Skip to contentSkip to conversation

About perf,arm -- oops in validate_event

PATCH P J P 3 participants 4 messages Aug 20, 2013

Subsystems (3)
lore
Share & tools
lore

P J P

author
#0 lore raw

Hello,

-> https://lkml.org/lkml/2013/8/7/259

I wanted to confirm if this above fix should also go into ARM64 build Or is ARM64 platform not vulnerable?

===

$ git diff
arch/arm64/kernel/perf_event.c +3 -0
index 9ba33c4..cbed82f 100644
@@ -317,6 +317,9 @@ validate_event(struct pmu_hw_events *hw_events,
        struct hw_perf_event fake_event = event->hw;
        struct pmu *leader_pmu = event->group_leader->pmu;
 
+    if (is_software_event(event))
+        return 1;
+
        if (event->pmu != leader_pmu || event->state <= PERF_EVENT_STATE_OFF)
                return 1;

===

Thank you.
--
Prasad J Pandit / Red Hat Security Response Team DB7A 84C5 D3F9 7CD1 B5EB C939 D048 7860 3655 602B

maintainer
#1 lore raw
Quoted 4 lines · P J P

-> https://lkml.org/lkml/2013/8/7/259

I wanted to confirm if this above fix should also go into ARM64 build Or is ARM64 platform not vulnerable?

It is and I'll push patches to mainline (Will is preparing them).

Catalin

#2 lore raw
Quoted 7 lines · Catalin Marinas
Quoted 4 lines · P J P

-> https://lkml.org/lkml/2013/8/7/259

I wanted to confirm if this above fix should also go into ARM64 build Or is ARM64 platform not vulnerable?

It is and I'll push patches to mainline (Will is preparing them).

You'll probably also want the equivalent of

    d9f966357b14e356dbd83b8f4a197a287ab4ff83

as well since arch/arm64/kernel/perf_event.c
has the same overflow in armpmu_map_event that is fixed in the mentioned commit.

Vince

maintainer
#3 lore raw
Quoted 8 lines · Vince Weaver
Quoted 6 lines · Catalin Marinas
Quoted 4 lines · P J P

-> https://lkml.org/lkml/2013/8/7/259

I wanted to confirm if this above fix should also go into ARM64 build Or is ARM64 platform not vulnerable?

Quoted 7 lines
Quoted 1 line

It is and I'll push patches to mainline (Will is preparing them).

You'll probably also want the equivalent of

    d9f966357b14e356dbd83b8f4a197a287ab4ff83

as well since arch/arm64/kernel/perf_event.c
has the same overflow in armpmu_map_event that is fixed in the mentioned commit.

Yes, I'm pushing both.

Thanks.

Signature · 1 line
Catalin