lkml.org 
[lkml]   [2011]   [May]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
SubjectRe: [PATCH 0/3] perf_events: update extra shared registers management (v2)
From
Date

How about something like the below on top of your patches?

---
Subject: perf, intel: Try alternative OFFCORE encoding
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon May 23 11:08:15 CEST 2011

Since the OFFCORE registers are fully symmetric, try the other when the
speficied one is already taken.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
arch/x86/kernel/cpu/perf_event.c | 5 ++-
arch/x86/kernel/cpu/perf_event_intel.c | 45 ++++++++++++++++++++++++++-------
2 files changed, 40 insertions(+), 10 deletions(-)

Index: linux-2.6/arch/x86/kernel/cpu/perf_event.c
===================================================================
--- linux-2.6.orig/arch/x86/kernel/cpu/perf_event.c
+++ linux-2.6/arch/x86/kernel/cpu/perf_event.c
@@ -326,9 +326,12 @@ struct x86_pmu {
* Extra registers for events
*/
struct extra_reg *extra_regs;
- bool regs_no_ht_sharing;
+ unsigned int er_flags;
};

+#define ERF_NO_HT_SHARING 1
+#define ERF_HAS_RSP_1 2
+
static struct x86_pmu x86_pmu __read_mostly;

static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
Index: linux-2.6/arch/x86/kernel/cpu/perf_event_intel.c
===================================================================
--- linux-2.6.orig/arch/x86/kernel/cpu/perf_event_intel.c
+++ linux-2.6/arch/x86/kernel/cpu/perf_event_intel.c
@@ -1019,6 +1019,27 @@ intel_bts_constraints(struct perf_event
return NULL;
}

+static bool intel_try_alt_er(struct perf_event *event, int idx)
+{
+ if (!(x86_pmu.er_flags & ERF_HAS_RSP_1))
+ return false;
+
+ if (event->hw.extra_reg.idx == EXTRA_REG_RSP_0) {
+ event->attr.config = 0x01bb;
+ event->hw.extra_reg.idx = EXTRA_REG_RSP_1;
+ event->hw.extra_reg.msr = MSR_OFFCORE_RSP_1;
+ } else if (event->hw.extra_reg.idx == EXTRA_REG_RSP_1) {
+ event->attr.config = 0x01b7;
+ event->hw.extra_reg.idx = EXTRA_REG_RSP_0;
+ event->hw.extra_reg.msr = MSR_OFFCORE_RSP_0;
+ }
+
+ if (event->hw.extra_reg.idx == idx)
+ return false;
+
+ return true;
+}
+
/*
* manage allocation of shared extra msr for certain events
*
@@ -1028,19 +1049,21 @@ intel_bts_constraints(struct perf_event
*/
static struct event_constraint *
__intel_shared_reg_get_constraints(struct cpu_hw_events *cpuc,
- struct hw_perf_event_extra *reg)
+ struct perf_event *event)
{
struct event_constraint *c = &emptyconstraint;
+ struct hw_perf_event_extra *reg = &event->hw.extra_reg;
struct er_account *era;
+ int idx = reg->idx;

/* already allocated shared msr */
if (reg->alloc)
return &unconstrained;

+again:
era = &cpuc->shared_regs->regs[reg->idx];

raw_spin_lock(&era->lock);
-
if (!atomic_read(&era->ref) || era->config == reg->config) {

/* lock in msr value */
@@ -1062,6 +1085,9 @@ __intel_shared_reg_get_constraints(struc
* the regular event constraint table.
*/
c = &unconstrained;
+ } else if (intel_try_alt_er(event, idx)) {
+ raw_spin_unlock(&era->lock);
+ goto again;
}
raw_spin_unlock(&era->lock);

@@ -1096,13 +1122,12 @@ intel_shared_regs_constraints(struct cpu
struct perf_event *event)
{
struct event_constraint *c = NULL;
- struct hw_perf_event_extra *xreg;

- xreg = &event->hw.extra_reg;
- if (xreg->idx != EXTRA_REG_NONE)
- c = __intel_shared_reg_get_constraints(cpuc, xreg);
+ if (event->hw.extra_reg.idx != EXTRA_REG_NONE)
+ c = __intel_shared_reg_get_contraints(cpuc, event);
+
return c;
- }
+}

static struct event_constraint *
intel_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
@@ -1261,7 +1286,7 @@ static void intel_pmu_cpu_starting(int c
*/
intel_pmu_lbr_reset();

- if (!cpuc->shared_regs || x86_pmu.regs_no_ht_sharing)
+ if (!cpuc->shared_regs || (x86_pmu.er_flags & ERF_NO_HT_SHARING))
return;

for_each_cpu(i, topology_thread_cpumask(cpu)) {
@@ -1486,6 +1511,7 @@ static __init int intel_pmu_init(void)
x86_pmu.enable_all = intel_pmu_nhm_enable_all;
x86_pmu.pebs_constraints = intel_westmere_pebs_event_constraints;
x86_pmu.extra_regs = intel_westmere_extra_regs;
+ x86_pmu.er_flags |= ERF_HAS_RSP_1;

/* UOPS_ISSUED.STALLED_CYCLES */
intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = 0x180010e;
@@ -1505,7 +1531,8 @@ static __init int intel_pmu_init(void)
x86_pmu.pebs_constraints = intel_snb_pebs_events;
x86_pmu.extra_regs = intel_snb_extra_regs;
/* all extra regs are per-cpu when HT is on */
- x86_pmu.regs_no_ht_sharing = true;
+ x86_pmu.er_flags |= ERF_HAS_RSP_1;
+ x86_pmu.er_flags |= ERF_NO_HT_SHARING;

/* UOPS_ISSUED.ANY,c=1,i=1 to count stall cycles */
intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = 0x180010e;


\
 
 \ /
  Last update: 2011-05-23 11:15    [W:0.084 / U:1.476 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site