linux-kernel - [tip:perf/urgent] perf, x86: Avoid double disable on throttle vs ioctl(PERF_IOC

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Date:	Wed, 10 Mar 2010 13:12:22 GMT
From:	tip-bot for Peter Zijlstra <a.p.zijlstra@...llo.nl>
To:	linux-tip-commits@...r.kernel.org
Cc:	linux-kernel@...r.kernel.org, hpa@...or.com, mingo@...hat.com,
	acme@...radead.org, a.p.zijlstra@...llo.nl, tglx@...utronix.de,
	mingo@...e.hu
Subject: [tip:perf/urgent] perf, x86: Avoid double disable on throttle vs ioctl(PERF_IOC_DISABLE)

Commit-ID:  71e2d2828046133ed985696a02e2e1499ca0bfb8
Gitweb:     http://git.kernel.org/tip/71e2d2828046133ed985696a02e2e1499ca0bfb8
Author:     Peter Zijlstra <a.p.zijlstra@...llo.nl>
AuthorDate: Mon, 8 Mar 2010 17:51:33 +0100
Committer:  Ingo Molnar <mingo@...e.hu>
CommitDate: Wed, 10 Mar 2010 13:22:31 +0100

perf, x86: Avoid double disable on throttle vs ioctl(PERF_IOC_DISABLE)

Calling ioctl(PERF_EVENT_IOC_DISABLE) on a thottled counter would result
in a double disable, cure this by using x86_pmu_{start,stop} for
throttle/unthrottle and teach x86_pmu_stop() to check ->active_mask.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@...llo.nl>
Cc: Arnaldo Carvalho de Melo <acme@...radead.org>
Cc: paulus@...ba.org
Cc: eranian@...gle.com
Cc: robert.richter@....com
Cc: fweisbec@...il.com
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@...e.hu>
---
 arch/x86/kernel/cpu/perf_event.c       |   20 ++++++--------------
 arch/x86/kernel/cpu/perf_event_intel.c |    2 +-
 2 files changed, 7 insertions(+), 15 deletions(-)

diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 9757b96..b68c4fb 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -983,14 +983,8 @@ static int x86_pmu_start(struct perf_event *event)
 
 static void x86_pmu_unthrottle(struct perf_event *event)
 {
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	struct hw_perf_event *hwc = &event->hw;
-
-	if (WARN_ON_ONCE(hwc->idx >= X86_PMC_IDX_MAX ||
-				cpuc->events[hwc->idx] != event))
-		return;
-
-	x86_pmu.enable(event);
+	int ret = x86_pmu_start(event);
+	WARN_ON_ONCE(ret);
 }
 
 void perf_event_print_debug(void)
@@ -1050,11 +1044,9 @@ static void x86_pmu_stop(struct perf_event *event)
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
 
-	/*
-	 * Must be done before we disable, otherwise the nmi handler
-	 * could reenable again:
-	 */
-	__clear_bit(idx, cpuc->active_mask);
+	if (!__test_and_clear_bit(idx, cpuc->active_mask))
+		return;
+
 	x86_pmu.disable(event);
 
 	/*
@@ -1123,7 +1115,7 @@ static int x86_pmu_handle_irq(struct pt_regs *regs)
 			continue;
 
 		if (perf_event_overflow(event, 1, &data, regs))
-			x86_pmu.disable(event);
+			x86_pmu_stop(event);
 	}
 
 	if (handled)
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index d87421c..84bfde6 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -774,7 +774,7 @@ again:
 		data.period = event->hw.last_period;
 
 		if (perf_event_overflow(event, 1, &data, regs))
-			intel_pmu_disable_event(event);
+			x86_pmu_stop(event);
 	}
 
 	intel_pmu_ack_status(ack);
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/