提交 93fcf72c 编写于 作者: M Maria Dimakopoulou 提交者: Ingo Molnar

perf/x86/intel: Enforce HT bug workaround for SNB/IVB/HSW

This patches activates the HT bug workaround for the
SNB/IVB/HSW processors. This covers non-PEBS mode.
Activation is done thru the constraint tables.

Both client and server processors needs this workaround.
Signed-off-by: NMaria Dimakopoulou <maria.n.dimakopoulou@gmail.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NStephane Eranian <eranian@google.com>
Cc: bp@alien8.de
Cc: jolsa@redhat.com
Cc: kan.liang@intel.com
Link: http://lkml.kernel.org/r/1416251225-17721-8-git-send-email-eranian@google.comSigned-off-by: NIngo Molnar <mingo@kernel.org>
上级 e979121b
...@@ -113,6 +113,12 @@ static struct event_constraint intel_snb_event_constraints[] __read_mostly = ...@@ -113,6 +113,12 @@ static struct event_constraint intel_snb_event_constraints[] __read_mostly =
INTEL_EVENT_CONSTRAINT(0xcd, 0x8), /* MEM_TRANS_RETIRED.LOAD_LATENCY */ INTEL_EVENT_CONSTRAINT(0xcd, 0x8), /* MEM_TRANS_RETIRED.LOAD_LATENCY */
INTEL_UEVENT_CONSTRAINT(0x04a3, 0xf), /* CYCLE_ACTIVITY.CYCLES_NO_DISPATCH */ INTEL_UEVENT_CONSTRAINT(0x04a3, 0xf), /* CYCLE_ACTIVITY.CYCLES_NO_DISPATCH */
INTEL_UEVENT_CONSTRAINT(0x02a3, 0x4), /* CYCLE_ACTIVITY.CYCLES_L1D_PENDING */ INTEL_UEVENT_CONSTRAINT(0x02a3, 0x4), /* CYCLE_ACTIVITY.CYCLES_L1D_PENDING */
INTEL_EXCLEVT_CONSTRAINT(0xd0, 0xf), /* MEM_UOPS_RETIRED.* */
INTEL_EXCLEVT_CONSTRAINT(0xd1, 0xf), /* MEM_LOAD_UOPS_RETIRED.* */
INTEL_EXCLEVT_CONSTRAINT(0xd2, 0xf), /* MEM_LOAD_UOPS_LLC_HIT_RETIRED.* */
INTEL_EXCLEVT_CONSTRAINT(0xd3, 0xf), /* MEM_LOAD_UOPS_LLC_MISS_RETIRED.* */
EVENT_CONSTRAINT_END EVENT_CONSTRAINT_END
}; };
...@@ -131,15 +137,12 @@ static struct event_constraint intel_ivb_event_constraints[] __read_mostly = ...@@ -131,15 +137,12 @@ static struct event_constraint intel_ivb_event_constraints[] __read_mostly =
INTEL_UEVENT_CONSTRAINT(0x08a3, 0x4), /* CYCLE_ACTIVITY.CYCLES_L1D_PENDING */ INTEL_UEVENT_CONSTRAINT(0x08a3, 0x4), /* CYCLE_ACTIVITY.CYCLES_L1D_PENDING */
INTEL_UEVENT_CONSTRAINT(0x0ca3, 0x4), /* CYCLE_ACTIVITY.STALLS_L1D_PENDING */ INTEL_UEVENT_CONSTRAINT(0x0ca3, 0x4), /* CYCLE_ACTIVITY.STALLS_L1D_PENDING */
INTEL_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PREC_DIST */ INTEL_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PREC_DIST */
/*
* Errata BV98 -- MEM_*_RETIRED events can leak between counters of SMT INTEL_EXCLEVT_CONSTRAINT(0xd0, 0xf), /* MEM_UOPS_RETIRED.* */
* siblings; disable these events because they can corrupt unrelated INTEL_EXCLEVT_CONSTRAINT(0xd1, 0xf), /* MEM_LOAD_UOPS_RETIRED.* */
* counters. INTEL_EXCLEVT_CONSTRAINT(0xd2, 0xf), /* MEM_LOAD_UOPS_LLC_HIT_RETIRED.* */
*/ INTEL_EXCLEVT_CONSTRAINT(0xd3, 0xf), /* MEM_LOAD_UOPS_LLC_MISS_RETIRED.* */
INTEL_EVENT_CONSTRAINT(0xd0, 0x0), /* MEM_UOPS_RETIRED.* */
INTEL_EVENT_CONSTRAINT(0xd1, 0x0), /* MEM_LOAD_UOPS_RETIRED.* */
INTEL_EVENT_CONSTRAINT(0xd2, 0x0), /* MEM_LOAD_UOPS_LLC_HIT_RETIRED.* */
INTEL_EVENT_CONSTRAINT(0xd3, 0x0), /* MEM_LOAD_UOPS_LLC_MISS_RETIRED.* */
EVENT_CONSTRAINT_END EVENT_CONSTRAINT_END
}; };
...@@ -217,6 +220,12 @@ static struct event_constraint intel_hsw_event_constraints[] = { ...@@ -217,6 +220,12 @@ static struct event_constraint intel_hsw_event_constraints[] = {
INTEL_UEVENT_CONSTRAINT(0x0ca3, 0x4), INTEL_UEVENT_CONSTRAINT(0x0ca3, 0x4),
/* CYCLE_ACTIVITY.CYCLES_NO_EXECUTE */ /* CYCLE_ACTIVITY.CYCLES_NO_EXECUTE */
INTEL_UEVENT_CONSTRAINT(0x04a3, 0xf), INTEL_UEVENT_CONSTRAINT(0x04a3, 0xf),
INTEL_EXCLEVT_CONSTRAINT(0xd0, 0xf), /* MEM_UOPS_RETIRED.* */
INTEL_EXCLEVT_CONSTRAINT(0xd1, 0xf), /* MEM_LOAD_UOPS_RETIRED.* */
