[RESEND PATCH v4 3/5] MIPS/Perf-events: Fix event check in validate_event()

[Deng-Cheng Zhu]

Ignore events that are in off/error state or belong to a different PMU.

This patch originates from the following commit for ARM by Will Deacon:

- 65b4711ff513767341aa1915c822de6ec0de65cb
    ARM: 6352/1: perf: fix event validation

    The validate_event function in the ARM perf events backend has the
    following problems:

    1.) Events that are disabled count towards the cost.
    2.) Events associated with other PMUs [for example, software events or
        breakpoints] do not count towards the cost, but do fail validation,
        causing the group to fail.

    This patch changes validate_event so that it ignores events in the
    PERF_EVENT_STATE_OFF state or that are scheduled for other PMUs.

Acked-by: Will Deacon <will.deacon@xxxxxxx>
Acked-by: David Daney <ddaney@xxxxxxxxxxxxxxxxxx>
Signed-off-by: Deng-Cheng Zhu <dengcheng.zhu@xxxxxxxxx>
---
Changes:
v4 - v3:
o None
v3 - v2:
o Keep all mentioned commits in the form of number + title + original
summary + (MIPS specific info when needed).
v2 - v1:
o Corrected the return value of the event check in validate_event().

 arch/mips/kernel/perf_event.c |    5 +++--
 1 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/arch/mips/kernel/perf_event.c b/arch/mips/kernel/perf_event.c
index 1ee44a3..3d55761 100644
--- a/arch/mips/kernel/perf_event.c
+++ b/arch/mips/kernel/perf_event.c
@@ -486,8 +486,9 @@ static int validate_event(struct cpu_hw_events *cpuc,
 {
 	struct hw_perf_event fake_hwc = event->hw;
 
-	if (event->pmu && event->pmu != &pmu)
-		return 0;
+	/* Allow mixed event group. So return 1 to pass validation. */
+	if (event->pmu != &pmu || event->state <= PERF_EVENT_STATE_OFF)
+		return 1;
 
 	return mipspmu->alloc_counter(cpuc, &fake_hwc) >= 0;
 }
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/