perf/core: Generalize event->group_flags

Currently, PERF_GROUP_SOFTWARE is used in the group_flags field of a
group's leader to indicate that is_software_event(event) is true for all
events in a group. This is the only usage of event->group_flags.

This pattern of setting a group level flags when all events in the group
share a property is useful for the flag introduced in the next patch and
for future CQM/CMT flags. So this patches generalizes group_flags to work
as an aggregate of event level flags.

PERF_GROUP_SOFTWARE denotes an inmutable event's property. All other flags
that I intend to add are also determinable at event initialization.
To better convey the above, this patch renames event's group_flags to
group_caps and PERF_GROUP_SOFTWARE to PERF_EV_CAP_SOFTWARE.

Individual event flags are stored in the new event->event_caps. Since the
cap flags do not change after event initialization, there is no need to
serialize event_caps. This new field is used when events are added to a
context, similarly to how PERF_GROUP_SOFTWARE and is_software_event()
worked.

Lastly, for consistency, updates is_software_event() to rely in event_cap
instead of the context index.

Signed-off-by: David Carrillo-Cisneros <davidcc@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul Turner <pjt@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Link: http://lkml.kernel.org/r/1471467307-61171-3-git-send-email-davidcc@google.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 529c41f..6f7459f 100644 (file)
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -510,9 +510,12 @@ typedef void (*perf_overflow_handler_t)(struct perf_event *,
                                        struct perf_sample_data *,
                                        struct pt_regs *regs);
 
-enum perf_group_flag {
-       PERF_GROUP_SOFTWARE             = 0x1,
-};
+/*
+ * Event capabilities. For event_caps and groups caps.
+ *
+ * PERF_EV_CAP_SOFTWARE: Is a software event.
+ */
+#define PERF_EV_CAP_SOFTWARE           BIT(0)
 
 #define SWEVENT_HLIST_BITS             8
 #define SWEVENT_HLIST_SIZE             (1 << SWEVENT_HLIST_BITS)
@@ -568,7 +571,12 @@ struct perf_event {
        struct hlist_node               hlist_entry;
        struct list_head                active_entry;
        int                             nr_siblings;
-       int                             group_flags;
+
+       /* Not serialized. Only written during event initialization. */
+       int                             event_caps;
+       /* The cumulative AND of all event_caps for events in this group. */
+       int                             group_caps;
+
        struct perf_event               *group_leader;
        struct pmu                      *pmu;
        void                            *pmu_private;
@@ -988,7 +996,7 @@ static inline bool is_sampling_event(struct perf_event *event)
  */
 static inline int is_software_event(struct perf_event *event)
 {
-       return event->pmu->task_ctx_nr == perf_sw_context;
+       return event->event_caps & PERF_EV_CAP_SOFTWARE;
 }
 
 extern struct static_key perf_swevent_enabled[PERF_COUNT_SW_MAX];

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 849919c..8c42a5a 100644 (file)
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -1475,8 +1475,7 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx)
        if (event->group_leader == event) {
                struct list_head *list;
 
-               if (is_software_event(event))
-                       event->group_flags |= PERF_GROUP_SOFTWARE;
+               event->group_caps = event->event_caps;
 
                list = ctx_group_list(event, ctx);
                list_add_tail(&event->group_entry, list);
@@ -1630,9 +1629,7 @@ static void perf_group_attach(struct perf_event *event)
 
        WARN_ON_ONCE(group_leader->ctx != event->ctx);
 
-       if (group_leader->group_flags & PERF_GROUP_SOFTWARE &&
-                       !is_software_event(event))
-               group_leader->group_flags &= ~PERF_GROUP_SOFTWARE;
+       group_leader->group_caps &= event->event_caps;
 
        list_add_tail(&event->group_entry, &group_leader->sibling_list);
        group_leader->nr_siblings++;
@@ -1723,7 +1720,7 @@ static void perf_group_detach(struct perf_event *event)
                sibling->group_leader = sibling;
 
                /* Inherit group flags from the previous leader */
-               sibling->group_flags = event->group_flags;
+               sibling->group_caps = event->group_caps;
 
                WARN_ON_ONCE(sibling->ctx != event->ctx);
        }
@@ -2149,7 +2146,7 @@ static int group_can_go_on(struct perf_event *event,
        /*
         * Groups consisting entirely of software events can always go on.
         */
-       if (event->group_flags & PERF_GROUP_SOFTWARE)
+       if (event->group_caps & PERF_EV_CAP_SOFTWARE)
                return 1;
        /*
         * If an exclusive group is already on, no other hardware
@@ -9490,6 +9487,9 @@ SYSCALL_DEFINE5(perf_event_open,
                        goto err_alloc;
        }
 
+       if (pmu->task_ctx_nr == perf_sw_context)
+               event->event_caps |= PERF_EV_CAP_SOFTWARE;
+
        if (group_leader &&
            (is_software_event(event) != is_software_event(group_leader))) {
                if (is_software_event(event)) {
@@ -9503,7 +9503,7 @@ SYSCALL_DEFINE5(perf_event_open,
                         */
                        pmu = group_leader->pmu;
                } else if (is_software_event(group_leader) &&
-                          (group_leader->group_flags & PERF_GROUP_SOFTWARE)) {
+                          (group_leader->group_caps & PERF_EV_CAP_SOFTWARE)) {
                        /*
                         * In case the group is a pure software group, and we
                         * try to add a hardware event, move the whole group to