[PATCH] perf: fix alloc_callchain_buffers()

[PATCH] perf: fix alloc_callchain_buffers()

[Eric Dumazet]

Commit 927c7a9e92c4 ( perf: Fix race in callchains ) introduced a
mismatch in the sizing of struct callchain_cpus_entries.

nr_cpu_ids must be used instead of num_possible_cpus(), or we might get
out of bound memory accesses on some machines.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Frederic Weisbecker <fweisbec@gmail.com>
CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
CC: Arnaldo Carvalho de Melo <acme@redhat.com>
CC: David Miller <davem@davemloft.net>
CC: Stephane Eranian <eranian@google.com>
CC: stable@kernel.org
---
 kernel/perf_event.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 126a302..852ae8c 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -1999,8 +1999,7 @@ static int alloc_callchain_buffers(void)
 	 * accessed from NMI. Use a temporary manual per cpu allocation
 	 * until that gets sorted out.
 	 */
-	size = sizeof(*entries) + sizeof(struct perf_callchain_entry *) *
-		num_possible_cpus();
+	size = offsetof(struct callchain_cpus_entries, cpu_entries[nr_cpu_ids]);
 
 	entries = kzalloc(size, GFP_KERNEL);
 	if (!entries)

[tip:perf/urgent] perf: Fix alloc_callchain_buffers()

[tip-bot for Eric Dumazet]

Commit-ID:  10f633a1a5cd8b9e2605d1f64a14ca008bce962b
Gitweb:     http://git.kernel.org/tip/10f633a1a5cd8b9e2605d1f64a14ca008bce962b
Author:     Eric Dumazet <eric.dumazet@gmail.com>
AuthorDate: Tue, 25 Jan 2011 19:40:51 +0100
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Wed, 26 Jan 2011 07:55:15 +0100

perf: Fix alloc_callchain_buffers()

Commit 927c7a9e92c4 ("perf: Fix race in callchains") introduced
a mismatch in the sizing of struct callchain_cpus_entries.

nr_cpu_ids must be used instead of num_possible_cpus(), or we
might get out of bound memory accesses on some machines.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: David Miller <davem@davemloft.net>
Cc: Stephane Eranian <eranian@google.com>
CC: stable@kernel.org
LKML-Reference: <1295980851.3588.351.camel@edumazet-laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/perf_event.c |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 126a302..852ae8c 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -1999,8 +1999,7 @@ static int alloc_callchain_buffers(void)
 	 * accessed from NMI. Use a temporary manual per cpu allocation
 	 * until that gets sorted out.
 	 */
-	size = sizeof(*entries) + sizeof(struct perf_callchain_entry *) *
-		num_possible_cpus();
+	size = offsetof(struct callchain_cpus_entries, cpu_entries[nr_cpu_ids]);
 
 	entries = kzalloc(size, GFP_KERNEL);
 	if (!entries)

Re: [PATCH] perf: fix alloc_callchain_buffers()

[Frederic Weisbecker]

On Tue, Jan 25, 2011 at 07:40:51PM +0100, Eric Dumazet wrote:
> Commit 927c7a9e92c4 ( perf: Fix race in callchains ) introduced a
> mismatch in the sizing of struct callchain_cpus_entries.
> 
> nr_cpu_ids must be used instead of num_possible_cpus(), or we might get
> out of bound memory accesses on some machines.
> 
> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
> CC: Frederic Weisbecker <fweisbec@gmail.com>
> CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
> CC: Arnaldo Carvalho de Melo <acme@redhat.com>
> CC: David Miller <davem@davemloft.net>
> CC: Stephane Eranian <eranian@google.com>
> CC: stable@kernel.org

Good catch, thanks!

[tip:perf/urgent] perf: Fix alloc_callchain_buffers()

[tip-bot for Eric Dumazet]

Commit-ID:  88d4f0db7fa8785859c1d637f9aac210932b6216
Gitweb:     http://git.kernel.org/tip/88d4f0db7fa8785859c1d637f9aac210932b6216
Author:     Eric Dumazet <eric.dumazet@gmail.com>
AuthorDate: Tue, 25 Jan 2011 19:40:51 +0100
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Thu, 27 Jan 2011 19:21:50 +0100

perf: Fix alloc_callchain_buffers()

Commit 927c7a9e92c4 ("perf: Fix race in callchains") introduced
a mismatch in the sizing of struct callchain_cpus_entries.

nr_cpu_ids must be used instead of num_possible_cpus(), or we
might get out of bound memory accesses on some machines.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: David Miller <davem@davemloft.net>
Cc: Stephane Eranian <eranian@google.com>
CC: stable@kernel.org
LKML-Reference: <1295980851.3588.351.camel@edumazet-laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/perf_event.c |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 126a302..852ae8c 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -1999,8 +1999,7 @@ static int alloc_callchain_buffers(void)
 	 * accessed from NMI. Use a temporary manual per cpu allocation
 	 * until that gets sorted out.
 	 */
-	size = sizeof(*entries) + sizeof(struct perf_callchain_entry *) *
-		num_possible_cpus();
+	size = offsetof(struct callchain_cpus_entries, cpu_entries[nr_cpu_ids]);
 
 	entries = kzalloc(size, GFP_KERNEL);
 	if (!entries)