lkml.org 
[lkml]   [2019]   [Oct]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Subject[PATCH v3 4/4] perf/core,x86: synchronize PMU task contexts on optimized context switches
From
Date

Install Intel specific PMU task context synchronization adapter and
extend optimized context switch path with PMU specific task context
synchronization to fix LBR callstack virtualization on context switches.

Signed-off-by: Alexey Budankov <alexey.budankov@linux.intel.com>
---
arch/x86/events/intel/core.c | 7 +++++++
kernel/events/core.c | 9 +++++++++
2 files changed, 16 insertions(+)

diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
index 43c966d1208e..7cfa658cce4b 100644
--- a/arch/x86/events/intel/core.c
+++ b/arch/x86/events/intel/core.c
@@ -3819,6 +3819,12 @@ static void intel_pmu_sched_task(struct perf_event_context *ctx,
intel_pmu_lbr_sched_task(ctx, sched_in);
}

+static void intel_pmu_sync_task_ctx(struct x86_perf_task_context *one,
+ struct x86_perf_task_context *another)
+{
+ intel_pmu_lbr_sync_task_ctx(one, another);
+}
+
static int intel_pmu_check_period(struct perf_event *event, u64 value)
{
return intel_pmu_has_bts_period(event, value) ? -EINVAL : 0;
@@ -3954,6 +3960,7 @@ static __initconst const struct x86_pmu intel_pmu = {

.guest_get_msrs = intel_guest_get_msrs,
.sched_task = intel_pmu_sched_task,
+ .sync_task_ctx = intel_pmu_sync_task_ctx,

.check_period = intel_pmu_check_period,

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 2aad959e6def..3c7edd8454ef 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -3204,11 +3204,20 @@ static void perf_event_context_sched_out(struct task_struct *task, int ctxn,
raw_spin_lock(&ctx->lock);
raw_spin_lock_nested(&next_ctx->lock, SINGLE_DEPTH_NESTING);
if (context_equiv(ctx, next_ctx)) {
+ struct pmu *pmu = ctx->pmu;
+
WRITE_ONCE(ctx->task, next);
WRITE_ONCE(next_ctx->task, task);

swap(ctx->task_ctx_data, next_ctx->task_ctx_data);

+ /*
+ * PMU specific parts of task perf context may require
+ * additional synchronization, at least for proper Intel
+ * LBR callstack data profiling;
+ */
+ pmu->sync_task_ctx(ctx->task_ctx_data,
+ next_ctx->task_ctx_data);
/*
* RCU_INIT_POINTER here is safe because we've not
* modified the ctx and the above modification of
--
2.20.1
\
 
 \ /
  Last update: 2019-10-18 11:45    [W:2.456 / U:0.552 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site