirk(p6_pmu_rdpmc_quirk);
break;
case 3: /* Pentium II - Klamath */
case 5: /* Pentium II - Deschutes */
case 6: /* Pentium II - Mendocino */
break;
case 7: /* Pentium III - Katmai */
case 8: /* Pentium III - Coppermine */
case 10: /* Pentium III Xeon */
case 11: /* Pentium III - Tualatin */
break;
case 9: /* Pentium M - Banias */
case 13: /* Pentium M - Dothan */
break;
default:
pr_cont("unsupported p6 CPU model %d ", boot_cpu_data.x86_model);
return -ENODEV;
}
memcpy(hw_cache_event_ids, p6_hw_cache_event_ids,
sizeof(hw_cache_event_ids));
return 0;
}