linux-kernel - [tip:perf/core] perf: Fix locking for PERF_FORMAT

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

Date:	Sat, 21 Nov 2009 13:44:00 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, paulus@...ba.org, hpa@...or.com,
	mingo@...hat.com, a.p.zijlstra@...llo.nl, tglx@...utronix.de,
	mingo@...e.hu
Subject: [tip:perf/core] perf: Fix locking for PERF_FORMAT_GROUP

Commit-ID:  6f10581aeaa5543a3b7a8c7a87a064375ec357f8
Gitweb:     http://git.kernel.org/tip/6f10581aeaa5543a3b7a8c7a87a064375ec357f8
Author:     Peter Zijlstra <a.p.zijlstra@...llo.nl>
AuthorDate: Fri, 20 Nov 2009 22:19:56 +0100
Committer:  Ingo Molnar <mingo@...e.hu>
CommitDate: Sat, 21 Nov 2009 14:11:40 +0100

perf: Fix locking for PERF_FORMAT_GROUP

We should hold event->child_mutex when iterating the inherited
counters, we should hold ctx->mutex when iterating siblings.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@...llo.nl>
Cc: Paul Mackerras <paulus@...ba.org>
LKML-Reference: <20091120212509.251030114@...llo.nl>
Signed-off-by: Ingo Molnar <mingo@...e.hu>
---
 kernel/perf_event.c |   20 +++++++++++++-------
 1 files changed, 13 insertions(+), 7 deletions(-)

diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 80f40da..3ede098 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -1782,6 +1782,7 @@ u64 perf_event_read_value(struct perf_event *event, u64 *enabled, u64 *running)
 	*enabled = 0;
 	*running = 0;
 
+	mutex_lock(&event->child_mutex);
 	total += perf_event_read(event);
 	*enabled += event->total_time_enabled +
 			atomic64_read(&event->child_total_time_enabled);
@@ -1793,6 +1794,7 @@ u64 perf_event_read_value(struct perf_event *event, u64 *enabled, u64 *running)
 		*enabled += child->total_time_enabled;
 		*running += child->total_time_running;
 	}
+	mutex_unlock(&event->child_mutex);
 
 	return total;
 }
@@ -1802,10 +1804,12 @@ static int perf_event_read_group(struct perf_event *event,
 				   u64 read_format, char __user *buf)
 {
 	struct perf_event *leader = event->group_leader, *sub;
-	int n = 0, size = 0, ret = 0;
+	int n = 0, size = 0, ret = -EFAULT;
+	struct perf_event_context *ctx = leader->ctx;
 	u64 values[5];
 	u64 count, enabled, running;
 
+	mutex_lock(&ctx->mutex);
 	count = perf_event_read_value(leader, &enabled, &running);
 
 	values[n++] = 1 + leader->nr_siblings;
@@ -1820,9 +1824,9 @@ static int perf_event_read_group(struct perf_event *event,
 	size = n * sizeof(u64);
 
 	if (copy_to_user(buf, values, size))
-		return -EFAULT;
+		goto unlock;
 
-	ret += size;
+	ret = size;
 
 	list_for_each_entry(sub, &leader->sibling_list, group_entry) {
 		n = 0;
@@ -1833,11 +1837,15 @@ static int perf_event_read_group(struct perf_event *event,
 
 		size = n * sizeof(u64);
 
-		if (copy_to_user(buf + size, values, size))
-			return -EFAULT;
+		if (copy_to_user(buf + size, values, size)) {
+			ret = -EFAULT;
+			goto unlock;
+		}
 
 		ret += size;
 	}
+unlock:
+	mutex_unlock(&ctx->mutex);
 
 	return ret;
 }
@@ -1884,12 +1892,10 @@ perf_read_hw(struct perf_event *event, char __user *buf, size_t count)
 		return -ENOSPC;
 
 	WARN_ON_ONCE(event->ctx->parent_ctx);
-	mutex_lock(&event->child_mutex);
 	if (read_format & PERF_FORMAT_GROUP)
 		ret = perf_event_read_group(event, read_format, buf);
 	else
 		ret = perf_event_read_one(event, read_format, buf);
-	mutex_unlock(&event->child_mutex);
 
 	return ret;
 }
--
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/