INTEL_EXCLEVT_CONSTRAINT(0xd2, 0xf), /* MEM_LOAD_UOPS_LLC_HIT_RETIRED.* */
INTEL_EXCLEVT_CONSTRAINT(0xd3, 0xf), /* MEM_LOAD_UOPS_LLC_MISS_RETIRED.* */
EVENT_CONSTRAINT_END EVENT_CONSTRAINT_END
}; };
...@@ -2865,6 +2874,27 @@ static __init void intel_nehalem_quirk(void) ...@@ -2865,6 +2874,27 @@ static __init void intel_nehalem_quirk(void)
} }
} }
/*
* enable software workaround for errata:
* SNB: BJ122
* IVB: BV98
* HSW: HSD29
*
* Only needed when HT is enabled. However detecting
* this is too difficult and model specific so we enable
* it even with HT off for now.
*/
static __init void intel_ht_bug(void)
{
x86_pmu.flags |= PMU_FL_EXCL_CNTRS;
x86_pmu.commit_scheduling = intel_commit_scheduling;
x86_pmu.start_scheduling = intel_start_scheduling;
x86_pmu.stop_scheduling = intel_stop_scheduling;
pr_info("CPU erratum BJ122, BV98, HSD29 worked around\n");
}
EVENT_ATTR_STR(mem-loads, mem_ld_hsw, "event=0xcd,umask=0x1,ldlat=3"); EVENT_ATTR_STR(mem-loads, mem_ld_hsw, "event=0xcd,umask=0x1,ldlat=3");
EVENT_ATTR_STR(mem-stores, mem_st_hsw, "event=0xd0,umask=0x82") EVENT_ATTR_STR(mem-stores, mem_st_hsw, "event=0xd0,umask=0x82")
...@@ -3079,6 +3109,7 @@ __init int intel_pmu_init(void) ...@@ -3079,6 +3109,7 @@ __init int intel_pmu_init(void)
case 42: /* 32nm SandyBridge */ case 42: /* 32nm SandyBridge */
case 45: /* 32nm SandyBridge-E/EN/EP */ case 45: /* 32nm SandyBridge-E/EN/EP */
x86_add_quirk(intel_sandybridge_quirk); x86_add_quirk(intel_sandybridge_quirk);
x86_add_quirk(intel_ht_bug);
memcpy(hw_cache_event_ids, snb_hw_cache_event_ids, memcpy(hw_cache_event_ids, snb_hw_cache_event_ids,
sizeof(hw_cache_event_ids)); sizeof(hw_cache_event_ids));
memcpy(hw_cache_extra_regs, snb_hw_cache_extra_regs, memcpy(hw_cache_extra_regs, snb_hw_cache_extra_regs,
...@@ -3093,6 +3124,8 @@ __init int intel_pmu_init(void) ...@@ -3093,6 +3124,8 @@ __init int intel_pmu_init(void)
x86_pmu.extra_regs = intel_snbep_extra_regs; x86_pmu.extra_regs = intel_snbep_extra_regs;
else else
x86_pmu.extra_regs = intel_snb_extra_regs; x86_pmu.extra_regs = intel_snb_extra_regs;
/* all extra regs are per-cpu when HT is on */ /* all extra regs are per-cpu when HT is on */
x86_pmu.flags |= PMU_FL_HAS_RSP_1; x86_pmu.flags |= PMU_FL_HAS_RSP_1;
x86_pmu.flags |= PMU_FL_NO_HT_SHARING; x86_pmu.flags |= PMU_FL_NO_HT_SHARING;
...@@ -3111,6 +3144,7 @@ __init int intel_pmu_init(void) ...@@ -3111,6 +3144,7 @@ __init int intel_pmu_init(void)
case 58: /* 22nm IvyBridge */ case 58: /* 22nm IvyBridge */
case 62: /* 22nm IvyBridge-EP/EX */ case 62: /* 22nm IvyBridge-EP/EX */
x86_add_quirk(intel_ht_bug);
memcpy(hw_cache_event_ids, snb_hw_cache_event_ids, memcpy(hw_cache_event_ids, snb_hw_cache_event_ids,
sizeof(hw_cache_event_ids)); sizeof(hw_cache_event_ids));
/* dTLB-load-misses on IVB is different than SNB */ /* dTLB-load-misses on IVB is different than SNB */
...@@ -3146,6 +3180,7 @@ __init int intel_pmu_init(void) ...@@ -3146,6 +3180,7 @@ __init int intel_pmu_init(void)
case 63: /* 22nm Haswell Server */ case 63: /* 22nm Haswell Server */
case 69: /* 22nm Haswell ULT */ case 69: /* 22nm Haswell ULT */
case 70: /* 22nm Haswell + GT3e (Intel Iris Pro graphics) */ case 70: /* 22nm Haswell + GT3e (Intel Iris Pro graphics) */
x86_add_quirk(intel_ht_bug);
x86_pmu.late_ack = true; x86_pmu.late_ack = true;
memcpy(hw_cache_event_ids, hsw_hw_cache_event_ids, sizeof(hw_cache_event_ids)); memcpy(hw_cache_event_ids, hsw_hw_cache_event_ids, sizeof(hw_cache_event_ids));
memcpy(hw_cache_extra_regs, hsw_hw_cache_extra_regs, sizeof(hw_cache_extra_regs)); memcpy(hw_cache_extra_regs, hsw_hw_cache_extra_regs, sizeof(hw_cache_extra_regs));
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册