perf/core: Fix exclusive events' grouping

So far, we tried to disallow grouping exclusive events for the fear of
complications they would cause with moving between contexts. Specifically,
moving a software group to a hardware context would violate the exclusivity
rules if both groups contain matching exclusive events.

This attempt was, however, unsuccessful: the check that we have in the
perf_event_open() syscall is both wrong (looks at wrong PMU) and
insufficient (group leader may still be exclusive), as can be illustrated
by running:

  $ perf record -e '{intel_pt//,cycles}' uname
  $ perf record -e '{cycles,intel_pt//}' uname

ultimately successfully.

Furthermore, we are completely free to trigger the exclusivity violation
by:

   perf -e '{cycles,intel_pt//}' -e '{intel_pt//,instructions}'

even though the helpful perf record will not allow that, the ABI will.

The warning later in the perf_event_open() path will also not trigger, because
it's also wrong.

Fix all this by validating the original group before moving, getting rid
of broken safeguards and placing a useful one to perf_install_in_context().

Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: <stable@vger.kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: mathieu.poirier@linaro.org
Cc: will.deacon@arm.com
Fixes: bed5b25ad9c8a ("perf: Add a pmu capability for "exclusive" events")
Link: https://lkml.kernel.org/r/20190701110755.24646-1-alexander.shishkin@linux.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 16e38c286d46be1f608dbcfd88e3dd5f383a78e4..e8ad3c590a231d7b1f64ca73a437c2105da4eac5 100644 (file)
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -1055,6 +1055,11 @@ static inline int in_software_context(struct perf_event *event)
        return event->ctx->pmu->task_ctx_nr == perf_sw_context;
 }
 
+static inline int is_exclusive_pmu(struct pmu *pmu)
+{
+       return pmu->capabilities & PERF_PMU_CAP_EXCLUSIVE;
+}
+
 extern struct static_key perf_swevent_enabled[PERF_COUNT_SW_MAX];
 
 extern void ___perf_sw_event(u32, u64, struct pt_regs *, u64);

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 5dd19bedbf64747d27332ba9b9b029d0e0cec510..eea9d52b010c7a611dd08af16b934ff5508e5d45 100644 (file)
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -2553,6 +2553,9 @@ unlock:
        return ret;
 }
 
+static bool exclusive_event_installable(struct perf_event *event,
+                                       struct perf_event_context *ctx);
+
 /*
  * Attach a performance event to a context.
  *
@@ -2567,6 +2570,8 @@ perf_install_in_context(struct perf_event_context *ctx,
 
        lockdep_assert_held(&ctx->mutex);
 
+       WARN_ON_ONCE(!exclusive_event_installable(event, ctx));
+
        if (event->cpu != -1)
                event->cpu = cpu;
 
@@ -4360,7 +4365,7 @@ static int exclusive_event_init(struct perf_event *event)
 {
        struct pmu *pmu = event->pmu;
 
-       if (!(pmu->capabilities & PERF_PMU_CAP_EXCLUSIVE))
+       if (!is_exclusive_pmu(pmu))
                return 0;
 
        /*
@@ -4391,7 +4396,7 @@ static void exclusive_event_destroy(struct perf_event *event)
 {
        struct pmu *pmu = event->pmu;
 
-       if (!(pmu->capabilities & PERF_PMU_CAP_EXCLUSIVE))
+       if (!is_exclusive_pmu(pmu))
                return;
 
        /* see comment in exclusive_event_init() */
@@ -4411,14 +4416,15 @@ static bool exclusive_event_match(struct perf_event *e1, struct perf_event *e2)
        return false;
 }
 
-/* Called under the same ctx::mutex as perf_install_in_context() */
 static bool exclusive_event_installable(struct perf_event *event,
                                        struct perf_event_context *ctx)
 {
        struct perf_event *iter_event;
        struct pmu *pmu = event->pmu;
 
-       if (!(pmu->capabilities & PERF_PMU_CAP_EXCLUSIVE))
+       lockdep_assert_held(&ctx->mutex);
+
+       if (!is_exclusive_pmu(pmu))
                return true;
 
        list_for_each_entry(iter_event, &ctx->event_list, event_entry) {
@@ -10947,11 +10953,6 @@ SYSCALL_DEFINE5(perf_event_open,
                goto err_alloc;
        }
 
-       if ((pmu->capabilities & PERF_PMU_CAP_EXCLUSIVE) && group_leader) {
-               err = -EBUSY;
-               goto err_context;
-       }
-
        /*
         * Look up the group leader (we will attach this event to it):
         */
@@ -11039,6 +11040,18 @@ SYSCALL_DEFINE5(perf_event_open,
                                move_group = 0;
                        }
                }
+
+               /*
+                * Failure to create exclusive events returns -EBUSY.
+                */
+               err = -EBUSY;
+               if (!exclusive_event_installable(group_leader, ctx))
+                       goto err_locked;
+
+               for_each_sibling_event(sibling, group_leader) {
+                       if (!exclusive_event_installable(sibling, ctx))
+                               goto err_locked;
+               }
        } else {
                mutex_lock(&ctx->mutex);
        }
@@ -11075,9 +11088,6 @@ SYSCALL_DEFINE5(perf_event_open,
         * because we need to serialize with concurrent event creation.
         */
        if (!exclusive_event_installable(event, ctx)) {
-               /* exclusive and group stuff are assumed mutually exclusive */
-               WARN_ON_ONCE(move_group);
-
                err = -EBUSY;
                goto err_locked;
        }