Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1602819
| From | Andy Lutomirski <luto@kernel.org> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | [PATCH 1/2] x86/perf: Fix CR4.PCE propagation to use active_mm instead of mm |
| Date | 2017-03-16 21:10 +0100 |
| Message-ID | <tlJSG-3pV-25@gated-at.bofh.it> (permalink) |
| References | <tlJSF-3pV-5@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
If one thread mmaps a perf event while another thread in the same mm
is in some context where active_mm != mm (which can happen in the
scheduler, for example), refresh_pce() would write the wrong value
to CR4.PCE. This broke some PAPI tests.
Cc: stable@vger.kernel.org
Fixes: 7911d3f7af14 ("perf/x86: Only allow rdpmc if a perf_event is mapped")
Reported-and-tested-by: Vince Weaver <vincent.weaver@maine.edu>
Signed-off-by: Andy Lutomirski <luto@kernel.org>
---
arch/x86/events/core.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
index 349d4d17aa7f..4f564df73b8f 100644
--- a/arch/x86/events/core.c
+++ b/arch/x86/events/core.c
@@ -2101,8 +2101,8 @@ static int x86_pmu_event_init(struct perf_event *event)
static void refresh_pce(void *ignored)
{
- if (current->mm)
- load_mm_cr4(current->mm);
+ if (current->active_mm)
+ load_mm_cr4(current->active_mm);
}
static void x86_pmu_event_mapped(struct perf_event *event)
--
2.9.3
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[PATCH 0/2] x86/perf: A CR4.PCE bugfix and clarification Andy Lutomirski <luto@kernel.org> - 2017-03-16 21:10 +0100
[PATCH 1/2] x86/perf: Fix CR4.PCE propagation to use active_mm instead of mm Andy Lutomirski <luto@kernel.org> - 2017-03-16 21:10 +0100
[tip:perf/urgent] x86/perf: Fix CR4.PCE propagation to use active_mm instead of mm tip-bot for Andy Lutomirski <tipbot@zytor.com> - 2017-03-17 11:50 +0100
[PATCH 2/2] x86/perf: Clarify why x86_pmu_event_mapped() isn't racy Andy Lutomirski <luto@kernel.org> - 2017-03-16 21:10 +0100
[tip:perf/urgent] x86/perf: Clarify why x86_pmu_event_mapped() isn't racy tip-bot for Andy Lutomirski <tipbot@zytor.com> - 2017-03-17 11:50 +0100
csiph-web