[tip:perf/core] perf/core: Introduce PMU_EV_CAP_READ_ACTIVE_PKG

From: tip-bot for David Carrillo-Cisneros
Date: Thu Aug 18 2016 - 07:46:06 EST


Commit-ID: d6a2f9035bfc27d0e9d78b13635dda9fb017ac01
Gitweb: http://git.kernel.org/tip/d6a2f9035bfc27d0e9d78b13635dda9fb017ac01
Author: David Carrillo-Cisneros <davidcc@xxxxxxxxxx>
AuthorDate: Wed, 17 Aug 2016 13:55:06 -0700
Committer: Ingo Molnar <mingo@xxxxxxxxxx>
CommitDate: Thu, 18 Aug 2016 10:53:59 +0200

perf/core: Introduce PMU_EV_CAP_READ_ACTIVE_PKG

Introduce the flag PMU_EV_CAP_READ_ACTIVE_PKG, useful for uncore events,
that allows a PMU to signal the generic perf code that an event is readable
in the current CPU if the event is active in a CPU in the same package as
the current CPU.

This is an optimization that avoids a unnecessary IPI for the common case
where uncore events are run and read in the same package but in
different CPUs.

As an example, the IPI removal speeds up perf_read() in my Haswell system
as follows:

- For event UNC_C_LLC_LOOKUP: From 260 us to 31 us.
- For event RAPL's power/energy-cores/: From to 255 us to 27 us.

For the optimization to work, all events in the group must have it
(similarly to PERF_EV_CAP_SOFTWARE).

Signed-off-by: David Carrillo-Cisneros <davidcc@xxxxxxxxxx>
Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Cc: Alexander Shishkin <alexander.shishkin@xxxxxxxxxxxxxxx>
Cc: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
Cc: David Carrillo-Cisneros <davidcc@xxxxxxxxxx>
Cc: Jiri Olsa <jolsa@xxxxxxxxxx>
Cc: Kan Liang <kan.liang@xxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Paul Turner <pjt@xxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Stephane Eranian <eranian@xxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Vegard Nossum <vegard.nossum@xxxxxxxxx>
Cc: Vince Weaver <vincent.weaver@xxxxxxxxx>
Link: http://lkml.kernel.org/r/1471467307-61171-4-git-send-email-davidcc@xxxxxxxxxx
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
---
include/linux/perf_event.h | 3 +++
kernel/events/core.c | 25 +++++++++++++++++++++++--
2 files changed, 26 insertions(+), 2 deletions(-)

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 6f7459f..5c53625 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -514,8 +514,11 @@ typedef void (*perf_overflow_handler_t)(struct perf_event *,
* Event capabilities. For event_caps and groups caps.
*
* PERF_EV_CAP_SOFTWARE: Is a software event.
+ * PERF_EV_CAP_READ_ACTIVE_PKG: A CPU event (or cgroup event) that can be read
+ * from any CPU in the package where it is active.
*/
#define PERF_EV_CAP_SOFTWARE BIT(0)
+#define PERF_EV_CAP_READ_ACTIVE_PKG BIT(1)

#define SWEVENT_HLIST_BITS 8
#define SWEVENT_HLIST_SIZE (1 << SWEVENT_HLIST_BITS)
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 8c42a5a..3f07e6c 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -3424,6 +3424,22 @@ struct perf_read_data {
int ret;
};

+static int find_cpu_to_read(struct perf_event *event, int local_cpu)
+{
+ int event_cpu = event->oncpu;
+ u16 local_pkg, event_pkg;
+
+ if (event->group_caps & PERF_EV_CAP_READ_ACTIVE_PKG) {
+ event_pkg = topology_physical_package_id(event_cpu);
+ local_pkg = topology_physical_package_id(local_cpu);
+
+ if (event_pkg == local_pkg)
+ return local_cpu;
+ }
+
+ return event_cpu;
+}
+
/*
* Cross CPU call to read the hardware event
*/
@@ -3545,7 +3561,7 @@ u64 perf_event_read_local(struct perf_event *event)

static int perf_event_read(struct perf_event *event, bool group)
{
- int ret = 0;
+ int ret = 0, cpu_to_read, local_cpu;

/*
* If event is enabled and currently active on a CPU, update the
@@ -3557,7 +3573,12 @@ static int perf_event_read(struct perf_event *event, bool group)
.group = group,
.ret = 0,
};
- ret = smp_call_function_single(event->oncpu, __perf_event_read, &data, 1);
+
+ local_cpu = get_cpu();
+ cpu_to_read = find_cpu_to_read(event, local_cpu);
+ put_cpu();
+
+ ret = smp_call_function_single(cpu_to_read, __perf_event_read, &data, 1);
/* The event must have been read from an online CPU: */
WARN_ON_ONCE(ret);
ret = ret ? : data.ret;